{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5884052641256665,
  "eval_steps": 200,
  "global_step": 8400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.004824572924602e-05,
      "grad_norm": 6.222772121429443,
      "learning_rate": 9.99930017513135e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 16384,
      "step": 1
    },
    {
      "epoch": 0.00014009649145849205,
      "grad_norm": 6.042057037353516,
      "learning_rate": 9.998600350262697e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 32768,
      "step": 2
    },
    {
      "epoch": 0.00021014473718773804,
      "grad_norm": 7.119229316711426,
      "learning_rate": 9.997900525394046e-05,
      "loss": 1.4047,
      "num_input_tokens_seen": 49152,
      "step": 3
    },
    {
      "epoch": 0.0002801929829169841,
      "grad_norm": 7.133191108703613,
      "learning_rate": 9.997200700525395e-05,
      "loss": 1.3921,
      "num_input_tokens_seen": 65536,
      "step": 4
    },
    {
      "epoch": 0.0003502412286462301,
      "grad_norm": 6.1078338623046875,
      "learning_rate": 9.996500875656743e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 81920,
      "step": 5
    },
    {
      "epoch": 0.0004202894743754761,
      "grad_norm": 6.466420650482178,
      "learning_rate": 9.995801050788092e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 97344,
      "step": 6
    },
    {
      "epoch": 0.0004903377201047221,
      "grad_norm": 5.578189849853516,
      "learning_rate": 9.99510122591944e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 113728,
      "step": 7
    },
    {
      "epoch": 0.0005603859658339682,
      "grad_norm": 7.197720527648926,
      "learning_rate": 9.994401401050789e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 129528,
      "step": 8
    },
    {
      "epoch": 0.0006304342115632141,
      "grad_norm": 6.618913650512695,
      "learning_rate": 9.993701576182136e-05,
      "loss": 1.3495,
      "num_input_tokens_seen": 145704,
      "step": 9
    },
    {
      "epoch": 0.0007004824572924602,
      "grad_norm": 6.955508232116699,
      "learning_rate": 9.993001751313485e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 161664,
      "step": 10
    },
    {
      "epoch": 0.0007705307030217062,
      "grad_norm": 6.6807074546813965,
      "learning_rate": 9.992301926444835e-05,
      "loss": 1.1693,
      "num_input_tokens_seen": 177960,
      "step": 11
    },
    {
      "epoch": 0.0008405789487509522,
      "grad_norm": 6.784447193145752,
      "learning_rate": 9.991602101576183e-05,
      "loss": 1.3744,
      "num_input_tokens_seen": 194344,
      "step": 12
    },
    {
      "epoch": 0.0009106271944801982,
      "grad_norm": 6.7418437004089355,
      "learning_rate": 9.990902276707532e-05,
      "loss": 1.22,
      "num_input_tokens_seen": 210728,
      "step": 13
    },
    {
      "epoch": 0.0009806754402094443,
      "grad_norm": 6.43395471572876,
      "learning_rate": 9.990202451838879e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 227112,
      "step": 14
    },
    {
      "epoch": 0.0010507236859386903,
      "grad_norm": 6.09422492980957,
      "learning_rate": 9.989502626970228e-05,
      "loss": 1.195,
      "num_input_tokens_seen": 243496,
      "step": 15
    },
    {
      "epoch": 0.0011207719316679364,
      "grad_norm": 6.238271236419678,
      "learning_rate": 9.988802802101577e-05,
      "loss": 1.2623,
      "num_input_tokens_seen": 259744,
      "step": 16
    },
    {
      "epoch": 0.0011908201773971822,
      "grad_norm": 6.56187629699707,
      "learning_rate": 9.988102977232926e-05,
      "loss": 1.2721,
      "num_input_tokens_seen": 276128,
      "step": 17
    },
    {
      "epoch": 0.0012608684231264283,
      "grad_norm": 6.818358898162842,
      "learning_rate": 9.987403152364275e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 292512,
      "step": 18
    },
    {
      "epoch": 0.0013309166688556743,
      "grad_norm": 5.950352191925049,
      "learning_rate": 9.986703327495622e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 308632,
      "step": 19
    },
    {
      "epoch": 0.0014009649145849204,
      "grad_norm": 6.387479305267334,
      "learning_rate": 9.986003502626971e-05,
      "loss": 1.2783,
      "num_input_tokens_seen": 325016,
      "step": 20
    },
    {
      "epoch": 0.0014710131603141664,
      "grad_norm": 6.187346458435059,
      "learning_rate": 9.985303677758318e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 341384,
      "step": 21
    },
    {
      "epoch": 0.0015410614060434125,
      "grad_norm": 5.371951103210449,
      "learning_rate": 9.984603852889667e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 357768,
      "step": 22
    },
    {
      "epoch": 0.0016111096517726585,
      "grad_norm": 6.2206807136535645,
      "learning_rate": 9.983904028021016e-05,
      "loss": 1.2516,
      "num_input_tokens_seen": 374152,
      "step": 23
    },
    {
      "epoch": 0.0016811578975019044,
      "grad_norm": 6.121264457702637,
      "learning_rate": 9.983204203152365e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 390536,
      "step": 24
    },
    {
      "epoch": 0.0017512061432311504,
      "grad_norm": 6.353756904602051,
      "learning_rate": 9.982504378283714e-05,
      "loss": 1.3118,
      "num_input_tokens_seen": 406920,
      "step": 25
    },
    {
      "epoch": 0.0018212543889603965,
      "grad_norm": 6.270686149597168,
      "learning_rate": 9.981804553415061e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 422728,
      "step": 26
    },
    {
      "epoch": 0.0018913026346896425,
      "grad_norm": 6.117632865905762,
      "learning_rate": 9.98110472854641e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 439112,
      "step": 27
    },
    {
      "epoch": 0.0019613508804188886,
      "grad_norm": 6.429015159606934,
      "learning_rate": 9.980404903677759e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 455144,
      "step": 28
    },
    {
      "epoch": 0.0020313991261481346,
      "grad_norm": 6.4467620849609375,
      "learning_rate": 9.979705078809107e-05,
      "loss": 1.3335,
      "num_input_tokens_seen": 470360,
      "step": 29
    },
    {
      "epoch": 0.0021014473718773807,
      "grad_norm": 6.57926082611084,
      "learning_rate": 9.979005253940455e-05,
      "loss": 1.2126,
      "num_input_tokens_seen": 486120,
      "step": 30
    },
    {
      "epoch": 0.0021714956176066267,
      "grad_norm": 5.650569915771484,
      "learning_rate": 9.978305429071804e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 501896,
      "step": 31
    },
    {
      "epoch": 0.0022415438633358728,
      "grad_norm": 6.380292892456055,
      "learning_rate": 9.977605604203153e-05,
      "loss": 1.2251,
      "num_input_tokens_seen": 517752,
      "step": 32
    },
    {
      "epoch": 0.002311592109065119,
      "grad_norm": 5.704173564910889,
      "learning_rate": 9.976905779334502e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 534136,
      "step": 33
    },
    {
      "epoch": 0.0023816403547943644,
      "grad_norm": 5.342978000640869,
      "learning_rate": 9.97620595446585e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 550216,
      "step": 34
    },
    {
      "epoch": 0.0024516886005236105,
      "grad_norm": 5.7014241218566895,
      "learning_rate": 9.975506129597198e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 566600,
      "step": 35
    },
    {
      "epoch": 0.0025217368462528565,
      "grad_norm": 6.26229190826416,
      "learning_rate": 9.974806304728546e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 582984,
      "step": 36
    },
    {
      "epoch": 0.0025917850919821026,
      "grad_norm": 6.583463191986084,
      "learning_rate": 9.974106479859896e-05,
      "loss": 1.3021,
      "num_input_tokens_seen": 598968,
      "step": 37
    },
    {
      "epoch": 0.0026618333377113486,
      "grad_norm": 5.58498477935791,
      "learning_rate": 9.973406654991245e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 614840,
      "step": 38
    },
    {
      "epoch": 0.0027318815834405947,
      "grad_norm": 5.906906604766846,
      "learning_rate": 9.972706830122592e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 631224,
      "step": 39
    },
    {
      "epoch": 0.0028019298291698407,
      "grad_norm": 5.962359428405762,
      "learning_rate": 9.972007005253941e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 647000,
      "step": 40
    },
    {
      "epoch": 0.002871978074899087,
      "grad_norm": 6.447500705718994,
      "learning_rate": 9.971307180385289e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 662480,
      "step": 41
    },
    {
      "epoch": 0.002942026320628333,
      "grad_norm": 5.7290520668029785,
      "learning_rate": 9.970607355516638e-05,
      "loss": 1.3585,
      "num_input_tokens_seen": 678480,
      "step": 42
    },
    {
      "epoch": 0.003012074566357579,
      "grad_norm": 6.063445568084717,
      "learning_rate": 9.969907530647987e-05,
      "loss": 1.2841,
      "num_input_tokens_seen": 694256,
      "step": 43
    },
    {
      "epoch": 0.003082122812086825,
      "grad_norm": 5.302809238433838,
      "learning_rate": 9.969207705779335e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 710152,
      "step": 44
    },
    {
      "epoch": 0.003152171057816071,
      "grad_norm": 5.634128093719482,
      "learning_rate": 9.968507880910684e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 726184,
      "step": 45
    },
    {
      "epoch": 0.003222219303545317,
      "grad_norm": 5.652642726898193,
      "learning_rate": 9.967808056042032e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 742520,
      "step": 46
    },
    {
      "epoch": 0.0032922675492745627,
      "grad_norm": 5.340751647949219,
      "learning_rate": 9.96710823117338e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 758904,
      "step": 47
    },
    {
      "epoch": 0.0033623157950038087,
      "grad_norm": 5.422239780426025,
      "learning_rate": 9.966408406304728e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 775040,
      "step": 48
    },
    {
      "epoch": 0.0034323640407330548,
      "grad_norm": 5.29241418838501,
      "learning_rate": 9.965708581436077e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 790856,
      "step": 49
    },
    {
      "epoch": 0.003502412286462301,
      "grad_norm": 5.146270275115967,
      "learning_rate": 9.965008756567426e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 807064,
      "step": 50
    },
    {
      "epoch": 0.003572460532191547,
      "grad_norm": 5.825758457183838,
      "learning_rate": 9.964308931698775e-05,
      "loss": 1.2108,
      "num_input_tokens_seen": 823448,
      "step": 51
    },
    {
      "epoch": 0.003642508777920793,
      "grad_norm": 6.179538726806641,
      "learning_rate": 9.963609106830124e-05,
      "loss": 1.322,
      "num_input_tokens_seen": 838888,
      "step": 52
    },
    {
      "epoch": 0.003712557023650039,
      "grad_norm": 6.464454174041748,
      "learning_rate": 9.962909281961471e-05,
      "loss": 1.5077,
      "num_input_tokens_seen": 855272,
      "step": 53
    },
    {
      "epoch": 0.003782605269379285,
      "grad_norm": 5.4227294921875,
      "learning_rate": 9.96220945709282e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 871656,
      "step": 54
    },
    {
      "epoch": 0.003852653515108531,
      "grad_norm": 5.949041366577148,
      "learning_rate": 9.961509632224169e-05,
      "loss": 1.3618,
      "num_input_tokens_seen": 888040,
      "step": 55
    },
    {
      "epoch": 0.003922701760837777,
      "grad_norm": 6.050904750823975,
      "learning_rate": 9.960809807355516e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 904400,
      "step": 56
    },
    {
      "epoch": 0.003992750006567023,
      "grad_norm": 6.048308849334717,
      "learning_rate": 9.960109982486866e-05,
      "loss": 1.3131,
      "num_input_tokens_seen": 919952,
      "step": 57
    },
    {
      "epoch": 0.004062798252296269,
      "grad_norm": 5.683863162994385,
      "learning_rate": 9.959410157618214e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 936336,
      "step": 58
    },
    {
      "epoch": 0.004132846498025515,
      "grad_norm": 5.449287414550781,
      "learning_rate": 9.958710332749563e-05,
      "loss": 1.0613,
      "num_input_tokens_seen": 952152,
      "step": 59
    },
    {
      "epoch": 0.004202894743754761,
      "grad_norm": 5.31496524810791,
      "learning_rate": 9.958010507880912e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 967824,
      "step": 60
    },
    {
      "epoch": 0.004272942989484007,
      "grad_norm": 5.57105016708374,
      "learning_rate": 9.957310683012259e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 983864,
      "step": 61
    },
    {
      "epoch": 0.004342991235213253,
      "grad_norm": 5.3456830978393555,
      "learning_rate": 9.956610858143608e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 1000248,
      "step": 62
    },
    {
      "epoch": 0.004413039480942499,
      "grad_norm": 5.453295707702637,
      "learning_rate": 9.955911033274957e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 1016632,
      "step": 63
    },
    {
      "epoch": 0.0044830877266717455,
      "grad_norm": 4.975449562072754,
      "learning_rate": 9.955211208406306e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 1033016,
      "step": 64
    },
    {
      "epoch": 0.004553135972400991,
      "grad_norm": 5.542137145996094,
      "learning_rate": 9.954511383537655e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 1049400,
      "step": 65
    },
    {
      "epoch": 0.004623184218130238,
      "grad_norm": 5.213950157165527,
      "learning_rate": 9.953811558669002e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 1065784,
      "step": 66
    },
    {
      "epoch": 0.004693232463859483,
      "grad_norm": 5.496099948883057,
      "learning_rate": 9.953111733800351e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 1082168,
      "step": 67
    },
    {
      "epoch": 0.004763280709588729,
      "grad_norm": 5.64145565032959,
      "learning_rate": 9.952411908931698e-05,
      "loss": 1.2301,
      "num_input_tokens_seen": 1098024,
      "step": 68
    },
    {
      "epoch": 0.004833328955317975,
      "grad_norm": 5.566709995269775,
      "learning_rate": 9.951712084063047e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 1114408,
      "step": 69
    },
    {
      "epoch": 0.004903377201047221,
      "grad_norm": 6.443673133850098,
      "learning_rate": 9.951012259194396e-05,
      "loss": 1.2313,
      "num_input_tokens_seen": 1130792,
      "step": 70
    },
    {
      "epoch": 0.0049734254467764675,
      "grad_norm": 5.882962226867676,
      "learning_rate": 9.950312434325745e-05,
      "loss": 1.4304,
      "num_input_tokens_seen": 1147176,
      "step": 71
    },
    {
      "epoch": 0.005043473692505713,
      "grad_norm": 6.0052666664123535,
      "learning_rate": 9.949612609457094e-05,
      "loss": 1.3027,
      "num_input_tokens_seen": 1160968,
      "step": 72
    },
    {
      "epoch": 0.0051135219382349596,
      "grad_norm": 5.260256767272949,
      "learning_rate": 9.948912784588441e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 1177352,
      "step": 73
    },
    {
      "epoch": 0.005183570183964205,
      "grad_norm": 5.641814708709717,
      "learning_rate": 9.94821295971979e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 1193032,
      "step": 74
    },
    {
      "epoch": 0.005253618429693452,
      "grad_norm": 5.121115207672119,
      "learning_rate": 9.947513134851138e-05,
      "loss": 1.2404,
      "num_input_tokens_seen": 1208952,
      "step": 75
    },
    {
      "epoch": 0.005323666675422697,
      "grad_norm": 5.63930082321167,
      "learning_rate": 9.946813309982487e-05,
      "loss": 1.5127,
      "num_input_tokens_seen": 1225000,
      "step": 76
    },
    {
      "epoch": 0.005393714921151944,
      "grad_norm": 4.880716800689697,
      "learning_rate": 9.946113485113837e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 1241384,
      "step": 77
    },
    {
      "epoch": 0.005463763166881189,
      "grad_norm": 5.59611177444458,
      "learning_rate": 9.945413660245184e-05,
      "loss": 1.1678,
      "num_input_tokens_seen": 1257680,
      "step": 78
    },
    {
      "epoch": 0.005533811412610436,
      "grad_norm": 5.052026271820068,
      "learning_rate": 9.944713835376533e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 1274064,
      "step": 79
    },
    {
      "epoch": 0.0056038596583396815,
      "grad_norm": 5.285096168518066,
      "learning_rate": 9.944014010507881e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 1290448,
      "step": 80
    },
    {
      "epoch": 0.005673907904068927,
      "grad_norm": 5.4286580085754395,
      "learning_rate": 9.94331418563923e-05,
      "loss": 1.3047,
      "num_input_tokens_seen": 1306832,
      "step": 81
    },
    {
      "epoch": 0.005743956149798174,
      "grad_norm": 5.937953472137451,
      "learning_rate": 9.942614360770578e-05,
      "loss": 1.4353,
      "num_input_tokens_seen": 1323216,
      "step": 82
    },
    {
      "epoch": 0.005814004395527419,
      "grad_norm": 5.129006385803223,
      "learning_rate": 9.941914535901927e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 1339408,
      "step": 83
    },
    {
      "epoch": 0.005884052641256666,
      "grad_norm": 5.179675102233887,
      "learning_rate": 9.941214711033276e-05,
      "loss": 1.2452,
      "num_input_tokens_seen": 1355792,
      "step": 84
    },
    {
      "epoch": 0.005954100886985911,
      "grad_norm": 4.912832736968994,
      "learning_rate": 9.940514886164624e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 1372176,
      "step": 85
    },
    {
      "epoch": 0.006024149132715158,
      "grad_norm": 5.190899848937988,
      "learning_rate": 9.939815061295973e-05,
      "loss": 1.2543,
      "num_input_tokens_seen": 1388560,
      "step": 86
    },
    {
      "epoch": 0.006094197378444403,
      "grad_norm": 5.1751275062561035,
      "learning_rate": 9.939115236427321e-05,
      "loss": 1.3145,
      "num_input_tokens_seen": 1404944,
      "step": 87
    },
    {
      "epoch": 0.00616424562417365,
      "grad_norm": 5.450705528259277,
      "learning_rate": 9.938415411558669e-05,
      "loss": 1.2844,
      "num_input_tokens_seen": 1421328,
      "step": 88
    },
    {
      "epoch": 0.0062342938699028955,
      "grad_norm": 5.593935012817383,
      "learning_rate": 9.937715586690018e-05,
      "loss": 1.3284,
      "num_input_tokens_seen": 1437464,
      "step": 89
    },
    {
      "epoch": 0.006304342115632142,
      "grad_norm": 5.156428813934326,
      "learning_rate": 9.937015761821367e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 1452952,
      "step": 90
    },
    {
      "epoch": 0.006374390361361388,
      "grad_norm": 4.673638820648193,
      "learning_rate": 9.936315936952715e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 1469336,
      "step": 91
    },
    {
      "epoch": 0.006444438607090634,
      "grad_norm": 4.996700763702393,
      "learning_rate": 9.935616112084064e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 1485448,
      "step": 92
    },
    {
      "epoch": 0.00651448685281988,
      "grad_norm": 4.817474365234375,
      "learning_rate": 9.934916287215412e-05,
      "loss": 1.151,
      "num_input_tokens_seen": 1501472,
      "step": 93
    },
    {
      "epoch": 0.006584535098549125,
      "grad_norm": 5.400479316711426,
      "learning_rate": 9.934216462346761e-05,
      "loss": 1.3144,
      "num_input_tokens_seen": 1516424,
      "step": 94
    },
    {
      "epoch": 0.006654583344278372,
      "grad_norm": 5.232216835021973,
      "learning_rate": 9.933516637478108e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 1532792,
      "step": 95
    },
    {
      "epoch": 0.006724631590007617,
      "grad_norm": 5.392521381378174,
      "learning_rate": 9.932816812609457e-05,
      "loss": 1.3195,
      "num_input_tokens_seen": 1548600,
      "step": 96
    },
    {
      "epoch": 0.006794679835736864,
      "grad_norm": 5.5280866622924805,
      "learning_rate": 9.932116987740806e-05,
      "loss": 1.283,
      "num_input_tokens_seen": 1564088,
      "step": 97
    },
    {
      "epoch": 0.0068647280814661095,
      "grad_norm": 4.963179588317871,
      "learning_rate": 9.931417162872155e-05,
      "loss": 1.2716,
      "num_input_tokens_seen": 1580040,
      "step": 98
    },
    {
      "epoch": 0.006934776327195356,
      "grad_norm": 4.920302391052246,
      "learning_rate": 9.930717338003504e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 1595880,
      "step": 99
    },
    {
      "epoch": 0.007004824572924602,
      "grad_norm": 4.935486793518066,
      "learning_rate": 9.930017513134851e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 1611864,
      "step": 100
    },
    {
      "epoch": 0.007074872818653848,
      "grad_norm": 5.099087238311768,
      "learning_rate": 9.9293176882662e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 1627472,
      "step": 101
    },
    {
      "epoch": 0.007144921064383094,
      "grad_norm": 5.3764328956604,
      "learning_rate": 9.928617863397548e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 1643856,
      "step": 102
    },
    {
      "epoch": 0.00721496931011234,
      "grad_norm": 5.281564712524414,
      "learning_rate": 9.927918038528898e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 1660240,
      "step": 103
    },
    {
      "epoch": 0.007285017555841586,
      "grad_norm": 5.395167827606201,
      "learning_rate": 9.927218213660247e-05,
      "loss": 1.6014,
      "num_input_tokens_seen": 1676624,
      "step": 104
    },
    {
      "epoch": 0.007355065801570832,
      "grad_norm": 5.322319507598877,
      "learning_rate": 9.926518388791594e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 1693008,
      "step": 105
    },
    {
      "epoch": 0.007425114047300078,
      "grad_norm": 5.301229953765869,
      "learning_rate": 9.925818563922943e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 1708424,
      "step": 106
    },
    {
      "epoch": 0.0074951622930293236,
      "grad_norm": 4.958597183227539,
      "learning_rate": 9.92511873905429e-05,
      "loss": 1.3285,
      "num_input_tokens_seen": 1724808,
      "step": 107
    },
    {
      "epoch": 0.00756521053875857,
      "grad_norm": 4.3913960456848145,
      "learning_rate": 9.924418914185639e-05,
      "loss": 0.9017,
      "num_input_tokens_seen": 1740752,
      "step": 108
    },
    {
      "epoch": 0.007635258784487816,
      "grad_norm": 5.401021480560303,
      "learning_rate": 9.923719089316988e-05,
      "loss": 1.3646,
      "num_input_tokens_seen": 1755176,
      "step": 109
    },
    {
      "epoch": 0.007705307030217062,
      "grad_norm": 4.894444942474365,
      "learning_rate": 9.923019264448337e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 1771560,
      "step": 110
    },
    {
      "epoch": 0.007775355275946308,
      "grad_norm": 4.878688335418701,
      "learning_rate": 9.922319439579686e-05,
      "loss": 1.1766,
      "num_input_tokens_seen": 1787944,
      "step": 111
    },
    {
      "epoch": 0.007845403521675554,
      "grad_norm": 4.9379777908325195,
      "learning_rate": 9.921619614711033e-05,
      "loss": 1.1631,
      "num_input_tokens_seen": 1803568,
      "step": 112
    },
    {
      "epoch": 0.0079154517674048,
      "grad_norm": 5.101811408996582,
      "learning_rate": 9.920919789842382e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 1819952,
      "step": 113
    },
    {
      "epoch": 0.007985500013134045,
      "grad_norm": 5.32574987411499,
      "learning_rate": 9.920219964973731e-05,
      "loss": 1.3012,
      "num_input_tokens_seen": 1835296,
      "step": 114
    },
    {
      "epoch": 0.008055548258863293,
      "grad_norm": 5.2391180992126465,
      "learning_rate": 9.919520140105079e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 1851224,
      "step": 115
    },
    {
      "epoch": 0.008125596504592538,
      "grad_norm": 4.865017890930176,
      "learning_rate": 9.918820315236427e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 1867608,
      "step": 116
    },
    {
      "epoch": 0.008195644750321784,
      "grad_norm": 4.943136215209961,
      "learning_rate": 9.918120490367776e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 1883696,
      "step": 117
    },
    {
      "epoch": 0.00826569299605103,
      "grad_norm": 4.769871711730957,
      "learning_rate": 9.917420665499125e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 1900080,
      "step": 118
    },
    {
      "epoch": 0.008335741241780275,
      "grad_norm": 4.785780429840088,
      "learning_rate": 9.916720840630474e-05,
      "loss": 1.2415,
      "num_input_tokens_seen": 1916464,
      "step": 119
    },
    {
      "epoch": 0.008405789487509523,
      "grad_norm": 4.802333831787109,
      "learning_rate": 9.916021015761822e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 1932848,
      "step": 120
    },
    {
      "epoch": 0.008475837733238768,
      "grad_norm": 5.22212553024292,
      "learning_rate": 9.91532119089317e-05,
      "loss": 1.2574,
      "num_input_tokens_seen": 1949232,
      "step": 121
    },
    {
      "epoch": 0.008545885978968014,
      "grad_norm": 5.104204177856445,
      "learning_rate": 9.914621366024518e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 1964184,
      "step": 122
    },
    {
      "epoch": 0.00861593422469726,
      "grad_norm": 5.11055326461792,
      "learning_rate": 9.913921541155868e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 1980568,
      "step": 123
    },
    {
      "epoch": 0.008685982470426507,
      "grad_norm": 4.784866809844971,
      "learning_rate": 9.913221716287216e-05,
      "loss": 1.2056,
      "num_input_tokens_seen": 1996952,
      "step": 124
    },
    {
      "epoch": 0.008756030716155752,
      "grad_norm": 4.763037204742432,
      "learning_rate": 9.912521891418564e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 2013336,
      "step": 125
    },
    {
      "epoch": 0.008826078961884998,
      "grad_norm": 4.813408851623535,
      "learning_rate": 9.911822066549913e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 2029720,
      "step": 126
    },
    {
      "epoch": 0.008896127207614244,
      "grad_norm": 4.79008674621582,
      "learning_rate": 9.911122241681261e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 2046104,
      "step": 127
    },
    {
      "epoch": 0.008966175453343491,
      "grad_norm": 4.843508720397949,
      "learning_rate": 9.91042241681261e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 2061592,
      "step": 128
    },
    {
      "epoch": 0.009036223699072737,
      "grad_norm": 4.917592525482178,
      "learning_rate": 9.909722591943959e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 2077792,
      "step": 129
    },
    {
      "epoch": 0.009106271944801982,
      "grad_norm": 4.9154133796691895,
      "learning_rate": 9.909022767075307e-05,
      "loss": 1.3284,
      "num_input_tokens_seen": 2094176,
      "step": 130
    },
    {
      "epoch": 0.009176320190531228,
      "grad_norm": 5.2125420570373535,
      "learning_rate": 9.908322942206656e-05,
      "loss": 1.3469,
      "num_input_tokens_seen": 2110480,
      "step": 131
    },
    {
      "epoch": 0.009246368436260475,
      "grad_norm": 4.715712547302246,
      "learning_rate": 9.907623117338004e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 2126864,
      "step": 132
    },
    {
      "epoch": 0.009316416681989721,
      "grad_norm": 4.805694580078125,
      "learning_rate": 9.906923292469353e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 2142848,
      "step": 133
    },
    {
      "epoch": 0.009386464927718966,
      "grad_norm": 4.961355209350586,
      "learning_rate": 9.9062234676007e-05,
      "loss": 1.3387,
      "num_input_tokens_seen": 2159232,
      "step": 134
    },
    {
      "epoch": 0.009456513173448212,
      "grad_norm": 4.582219123840332,
      "learning_rate": 9.905523642732049e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 2175616,
      "step": 135
    },
    {
      "epoch": 0.009526561419177458,
      "grad_norm": 5.195998191833496,
      "learning_rate": 9.904823817863398e-05,
      "loss": 1.2552,
      "num_input_tokens_seen": 2191872,
      "step": 136
    },
    {
      "epoch": 0.009596609664906705,
      "grad_norm": 4.934189319610596,
      "learning_rate": 9.904123992994747e-05,
      "loss": 1.2961,
      "num_input_tokens_seen": 2208208,
      "step": 137
    },
    {
      "epoch": 0.00966665791063595,
      "grad_norm": 4.981037616729736,
      "learning_rate": 9.903424168126096e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 2224592,
      "step": 138
    },
    {
      "epoch": 0.009736706156365196,
      "grad_norm": 5.469496250152588,
      "learning_rate": 9.902724343257443e-05,
      "loss": 1.3833,
      "num_input_tokens_seen": 2240976,
      "step": 139
    },
    {
      "epoch": 0.009806754402094442,
      "grad_norm": 4.889583587646484,
      "learning_rate": 9.902024518388792e-05,
      "loss": 1.2095,
      "num_input_tokens_seen": 2257360,
      "step": 140
    },
    {
      "epoch": 0.00987680264782369,
      "grad_norm": 4.532052516937256,
      "learning_rate": 9.901324693520141e-05,
      "loss": 1.143,
      "num_input_tokens_seen": 2272848,
      "step": 141
    },
    {
      "epoch": 0.009946850893552935,
      "grad_norm": 5.278079032897949,
      "learning_rate": 9.900624868651488e-05,
      "loss": 1.2849,
      "num_input_tokens_seen": 2289232,
      "step": 142
    },
    {
      "epoch": 0.01001689913928218,
      "grad_norm": 4.549891948699951,
      "learning_rate": 9.899925043782839e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 2305424,
      "step": 143
    },
    {
      "epoch": 0.010086947385011426,
      "grad_norm": 4.7777180671691895,
      "learning_rate": 9.899225218914186e-05,
      "loss": 1.1926,
      "num_input_tokens_seen": 2320968,
      "step": 144
    },
    {
      "epoch": 0.010156995630740673,
      "grad_norm": 4.320313453674316,
      "learning_rate": 9.898525394045535e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 2337352,
      "step": 145
    },
    {
      "epoch": 0.010227043876469919,
      "grad_norm": 4.915202617645264,
      "learning_rate": 9.897825569176882e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 2353064,
      "step": 146
    },
    {
      "epoch": 0.010297092122199165,
      "grad_norm": 4.569783687591553,
      "learning_rate": 9.897125744308231e-05,
      "loss": 0.8586,
      "num_input_tokens_seen": 2369128,
      "step": 147
    },
    {
      "epoch": 0.01036714036792841,
      "grad_norm": 4.591664791107178,
      "learning_rate": 9.89642591943958e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 2385512,
      "step": 148
    },
    {
      "epoch": 0.010437188613657656,
      "grad_norm": 4.913016319274902,
      "learning_rate": 9.895726094570929e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 2401208,
      "step": 149
    },
    {
      "epoch": 0.010507236859386903,
      "grad_norm": 4.908018112182617,
      "learning_rate": 9.895026269702278e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 2417592,
      "step": 150
    },
    {
      "epoch": 0.010577285105116149,
      "grad_norm": 4.536910057067871,
      "learning_rate": 9.894326444833625e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 2433976,
      "step": 151
    },
    {
      "epoch": 0.010647333350845395,
      "grad_norm": 4.899227142333984,
      "learning_rate": 9.893626619964974e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 2448072,
      "step": 152
    },
    {
      "epoch": 0.01071738159657464,
      "grad_norm": 4.600861072540283,
      "learning_rate": 9.892926795096323e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 2464240,
      "step": 153
    },
    {
      "epoch": 0.010787429842303888,
      "grad_norm": 4.707681179046631,
      "learning_rate": 9.89222697022767e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 2480624,
      "step": 154
    },
    {
      "epoch": 0.010857478088033133,
      "grad_norm": 4.748518466949463,
      "learning_rate": 9.89152714535902e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 2497008,
      "step": 155
    },
    {
      "epoch": 0.010927526333762379,
      "grad_norm": 4.794179439544678,
      "learning_rate": 9.890827320490368e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 2513392,
      "step": 156
    },
    {
      "epoch": 0.010997574579491624,
      "grad_norm": 4.593925476074219,
      "learning_rate": 9.890127495621717e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 2529776,
      "step": 157
    },
    {
      "epoch": 0.011067622825220872,
      "grad_norm": 4.318257808685303,
      "learning_rate": 9.889427670753066e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 2546160,
      "step": 158
    },
    {
      "epoch": 0.011137671070950117,
      "grad_norm": 4.631777286529541,
      "learning_rate": 9.888727845884414e-05,
      "loss": 1.1553,
      "num_input_tokens_seen": 2562544,
      "step": 159
    },
    {
      "epoch": 0.011207719316679363,
      "grad_norm": 4.896609783172607,
      "learning_rate": 9.888028021015762e-05,
      "loss": 1.1779,
      "num_input_tokens_seen": 2578088,
      "step": 160
    },
    {
      "epoch": 0.011277767562408609,
      "grad_norm": 4.3978681564331055,
      "learning_rate": 9.88732819614711e-05,
      "loss": 1.1778,
      "num_input_tokens_seen": 2594416,
      "step": 161
    },
    {
      "epoch": 0.011347815808137854,
      "grad_norm": 4.82927942276001,
      "learning_rate": 9.886628371278459e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 2609776,
      "step": 162
    },
    {
      "epoch": 0.011417864053867102,
      "grad_norm": 4.413319110870361,
      "learning_rate": 9.885928546409809e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 2626160,
      "step": 163
    },
    {
      "epoch": 0.011487912299596347,
      "grad_norm": 4.626354694366455,
      "learning_rate": 9.885228721541156e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 2642464,
      "step": 164
    },
    {
      "epoch": 0.011557960545325593,
      "grad_norm": 4.328434467315674,
      "learning_rate": 9.884528896672505e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 2658528,
      "step": 165
    },
    {
      "epoch": 0.011628008791054838,
      "grad_norm": 4.57839822769165,
      "learning_rate": 9.883829071803853e-05,
      "loss": 1.0775,
      "num_input_tokens_seen": 2674912,
      "step": 166
    },
    {
      "epoch": 0.011698057036784086,
      "grad_norm": 5.103973865509033,
      "learning_rate": 9.883129246935202e-05,
      "loss": 1.2458,
      "num_input_tokens_seen": 2690792,
      "step": 167
    },
    {
      "epoch": 0.011768105282513331,
      "grad_norm": 4.558016300201416,
      "learning_rate": 9.88242942206655e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 2705616,
      "step": 168
    },
    {
      "epoch": 0.011838153528242577,
      "grad_norm": 4.811260223388672,
      "learning_rate": 9.8817295971979e-05,
      "loss": 1.2989,
      "num_input_tokens_seen": 2721704,
      "step": 169
    },
    {
      "epoch": 0.011908201773971823,
      "grad_norm": 4.726966857910156,
      "learning_rate": 9.881029772329248e-05,
      "loss": 1.176,
      "num_input_tokens_seen": 2738088,
      "step": 170
    },
    {
      "epoch": 0.01197825001970107,
      "grad_norm": 4.874902725219727,
      "learning_rate": 9.880329947460596e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 2754040,
      "step": 171
    },
    {
      "epoch": 0.012048298265430316,
      "grad_norm": 4.379549980163574,
      "learning_rate": 9.879630122591945e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 2770424,
      "step": 172
    },
    {
      "epoch": 0.012118346511159561,
      "grad_norm": 4.455331802368164,
      "learning_rate": 9.878930297723292e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 2786808,
      "step": 173
    },
    {
      "epoch": 0.012188394756888807,
      "grad_norm": 4.42273473739624,
      "learning_rate": 9.878230472854641e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 2803176,
      "step": 174
    },
    {
      "epoch": 0.012258443002618052,
      "grad_norm": 4.4078874588012695,
      "learning_rate": 9.87753064798599e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 2819448,
      "step": 175
    },
    {
      "epoch": 0.0123284912483473,
      "grad_norm": 4.79048490524292,
      "learning_rate": 9.876830823117339e-05,
      "loss": 1.3331,
      "num_input_tokens_seen": 2835832,
      "step": 176
    },
    {
      "epoch": 0.012398539494076545,
      "grad_norm": 4.212133884429932,
      "learning_rate": 9.876130998248688e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 2851776,
      "step": 177
    },
    {
      "epoch": 0.012468587739805791,
      "grad_norm": 5.7587738037109375,
      "learning_rate": 9.875431173380035e-05,
      "loss": 1.4729,
      "num_input_tokens_seen": 2867896,
      "step": 178
    },
    {
      "epoch": 0.012538635985535037,
      "grad_norm": 4.3469462394714355,
      "learning_rate": 9.874731348511384e-05,
      "loss": 0.957,
      "num_input_tokens_seen": 2884280,
      "step": 179
    },
    {
      "epoch": 0.012608684231264284,
      "grad_norm": 4.584625244140625,
      "learning_rate": 9.874031523642733e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 2899208,
      "step": 180
    },
    {
      "epoch": 0.01267873247699353,
      "grad_norm": 4.544627666473389,
      "learning_rate": 9.87333169877408e-05,
      "loss": 1.1706,
      "num_input_tokens_seen": 2915416,
      "step": 181
    },
    {
      "epoch": 0.012748780722722775,
      "grad_norm": 4.8749237060546875,
      "learning_rate": 9.872631873905429e-05,
      "loss": 1.3382,
      "num_input_tokens_seen": 2931360,
      "step": 182
    },
    {
      "epoch": 0.01281882896845202,
      "grad_norm": 4.593903541564941,
      "learning_rate": 9.871932049036778e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 2947744,
      "step": 183
    },
    {
      "epoch": 0.012888877214181268,
      "grad_norm": 4.478219509124756,
      "learning_rate": 9.871232224168127e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 2963664,
      "step": 184
    },
    {
      "epoch": 0.012958925459910514,
      "grad_norm": 5.028106212615967,
      "learning_rate": 9.870532399299476e-05,
      "loss": 1.3223,
      "num_input_tokens_seen": 2980048,
      "step": 185
    },
    {
      "epoch": 0.01302897370563976,
      "grad_norm": 4.866946697235107,
      "learning_rate": 9.869832574430823e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 2995992,
      "step": 186
    },
    {
      "epoch": 0.013099021951369005,
      "grad_norm": 4.421341419219971,
      "learning_rate": 9.869132749562172e-05,
      "loss": 1.2252,
      "num_input_tokens_seen": 3012000,
      "step": 187
    },
    {
      "epoch": 0.01316907019709825,
      "grad_norm": 4.88083028793335,
      "learning_rate": 9.86843292469352e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 3028384,
      "step": 188
    },
    {
      "epoch": 0.013239118442827498,
      "grad_norm": 4.654318809509277,
      "learning_rate": 9.86773309982487e-05,
      "loss": 1.2839,
      "num_input_tokens_seen": 3044768,
      "step": 189
    },
    {
      "epoch": 0.013309166688556744,
      "grad_norm": 4.626763820648193,
      "learning_rate": 9.867033274956219e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 3061152,
      "step": 190
    },
    {
      "epoch": 0.01337921493428599,
      "grad_norm": 4.178484916687012,
      "learning_rate": 9.866333450087566e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 3077056,
      "step": 191
    },
    {
      "epoch": 0.013449263180015235,
      "grad_norm": 4.755034923553467,
      "learning_rate": 9.865633625218915e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 3093400,
      "step": 192
    },
    {
      "epoch": 0.013519311425744482,
      "grad_norm": 4.437506198883057,
      "learning_rate": 9.864933800350263e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 3109784,
      "step": 193
    },
    {
      "epoch": 0.013589359671473728,
      "grad_norm": 5.140488624572754,
      "learning_rate": 9.864233975481611e-05,
      "loss": 1.4312,
      "num_input_tokens_seen": 3124976,
      "step": 194
    },
    {
      "epoch": 0.013659407917202973,
      "grad_norm": 4.72155237197876,
      "learning_rate": 9.86353415061296e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 3140632,
      "step": 195
    },
    {
      "epoch": 0.013729456162932219,
      "grad_norm": 4.914645671844482,
      "learning_rate": 9.862834325744309e-05,
      "loss": 1.2464,
      "num_input_tokens_seen": 3156616,
      "step": 196
    },
    {
      "epoch": 0.013799504408661466,
      "grad_norm": 4.23387336730957,
      "learning_rate": 9.862134500875658e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 3172840,
      "step": 197
    },
    {
      "epoch": 0.013869552654390712,
      "grad_norm": 4.659370422363281,
      "learning_rate": 9.861434676007005e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 3188584,
      "step": 198
    },
    {
      "epoch": 0.013939600900119958,
      "grad_norm": 4.580902576446533,
      "learning_rate": 9.860734851138354e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 3204432,
      "step": 199
    },
    {
      "epoch": 0.014009649145849203,
      "grad_norm": 4.208237648010254,
      "learning_rate": 9.860035026269702e-05,
      "loss": 1.2056,
      "num_input_tokens_seen": 3220816,
      "step": 200
    },
    {
      "epoch": 0.014009649145849203,
      "eval_loss": 1.2226407527923584,
      "eval_runtime": 0.3992,
      "eval_samples_per_second": 2.505,
      "eval_steps_per_second": 2.505,
      "num_input_tokens_seen": 3220816,
      "step": 200
    },
    {
      "epoch": 0.014079697391578449,
      "grad_norm": 4.526260852813721,
      "learning_rate": 9.85933520140105e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 3237200,
      "step": 201
    },
    {
      "epoch": 0.014149745637307696,
      "grad_norm": 4.46895170211792,
      "learning_rate": 9.8586353765324e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 3253336,
      "step": 202
    },
    {
      "epoch": 0.014219793883036942,
      "grad_norm": 4.367347717285156,
      "learning_rate": 9.857935551663748e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 3269632,
      "step": 203
    },
    {
      "epoch": 0.014289842128766187,
      "grad_norm": 4.860860347747803,
      "learning_rate": 9.857235726795097e-05,
      "loss": 1.4068,
      "num_input_tokens_seen": 3285432,
      "step": 204
    },
    {
      "epoch": 0.014359890374495433,
      "grad_norm": 4.336480617523193,
      "learning_rate": 9.856535901926445e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 3301632,
      "step": 205
    },
    {
      "epoch": 0.01442993862022468,
      "grad_norm": 4.587873458862305,
      "learning_rate": 9.855836077057794e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 3318016,
      "step": 206
    },
    {
      "epoch": 0.014499986865953926,
      "grad_norm": 4.719262599945068,
      "learning_rate": 9.855136252189142e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 3333168,
      "step": 207
    },
    {
      "epoch": 0.014570035111683172,
      "grad_norm": 4.419138431549072,
      "learning_rate": 9.85443642732049e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 3349384,
      "step": 208
    },
    {
      "epoch": 0.014640083357412417,
      "grad_norm": 4.3150835037231445,
      "learning_rate": 9.85373660245184e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 3365768,
      "step": 209
    },
    {
      "epoch": 0.014710131603141665,
      "grad_norm": 4.5917649269104,
      "learning_rate": 9.853036777583188e-05,
      "loss": 1.2821,
      "num_input_tokens_seen": 3382152,
      "step": 210
    },
    {
      "epoch": 0.01478017984887091,
      "grad_norm": 4.9094343185424805,
      "learning_rate": 9.852336952714537e-05,
      "loss": 1.2415,
      "num_input_tokens_seen": 3397896,
      "step": 211
    },
    {
      "epoch": 0.014850228094600156,
      "grad_norm": 4.394861698150635,
      "learning_rate": 9.851637127845885e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 3414280,
      "step": 212
    },
    {
      "epoch": 0.014920276340329401,
      "grad_norm": 4.196374416351318,
      "learning_rate": 9.850937302977233e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 3430584,
      "step": 213
    },
    {
      "epoch": 0.014990324586058647,
      "grad_norm": 4.728682518005371,
      "learning_rate": 9.850237478108582e-05,
      "loss": 1.2686,
      "num_input_tokens_seen": 3446968,
      "step": 214
    },
    {
      "epoch": 0.015060372831787894,
      "grad_norm": 4.291411876678467,
      "learning_rate": 9.84953765323993e-05,
      "loss": 1.1877,
      "num_input_tokens_seen": 3462568,
      "step": 215
    },
    {
      "epoch": 0.01513042107751714,
      "grad_norm": 4.405060768127441,
      "learning_rate": 9.84883782837128e-05,
      "loss": 1.2873,
      "num_input_tokens_seen": 3478952,
      "step": 216
    },
    {
      "epoch": 0.015200469323246386,
      "grad_norm": 4.254365921020508,
      "learning_rate": 9.848138003502628e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 3495304,
      "step": 217
    },
    {
      "epoch": 0.015270517568975631,
      "grad_norm": 4.741672039031982,
      "learning_rate": 9.847438178633976e-05,
      "loss": 1.1983,
      "num_input_tokens_seen": 3511688,
      "step": 218
    },
    {
      "epoch": 0.015340565814704879,
      "grad_norm": 4.352742671966553,
      "learning_rate": 9.846738353765325e-05,
      "loss": 1.2028,
      "num_input_tokens_seen": 3528072,
      "step": 219
    },
    {
      "epoch": 0.015410614060434124,
      "grad_norm": 4.996603488922119,
      "learning_rate": 9.846038528896672e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 3542904,
      "step": 220
    },
    {
      "epoch": 0.01548066230616337,
      "grad_norm": 4.911815166473389,
      "learning_rate": 9.845338704028021e-05,
      "loss": 1.3375,
      "num_input_tokens_seen": 3558352,
      "step": 221
    },
    {
      "epoch": 0.015550710551892616,
      "grad_norm": 4.638419151306152,
      "learning_rate": 9.84463887915937e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 3574736,
      "step": 222
    },
    {
      "epoch": 0.015620758797621863,
      "grad_norm": 4.323521614074707,
      "learning_rate": 9.843939054290719e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 3591120,
      "step": 223
    },
    {
      "epoch": 0.01569080704335111,
      "grad_norm": 4.466544151306152,
      "learning_rate": 9.843239229422068e-05,
      "loss": 1.3988,
      "num_input_tokens_seen": 3607392,
      "step": 224
    },
    {
      "epoch": 0.015760855289080354,
      "grad_norm": 4.476973533630371,
      "learning_rate": 9.842539404553415e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 3623776,
      "step": 225
    },
    {
      "epoch": 0.0158309035348096,
      "grad_norm": 4.648625373840332,
      "learning_rate": 9.841839579684764e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 3640008,
      "step": 226
    },
    {
      "epoch": 0.015900951780538845,
      "grad_norm": 4.364476203918457,
      "learning_rate": 9.841139754816112e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 3656392,
      "step": 227
    },
    {
      "epoch": 0.01597100002626809,
      "grad_norm": 4.3054633140563965,
      "learning_rate": 9.84043992994746e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 3672392,
      "step": 228
    },
    {
      "epoch": 0.016041048271997337,
      "grad_norm": 4.83436918258667,
      "learning_rate": 9.83974010507881e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 3688776,
      "step": 229
    },
    {
      "epoch": 0.016111096517726586,
      "grad_norm": 4.447519779205322,
      "learning_rate": 9.839040280210158e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 3705080,
      "step": 230
    },
    {
      "epoch": 0.01618114476345583,
      "grad_norm": 4.269217491149902,
      "learning_rate": 9.838340455341507e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 3721464,
      "step": 231
    },
    {
      "epoch": 0.016251193009185077,
      "grad_norm": 4.41223669052124,
      "learning_rate": 9.837640630472854e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 3737184,
      "step": 232
    },
    {
      "epoch": 0.016321241254914323,
      "grad_norm": 4.632737159729004,
      "learning_rate": 9.836940805604203e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 3753192,
      "step": 233
    },
    {
      "epoch": 0.016391289500643568,
      "grad_norm": 4.379425525665283,
      "learning_rate": 9.836240980735552e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 3767976,
      "step": 234
    },
    {
      "epoch": 0.016461337746372814,
      "grad_norm": 4.28551721572876,
      "learning_rate": 9.835541155866901e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 3784008,
      "step": 235
    },
    {
      "epoch": 0.01653138599210206,
      "grad_norm": 4.642453670501709,
      "learning_rate": 9.83484133099825e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 3800000,
      "step": 236
    },
    {
      "epoch": 0.016601434237831305,
      "grad_norm": 4.367178440093994,
      "learning_rate": 9.834141506129597e-05,
      "loss": 1.2877,
      "num_input_tokens_seen": 3816384,
      "step": 237
    },
    {
      "epoch": 0.01667148248356055,
      "grad_norm": 4.5724005699157715,
      "learning_rate": 9.833441681260946e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 3830328,
      "step": 238
    },
    {
      "epoch": 0.0167415307292898,
      "grad_norm": 4.318159580230713,
      "learning_rate": 9.832741856392295e-05,
      "loss": 1.1143,
      "num_input_tokens_seen": 3846712,
      "step": 239
    },
    {
      "epoch": 0.016811578975019045,
      "grad_norm": 4.408501625061035,
      "learning_rate": 9.832042031523643e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 3861776,
      "step": 240
    },
    {
      "epoch": 0.01688162722074829,
      "grad_norm": 4.20060920715332,
      "learning_rate": 9.831342206654991e-05,
      "loss": 1.209,
      "num_input_tokens_seen": 3877736,
      "step": 241
    },
    {
      "epoch": 0.016951675466477537,
      "grad_norm": 4.431649208068848,
      "learning_rate": 9.83064238178634e-05,
      "loss": 1.2458,
      "num_input_tokens_seen": 3893320,
      "step": 242
    },
    {
      "epoch": 0.017021723712206782,
      "grad_norm": 4.000490188598633,
      "learning_rate": 9.829942556917689e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 3909704,
      "step": 243
    },
    {
      "epoch": 0.017091771957936028,
      "grad_norm": 4.703495025634766,
      "learning_rate": 9.829242732049038e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 3925808,
      "step": 244
    },
    {
      "epoch": 0.017161820203665273,
      "grad_norm": 4.639338970184326,
      "learning_rate": 9.828542907180386e-05,
      "loss": 1.3046,
      "num_input_tokens_seen": 3942192,
      "step": 245
    },
    {
      "epoch": 0.01723186844939452,
      "grad_norm": 4.414276599884033,
      "learning_rate": 9.827843082311734e-05,
      "loss": 1.271,
      "num_input_tokens_seen": 3958528,
      "step": 246
    },
    {
      "epoch": 0.017301916695123768,
      "grad_norm": 4.404853820800781,
      "learning_rate": 9.827143257443082e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 3974912,
      "step": 247
    },
    {
      "epoch": 0.017371964940853014,
      "grad_norm": 4.519491195678711,
      "learning_rate": 9.826443432574431e-05,
      "loss": 1.2894,
      "num_input_tokens_seen": 3991296,
      "step": 248
    },
    {
      "epoch": 0.01744201318658226,
      "grad_norm": 4.261727809906006,
      "learning_rate": 9.825743607705781e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 4006544,
      "step": 249
    },
    {
      "epoch": 0.017512061432311505,
      "grad_norm": 4.102485656738281,
      "learning_rate": 9.825043782837129e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 4022320,
      "step": 250
    },
    {
      "epoch": 0.01758210967804075,
      "grad_norm": 4.804764270782471,
      "learning_rate": 9.824343957968477e-05,
      "loss": 1.3344,
      "num_input_tokens_seen": 4037048,
      "step": 251
    },
    {
      "epoch": 0.017652157923769996,
      "grad_norm": 4.130600452423096,
      "learning_rate": 9.823644133099825e-05,
      "loss": 1.2349,
      "num_input_tokens_seen": 4053432,
      "step": 252
    },
    {
      "epoch": 0.017722206169499242,
      "grad_norm": 4.234742641448975,
      "learning_rate": 9.822944308231174e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 4069816,
      "step": 253
    },
    {
      "epoch": 0.017792254415228487,
      "grad_norm": 4.754928112030029,
      "learning_rate": 9.822244483362521e-05,
      "loss": 1.5168,
      "num_input_tokens_seen": 4085864,
      "step": 254
    },
    {
      "epoch": 0.017862302660957733,
      "grad_norm": 4.542768478393555,
      "learning_rate": 9.821544658493871e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 4102240,
      "step": 255
    },
    {
      "epoch": 0.017932350906686982,
      "grad_norm": 4.411310195922852,
      "learning_rate": 9.82084483362522e-05,
      "loss": 1.2694,
      "num_input_tokens_seen": 4118544,
      "step": 256
    },
    {
      "epoch": 0.018002399152416228,
      "grad_norm": 4.205377101898193,
      "learning_rate": 9.820145008756568e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 4134928,
      "step": 257
    },
    {
      "epoch": 0.018072447398145473,
      "grad_norm": 4.451165199279785,
      "learning_rate": 9.819445183887917e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 4150848,
      "step": 258
    },
    {
      "epoch": 0.01814249564387472,
      "grad_norm": 4.366336822509766,
      "learning_rate": 9.818745359019264e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 4167184,
      "step": 259
    },
    {
      "epoch": 0.018212543889603965,
      "grad_norm": 4.394649982452393,
      "learning_rate": 9.818045534150613e-05,
      "loss": 1.0741,
      "num_input_tokens_seen": 4183376,
      "step": 260
    },
    {
      "epoch": 0.01828259213533321,
      "grad_norm": 4.344518184661865,
      "learning_rate": 9.817345709281962e-05,
      "loss": 1.2282,
      "num_input_tokens_seen": 4199760,
      "step": 261
    },
    {
      "epoch": 0.018352640381062456,
      "grad_norm": 4.403041362762451,
      "learning_rate": 9.816645884413311e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 4215816,
      "step": 262
    },
    {
      "epoch": 0.0184226886267917,
      "grad_norm": 4.715320110321045,
      "learning_rate": 9.81594605954466e-05,
      "loss": 1.3074,
      "num_input_tokens_seen": 4231504,
      "step": 263
    },
    {
      "epoch": 0.01849273687252095,
      "grad_norm": 4.5754265785217285,
      "learning_rate": 9.815246234676007e-05,
      "loss": 1.253,
      "num_input_tokens_seen": 4247888,
      "step": 264
    },
    {
      "epoch": 0.018562785118250196,
      "grad_norm": 4.2346930503845215,
      "learning_rate": 9.814546409807356e-05,
      "loss": 1.1727,
      "num_input_tokens_seen": 4264248,
      "step": 265
    },
    {
      "epoch": 0.018632833363979442,
      "grad_norm": 4.186713218688965,
      "learning_rate": 9.813846584938705e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 4280632,
      "step": 266
    },
    {
      "epoch": 0.018702881609708687,
      "grad_norm": 4.6356706619262695,
      "learning_rate": 9.813146760070052e-05,
      "loss": 1.3755,
      "num_input_tokens_seen": 4296648,
      "step": 267
    },
    {
      "epoch": 0.018772929855437933,
      "grad_norm": 4.466466903686523,
      "learning_rate": 9.812446935201401e-05,
      "loss": 1.283,
      "num_input_tokens_seen": 4311408,
      "step": 268
    },
    {
      "epoch": 0.01884297810116718,
      "grad_norm": 4.3369140625,
      "learning_rate": 9.81174711033275e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 4326736,
      "step": 269
    },
    {
      "epoch": 0.018913026346896424,
      "grad_norm": 4.434782028198242,
      "learning_rate": 9.811047285464099e-05,
      "loss": 1.2859,
      "num_input_tokens_seen": 4343120,
      "step": 270
    },
    {
      "epoch": 0.01898307459262567,
      "grad_norm": 4.346708297729492,
      "learning_rate": 9.810347460595448e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 4359504,
      "step": 271
    },
    {
      "epoch": 0.019053122838354915,
      "grad_norm": 4.529878616333008,
      "learning_rate": 9.809647635726795e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 4375888,
      "step": 272
    },
    {
      "epoch": 0.019123171084084165,
      "grad_norm": 4.051745891571045,
      "learning_rate": 9.808947810858144e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 4392224,
      "step": 273
    },
    {
      "epoch": 0.01919321932981341,
      "grad_norm": 4.403522491455078,
      "learning_rate": 9.808247985989492e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 4408608,
      "step": 274
    },
    {
      "epoch": 0.019263267575542656,
      "grad_norm": 4.166261196136475,
      "learning_rate": 9.807548161120842e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 4424992,
      "step": 275
    },
    {
      "epoch": 0.0193333158212719,
      "grad_norm": 4.29187536239624,
      "learning_rate": 9.806848336252191e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 4441376,
      "step": 276
    },
    {
      "epoch": 0.019403364067001147,
      "grad_norm": 4.4056172370910645,
      "learning_rate": 9.806148511383538e-05,
      "loss": 1.1965,
      "num_input_tokens_seen": 4457760,
      "step": 277
    },
    {
      "epoch": 0.019473412312730393,
      "grad_norm": 4.355875015258789,
      "learning_rate": 9.805448686514887e-05,
      "loss": 1.1024,
      "num_input_tokens_seen": 4474144,
      "step": 278
    },
    {
      "epoch": 0.019543460558459638,
      "grad_norm": 4.46420955657959,
      "learning_rate": 9.804748861646235e-05,
      "loss": 1.203,
      "num_input_tokens_seen": 4488912,
      "step": 279
    },
    {
      "epoch": 0.019613508804188884,
      "grad_norm": 4.48052453994751,
      "learning_rate": 9.804049036777583e-05,
      "loss": 1.2089,
      "num_input_tokens_seen": 4505296,
      "step": 280
    },
    {
      "epoch": 0.01968355704991813,
      "grad_norm": 4.458749294281006,
      "learning_rate": 9.803349211908932e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 4520576,
      "step": 281
    },
    {
      "epoch": 0.01975360529564738,
      "grad_norm": 4.551771640777588,
      "learning_rate": 9.802649387040281e-05,
      "loss": 1.1671,
      "num_input_tokens_seen": 4536960,
      "step": 282
    },
    {
      "epoch": 0.019823653541376624,
      "grad_norm": 4.038064956665039,
      "learning_rate": 9.80194956217163e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 4553344,
      "step": 283
    },
    {
      "epoch": 0.01989370178710587,
      "grad_norm": 4.647075653076172,
      "learning_rate": 9.801249737302978e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 4568928,
      "step": 284
    },
    {
      "epoch": 0.019963750032835115,
      "grad_norm": 4.258941650390625,
      "learning_rate": 9.800549912434326e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 4585312,
      "step": 285
    },
    {
      "epoch": 0.02003379827856436,
      "grad_norm": 4.348769664764404,
      "learning_rate": 9.799850087565674e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 4601696,
      "step": 286
    },
    {
      "epoch": 0.020103846524293607,
      "grad_norm": 4.105901718139648,
      "learning_rate": 9.799150262697023e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 4617312,
      "step": 287
    },
    {
      "epoch": 0.020173894770022852,
      "grad_norm": 4.079495429992676,
      "learning_rate": 9.798450437828372e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 4633696,
      "step": 288
    },
    {
      "epoch": 0.020243943015752098,
      "grad_norm": 4.03472375869751,
      "learning_rate": 9.79775061295972e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 4650080,
      "step": 289
    },
    {
      "epoch": 0.020313991261481347,
      "grad_norm": 4.077049732208252,
      "learning_rate": 9.797050788091069e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 4666328,
      "step": 290
    },
    {
      "epoch": 0.020384039507210593,
      "grad_norm": 4.086606025695801,
      "learning_rate": 9.796350963222417e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 4682256,
      "step": 291
    },
    {
      "epoch": 0.020454087752939838,
      "grad_norm": 4.296900749206543,
      "learning_rate": 9.795651138353766e-05,
      "loss": 1.2964,
      "num_input_tokens_seen": 4698640,
      "step": 292
    },
    {
      "epoch": 0.020524135998669084,
      "grad_norm": 4.040759086608887,
      "learning_rate": 9.794951313485115e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 4714928,
      "step": 293
    },
    {
      "epoch": 0.02059418424439833,
      "grad_norm": 3.8260273933410645,
      "learning_rate": 9.794251488616462e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 4731312,
      "step": 294
    },
    {
      "epoch": 0.020664232490127575,
      "grad_norm": 4.294517993927002,
      "learning_rate": 9.793551663747811e-05,
      "loss": 1.2704,
      "num_input_tokens_seen": 4747544,
      "step": 295
    },
    {
      "epoch": 0.02073428073585682,
      "grad_norm": 4.206037521362305,
      "learning_rate": 9.79285183887916e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 4763928,
      "step": 296
    },
    {
      "epoch": 0.020804328981586066,
      "grad_norm": 4.147867202758789,
      "learning_rate": 9.792152014010509e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 4780312,
      "step": 297
    },
    {
      "epoch": 0.020874377227315312,
      "grad_norm": 4.23718786239624,
      "learning_rate": 9.791452189141857e-05,
      "loss": 1.2353,
      "num_input_tokens_seen": 4796384,
      "step": 298
    },
    {
      "epoch": 0.02094442547304456,
      "grad_norm": 4.172685146331787,
      "learning_rate": 9.790752364273205e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 4812768,
      "step": 299
    },
    {
      "epoch": 0.021014473718773807,
      "grad_norm": 4.167289733886719,
      "learning_rate": 9.790052539404554e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 4829152,
      "step": 300
    },
    {
      "epoch": 0.021084521964503052,
      "grad_norm": 4.096963882446289,
      "learning_rate": 9.789352714535903e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 4845384,
      "step": 301
    },
    {
      "epoch": 0.021154570210232298,
      "grad_norm": 4.223779678344727,
      "learning_rate": 9.788652889667252e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 4861768,
      "step": 302
    },
    {
      "epoch": 0.021224618455961543,
      "grad_norm": 3.8243472576141357,
      "learning_rate": 9.7879530647986e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 4878152,
      "step": 303
    },
    {
      "epoch": 0.02129466670169079,
      "grad_norm": 4.092590808868408,
      "learning_rate": 9.787253239929948e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 4894536,
      "step": 304
    },
    {
      "epoch": 0.021364714947420035,
      "grad_norm": 4.42412805557251,
      "learning_rate": 9.786553415061297e-05,
      "loss": 0.9915,
      "num_input_tokens_seen": 4910320,
      "step": 305
    },
    {
      "epoch": 0.02143476319314928,
      "grad_norm": 4.488316535949707,
      "learning_rate": 9.785853590192644e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 4926704,
      "step": 306
    },
    {
      "epoch": 0.021504811438878526,
      "grad_norm": 4.110256195068359,
      "learning_rate": 9.785153765323993e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 4943088,
      "step": 307
    },
    {
      "epoch": 0.021574859684607775,
      "grad_norm": 4.246950149536133,
      "learning_rate": 9.784453940455342e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 4958736,
      "step": 308
    },
    {
      "epoch": 0.02164490793033702,
      "grad_norm": 4.175214767456055,
      "learning_rate": 9.783754115586691e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 4975120,
      "step": 309
    },
    {
      "epoch": 0.021714956176066266,
      "grad_norm": 4.427795886993408,
      "learning_rate": 9.78305429071804e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 4991504,
      "step": 310
    },
    {
      "epoch": 0.021785004421795512,
      "grad_norm": 4.158191204071045,
      "learning_rate": 9.782354465849387e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 5007152,
      "step": 311
    },
    {
      "epoch": 0.021855052667524758,
      "grad_norm": 4.184347629547119,
      "learning_rate": 9.781654640980736e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 5023536,
      "step": 312
    },
    {
      "epoch": 0.021925100913254003,
      "grad_norm": 3.8506295680999756,
      "learning_rate": 9.780954816112084e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 5039728,
      "step": 313
    },
    {
      "epoch": 0.02199514915898325,
      "grad_norm": 4.310062408447266,
      "learning_rate": 9.780254991243432e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 5056008,
      "step": 314
    },
    {
      "epoch": 0.022065197404712494,
      "grad_norm": 4.215006351470947,
      "learning_rate": 9.779555166374781e-05,
      "loss": 1.1715,
      "num_input_tokens_seen": 5072096,
      "step": 315
    },
    {
      "epoch": 0.022135245650441743,
      "grad_norm": 4.219073295593262,
      "learning_rate": 9.77885534150613e-05,
      "loss": 1.219,
      "num_input_tokens_seen": 5088432,
      "step": 316
    },
    {
      "epoch": 0.02220529389617099,
      "grad_norm": 4.319522857666016,
      "learning_rate": 9.778155516637479e-05,
      "loss": 1.3085,
      "num_input_tokens_seen": 5104240,
      "step": 317
    },
    {
      "epoch": 0.022275342141900235,
      "grad_norm": 4.118961334228516,
      "learning_rate": 9.777455691768827e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 5120624,
      "step": 318
    },
    {
      "epoch": 0.02234539038762948,
      "grad_norm": 4.195051193237305,
      "learning_rate": 9.776755866900175e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 5137008,
      "step": 319
    },
    {
      "epoch": 0.022415438633358726,
      "grad_norm": 4.114197254180908,
      "learning_rate": 9.776056042031524e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 5153272,
      "step": 320
    },
    {
      "epoch": 0.02248548687908797,
      "grad_norm": 4.014908313751221,
      "learning_rate": 9.775356217162872e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 5169472,
      "step": 321
    },
    {
      "epoch": 0.022555535124817217,
      "grad_norm": 4.190642356872559,
      "learning_rate": 9.774656392294222e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 5185856,
      "step": 322
    },
    {
      "epoch": 0.022625583370546463,
      "grad_norm": 4.562993049621582,
      "learning_rate": 9.77395656742557e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 5202240,
      "step": 323
    },
    {
      "epoch": 0.02269563161627571,
      "grad_norm": 4.607022762298584,
      "learning_rate": 9.773256742556918e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 5218168,
      "step": 324
    },
    {
      "epoch": 0.022765679862004957,
      "grad_norm": 3.956439256668091,
      "learning_rate": 9.772556917688267e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 5234368,
      "step": 325
    },
    {
      "epoch": 0.022835728107734203,
      "grad_norm": 4.20713472366333,
      "learning_rate": 9.771857092819615e-05,
      "loss": 1.2332,
      "num_input_tokens_seen": 5249808,
      "step": 326
    },
    {
      "epoch": 0.02290577635346345,
      "grad_norm": 4.4092864990234375,
      "learning_rate": 9.771157267950964e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 5266120,
      "step": 327
    },
    {
      "epoch": 0.022975824599192694,
      "grad_norm": 4.529845237731934,
      "learning_rate": 9.770457443082312e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 5282504,
      "step": 328
    },
    {
      "epoch": 0.02304587284492194,
      "grad_norm": 4.221986293792725,
      "learning_rate": 9.769757618213661e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 5298344,
      "step": 329
    },
    {
      "epoch": 0.023115921090651186,
      "grad_norm": 4.29000186920166,
      "learning_rate": 9.76905779334501e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 5314728,
      "step": 330
    },
    {
      "epoch": 0.02318596933638043,
      "grad_norm": 4.426812648773193,
      "learning_rate": 9.768357968476358e-05,
      "loss": 1.514,
      "num_input_tokens_seen": 5330816,
      "step": 331
    },
    {
      "epoch": 0.023256017582109677,
      "grad_norm": 4.210752964019775,
      "learning_rate": 9.767658143607706e-05,
      "loss": 1.0854,
      "num_input_tokens_seen": 5346552,
      "step": 332
    },
    {
      "epoch": 0.023326065827838922,
      "grad_norm": 4.216427326202393,
      "learning_rate": 9.766958318739054e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 5362936,
      "step": 333
    },
    {
      "epoch": 0.02339611407356817,
      "grad_norm": 4.132325649261475,
      "learning_rate": 9.766258493870403e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 5379320,
      "step": 334
    },
    {
      "epoch": 0.023466162319297417,
      "grad_norm": 4.277027130126953,
      "learning_rate": 9.765558669001752e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 5395704,
      "step": 335
    },
    {
      "epoch": 0.023536210565026663,
      "grad_norm": 4.228096961975098,
      "learning_rate": 9.7648588441331e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 5412088,
      "step": 336
    },
    {
      "epoch": 0.02360625881075591,
      "grad_norm": 4.194522380828857,
      "learning_rate": 9.76415901926445e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 5428472,
      "step": 337
    },
    {
      "epoch": 0.023676307056485154,
      "grad_norm": 4.336326599121094,
      "learning_rate": 9.763459194395797e-05,
      "loss": 1.2251,
      "num_input_tokens_seen": 5444856,
      "step": 338
    },
    {
      "epoch": 0.0237463553022144,
      "grad_norm": 4.2723307609558105,
      "learning_rate": 9.762759369527146e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 5460304,
      "step": 339
    },
    {
      "epoch": 0.023816403547943645,
      "grad_norm": 4.190036773681641,
      "learning_rate": 9.762059544658493e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 5476688,
      "step": 340
    },
    {
      "epoch": 0.02388645179367289,
      "grad_norm": 4.477560043334961,
      "learning_rate": 9.761359719789842e-05,
      "loss": 1.362,
      "num_input_tokens_seen": 5493072,
      "step": 341
    },
    {
      "epoch": 0.02395650003940214,
      "grad_norm": 4.160232067108154,
      "learning_rate": 9.760659894921192e-05,
      "loss": 1.1602,
      "num_input_tokens_seen": 5509456,
      "step": 342
    },
    {
      "epoch": 0.024026548285131386,
      "grad_norm": 3.857335090637207,
      "learning_rate": 9.75996007005254e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 5525840,
      "step": 343
    },
    {
      "epoch": 0.02409659653086063,
      "grad_norm": 4.141246318817139,
      "learning_rate": 9.759260245183889e-05,
      "loss": 1.2009,
      "num_input_tokens_seen": 5541888,
      "step": 344
    },
    {
      "epoch": 0.024166644776589877,
      "grad_norm": 4.50364875793457,
      "learning_rate": 9.758560420315236e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 5557848,
      "step": 345
    },
    {
      "epoch": 0.024236693022319122,
      "grad_norm": 4.3343353271484375,
      "learning_rate": 9.757860595446585e-05,
      "loss": 1.3594,
      "num_input_tokens_seen": 5573504,
      "step": 346
    },
    {
      "epoch": 0.024306741268048368,
      "grad_norm": 4.050408363342285,
      "learning_rate": 9.757160770577934e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 5589544,
      "step": 347
    },
    {
      "epoch": 0.024376789513777614,
      "grad_norm": 4.051811695098877,
      "learning_rate": 9.756460945709283e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 5605368,
      "step": 348
    },
    {
      "epoch": 0.02444683775950686,
      "grad_norm": 4.365113258361816,
      "learning_rate": 9.755761120840632e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 5621752,
      "step": 349
    },
    {
      "epoch": 0.024516886005236105,
      "grad_norm": 4.0057501792907715,
      "learning_rate": 9.755061295971979e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 5638136,
      "step": 350
    },
    {
      "epoch": 0.024586934250965354,
      "grad_norm": 4.254896640777588,
      "learning_rate": 9.754361471103328e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 5653168,
      "step": 351
    },
    {
      "epoch": 0.0246569824966946,
      "grad_norm": 3.8119771480560303,
      "learning_rate": 9.753661646234677e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 5669504,
      "step": 352
    },
    {
      "epoch": 0.024727030742423845,
      "grad_norm": 4.5082621574401855,
      "learning_rate": 9.752961821366024e-05,
      "loss": 1.2537,
      "num_input_tokens_seen": 5685168,
      "step": 353
    },
    {
      "epoch": 0.02479707898815309,
      "grad_norm": 4.392731189727783,
      "learning_rate": 9.752261996497373e-05,
      "loss": 1.2534,
      "num_input_tokens_seen": 5701240,
      "step": 354
    },
    {
      "epoch": 0.024867127233882336,
      "grad_norm": 4.293395519256592,
      "learning_rate": 9.751562171628722e-05,
      "loss": 1.2774,
      "num_input_tokens_seen": 5717624,
      "step": 355
    },
    {
      "epoch": 0.024937175479611582,
      "grad_norm": 4.64813756942749,
      "learning_rate": 9.750862346760071e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 5733104,
      "step": 356
    },
    {
      "epoch": 0.025007223725340828,
      "grad_norm": 4.5166778564453125,
      "learning_rate": 9.75016252189142e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 5749488,
      "step": 357
    },
    {
      "epoch": 0.025077271971070073,
      "grad_norm": 3.894291400909424,
      "learning_rate": 9.749462697022767e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 5765872,
      "step": 358
    },
    {
      "epoch": 0.02514732021679932,
      "grad_norm": 4.10056209564209,
      "learning_rate": 9.748762872154116e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 5780856,
      "step": 359
    },
    {
      "epoch": 0.025217368462528568,
      "grad_norm": 4.6277666091918945,
      "learning_rate": 9.748063047285464e-05,
      "loss": 1.3649,
      "num_input_tokens_seen": 5796856,
      "step": 360
    },
    {
      "epoch": 0.025287416708257814,
      "grad_norm": 4.029720306396484,
      "learning_rate": 9.747363222416813e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 5812176,
      "step": 361
    },
    {
      "epoch": 0.02535746495398706,
      "grad_norm": 3.7772202491760254,
      "learning_rate": 9.746663397548161e-05,
      "loss": 1.0448,
      "num_input_tokens_seen": 5828064,
      "step": 362
    },
    {
      "epoch": 0.025427513199716305,
      "grad_norm": 4.379861354827881,
      "learning_rate": 9.74596357267951e-05,
      "loss": 1.3274,
      "num_input_tokens_seen": 5843680,
      "step": 363
    },
    {
      "epoch": 0.02549756144544555,
      "grad_norm": 4.254587173461914,
      "learning_rate": 9.745263747810859e-05,
      "loss": 1.1502,
      "num_input_tokens_seen": 5859024,
      "step": 364
    },
    {
      "epoch": 0.025567609691174796,
      "grad_norm": 4.271276473999023,
      "learning_rate": 9.744563922942207e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 5874320,
      "step": 365
    },
    {
      "epoch": 0.02563765793690404,
      "grad_norm": 4.224324703216553,
      "learning_rate": 9.743864098073555e-05,
      "loss": 1.0926,
      "num_input_tokens_seen": 5890704,
      "step": 366
    },
    {
      "epoch": 0.025707706182633287,
      "grad_norm": 4.289444446563721,
      "learning_rate": 9.743164273204903e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 5906016,
      "step": 367
    },
    {
      "epoch": 0.025777754428362536,
      "grad_norm": 4.280707359313965,
      "learning_rate": 9.742464448336253e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 5921784,
      "step": 368
    },
    {
      "epoch": 0.025847802674091782,
      "grad_norm": 4.554803848266602,
      "learning_rate": 9.741764623467602e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 5938072,
      "step": 369
    },
    {
      "epoch": 0.025917850919821028,
      "grad_norm": 4.677784442901611,
      "learning_rate": 9.74106479859895e-05,
      "loss": 1.2387,
      "num_input_tokens_seen": 5954456,
      "step": 370
    },
    {
      "epoch": 0.025987899165550273,
      "grad_norm": 4.268225193023682,
      "learning_rate": 9.740364973730298e-05,
      "loss": 1.2983,
      "num_input_tokens_seen": 5970664,
      "step": 371
    },
    {
      "epoch": 0.02605794741127952,
      "grad_norm": 4.361818790435791,
      "learning_rate": 9.739665148861646e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 5987048,
      "step": 372
    },
    {
      "epoch": 0.026127995657008764,
      "grad_norm": 3.9990735054016113,
      "learning_rate": 9.738965323992995e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 6003432,
      "step": 373
    },
    {
      "epoch": 0.02619804390273801,
      "grad_norm": 3.992142915725708,
      "learning_rate": 9.738265499124344e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 6019816,
      "step": 374
    },
    {
      "epoch": 0.026268092148467256,
      "grad_norm": 4.270167827606201,
      "learning_rate": 9.737565674255693e-05,
      "loss": 1.1764,
      "num_input_tokens_seen": 6036200,
      "step": 375
    },
    {
      "epoch": 0.0263381403941965,
      "grad_norm": 4.362086296081543,
      "learning_rate": 9.736865849387041e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 6052120,
      "step": 376
    },
    {
      "epoch": 0.02640818863992575,
      "grad_norm": 3.6900475025177,
      "learning_rate": 9.736166024518389e-05,
      "loss": 0.8729,
      "num_input_tokens_seen": 6068264,
      "step": 377
    },
    {
      "epoch": 0.026478236885654996,
      "grad_norm": 3.8281285762786865,
      "learning_rate": 9.735466199649738e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 6084504,
      "step": 378
    },
    {
      "epoch": 0.02654828513138424,
      "grad_norm": 3.9335553646087646,
      "learning_rate": 9.734766374781087e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 6100592,
      "step": 379
    },
    {
      "epoch": 0.026618333377113487,
      "grad_norm": 4.332645416259766,
      "learning_rate": 9.734066549912434e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 6116976,
      "step": 380
    },
    {
      "epoch": 0.026688381622842733,
      "grad_norm": 4.160863399505615,
      "learning_rate": 9.733366725043783e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 6133360,
      "step": 381
    },
    {
      "epoch": 0.02675842986857198,
      "grad_norm": 4.388178825378418,
      "learning_rate": 9.732666900175132e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 6149744,
      "step": 382
    },
    {
      "epoch": 0.026828478114301224,
      "grad_norm": 4.354910373687744,
      "learning_rate": 9.73196707530648e-05,
      "loss": 1.4115,
      "num_input_tokens_seen": 6166048,
      "step": 383
    },
    {
      "epoch": 0.02689852636003047,
      "grad_norm": 4.058071613311768,
      "learning_rate": 9.73126725043783e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 6181840,
      "step": 384
    },
    {
      "epoch": 0.026968574605759715,
      "grad_norm": 4.060855865478516,
      "learning_rate": 9.730567425569177e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 6198224,
      "step": 385
    },
    {
      "epoch": 0.027038622851488964,
      "grad_norm": 4.316681385040283,
      "learning_rate": 9.729867600700526e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 6214608,
      "step": 386
    },
    {
      "epoch": 0.02710867109721821,
      "grad_norm": 4.322516918182373,
      "learning_rate": 9.729167775831873e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 6230992,
      "step": 387
    },
    {
      "epoch": 0.027178719342947456,
      "grad_norm": 4.090857028961182,
      "learning_rate": 9.728467950963224e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 6246760,
      "step": 388
    },
    {
      "epoch": 0.0272487675886767,
      "grad_norm": 4.0143961906433105,
      "learning_rate": 9.727768126094571e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 6261968,
      "step": 389
    },
    {
      "epoch": 0.027318815834405947,
      "grad_norm": 4.911194324493408,
      "learning_rate": 9.72706830122592e-05,
      "loss": 1.3016,
      "num_input_tokens_seen": 6276664,
      "step": 390
    },
    {
      "epoch": 0.027388864080135192,
      "grad_norm": 4.057498931884766,
      "learning_rate": 9.726368476357269e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 6293048,
      "step": 391
    },
    {
      "epoch": 0.027458912325864438,
      "grad_norm": 3.9827401638031006,
      "learning_rate": 9.725668651488616e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 6309432,
      "step": 392
    },
    {
      "epoch": 0.027528960571593684,
      "grad_norm": 4.640822887420654,
      "learning_rate": 9.724968826619965e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 6325568,
      "step": 393
    },
    {
      "epoch": 0.027599008817322933,
      "grad_norm": 4.372538089752197,
      "learning_rate": 9.724269001751314e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 6341952,
      "step": 394
    },
    {
      "epoch": 0.02766905706305218,
      "grad_norm": 4.018289566040039,
      "learning_rate": 9.723569176882663e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 6358336,
      "step": 395
    },
    {
      "epoch": 0.027739105308781424,
      "grad_norm": 4.440858364105225,
      "learning_rate": 9.722869352014012e-05,
      "loss": 1.2272,
      "num_input_tokens_seen": 6374680,
      "step": 396
    },
    {
      "epoch": 0.02780915355451067,
      "grad_norm": 4.246788024902344,
      "learning_rate": 9.722169527145359e-05,
      "loss": 1.0161,
      "num_input_tokens_seen": 6390672,
      "step": 397
    },
    {
      "epoch": 0.027879201800239915,
      "grad_norm": 4.27274751663208,
      "learning_rate": 9.721469702276708e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 6407056,
      "step": 398
    },
    {
      "epoch": 0.02794925004596916,
      "grad_norm": 4.171760559082031,
      "learning_rate": 9.720769877408056e-05,
      "loss": 1.2766,
      "num_input_tokens_seen": 6423440,
      "step": 399
    },
    {
      "epoch": 0.028019298291698407,
      "grad_norm": 4.174622535705566,
      "learning_rate": 9.720070052539405e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 6439824,
      "step": 400
    },
    {
      "epoch": 0.028019298291698407,
      "eval_loss": 1.1994441747665405,
      "eval_runtime": 0.2131,
      "eval_samples_per_second": 4.693,
      "eval_steps_per_second": 4.693,
      "num_input_tokens_seen": 6439824,
      "step": 400
    },
    {
      "epoch": 0.028089346537427652,
      "grad_norm": 4.199150562286377,
      "learning_rate": 9.719370227670753e-05,
      "loss": 1.3432,
      "num_input_tokens_seen": 6456208,
      "step": 401
    },
    {
      "epoch": 0.028159394783156898,
      "grad_norm": 3.9011733531951904,
      "learning_rate": 9.718670402802102e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 6472592,
      "step": 402
    },
    {
      "epoch": 0.028229443028886147,
      "grad_norm": 4.142306327819824,
      "learning_rate": 9.717970577933451e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 6488976,
      "step": 403
    },
    {
      "epoch": 0.028299491274615392,
      "grad_norm": 3.9745633602142334,
      "learning_rate": 9.717270753064799e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 6505360,
      "step": 404
    },
    {
      "epoch": 0.028369539520344638,
      "grad_norm": 3.838865280151367,
      "learning_rate": 9.716570928196147e-05,
      "loss": 0.809,
      "num_input_tokens_seen": 6521744,
      "step": 405
    },
    {
      "epoch": 0.028439587766073884,
      "grad_norm": 4.48146390914917,
      "learning_rate": 9.715871103327496e-05,
      "loss": 1.4985,
      "num_input_tokens_seen": 6538128,
      "step": 406
    },
    {
      "epoch": 0.02850963601180313,
      "grad_norm": 4.393556594848633,
      "learning_rate": 9.715171278458844e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 6554512,
      "step": 407
    },
    {
      "epoch": 0.028579684257532375,
      "grad_norm": 3.970860004425049,
      "learning_rate": 9.714471453590194e-05,
      "loss": 1.1513,
      "num_input_tokens_seen": 6570896,
      "step": 408
    },
    {
      "epoch": 0.02864973250326162,
      "grad_norm": 4.166610240936279,
      "learning_rate": 9.713771628721542e-05,
      "loss": 1.108,
      "num_input_tokens_seen": 6587216,
      "step": 409
    },
    {
      "epoch": 0.028719780748990866,
      "grad_norm": 3.9887096881866455,
      "learning_rate": 9.71307180385289e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 6603600,
      "step": 410
    },
    {
      "epoch": 0.028789828994720112,
      "grad_norm": 4.195802211761475,
      "learning_rate": 9.712371978984239e-05,
      "loss": 1.1478,
      "num_input_tokens_seen": 6619984,
      "step": 411
    },
    {
      "epoch": 0.02885987724044936,
      "grad_norm": 4.011331081390381,
      "learning_rate": 9.711672154115587e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 6635904,
      "step": 412
    },
    {
      "epoch": 0.028929925486178606,
      "grad_norm": 4.4170026779174805,
      "learning_rate": 9.710972329246936e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 6651944,
      "step": 413
    },
    {
      "epoch": 0.028999973731907852,
      "grad_norm": 4.073450088500977,
      "learning_rate": 9.710272504378284e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 6668096,
      "step": 414
    },
    {
      "epoch": 0.029070021977637098,
      "grad_norm": 4.161722183227539,
      "learning_rate": 9.709572679509633e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 6684480,
      "step": 415
    },
    {
      "epoch": 0.029140070223366343,
      "grad_norm": 4.540097713470459,
      "learning_rate": 9.708872854640981e-05,
      "loss": 1.2143,
      "num_input_tokens_seen": 6700536,
      "step": 416
    },
    {
      "epoch": 0.02921011846909559,
      "grad_norm": 4.030871868133545,
      "learning_rate": 9.70817302977233e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 6716920,
      "step": 417
    },
    {
      "epoch": 0.029280166714824835,
      "grad_norm": 4.1743268966674805,
      "learning_rate": 9.707473204903679e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 6733304,
      "step": 418
    },
    {
      "epoch": 0.02935021496055408,
      "grad_norm": 4.227272987365723,
      "learning_rate": 9.706773380035026e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 6749688,
      "step": 419
    },
    {
      "epoch": 0.02942026320628333,
      "grad_norm": 4.406428813934326,
      "learning_rate": 9.706073555166375e-05,
      "loss": 1.0302,
      "num_input_tokens_seen": 6766072,
      "step": 420
    },
    {
      "epoch": 0.029490311452012575,
      "grad_norm": 4.17899227142334,
      "learning_rate": 9.705373730297724e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 6782456,
      "step": 421
    },
    {
      "epoch": 0.02956035969774182,
      "grad_norm": 4.034752368927002,
      "learning_rate": 9.704673905429073e-05,
      "loss": 1.2639,
      "num_input_tokens_seen": 6798840,
      "step": 422
    },
    {
      "epoch": 0.029630407943471066,
      "grad_norm": 4.795727729797363,
      "learning_rate": 9.703974080560421e-05,
      "loss": 1.2448,
      "num_input_tokens_seen": 6814912,
      "step": 423
    },
    {
      "epoch": 0.029700456189200312,
      "grad_norm": 4.509056568145752,
      "learning_rate": 9.703274255691769e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 6830720,
      "step": 424
    },
    {
      "epoch": 0.029770504434929557,
      "grad_norm": 4.064620494842529,
      "learning_rate": 9.702574430823118e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 6847104,
      "step": 425
    },
    {
      "epoch": 0.029840552680658803,
      "grad_norm": 3.9060182571411133,
      "learning_rate": 9.701874605954465e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 6862952,
      "step": 426
    },
    {
      "epoch": 0.02991060092638805,
      "grad_norm": 3.9900951385498047,
      "learning_rate": 9.701174781085814e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 6879336,
      "step": 427
    },
    {
      "epoch": 0.029980649172117294,
      "grad_norm": 4.371436595916748,
      "learning_rate": 9.700474956217164e-05,
      "loss": 1.2731,
      "num_input_tokens_seen": 6895720,
      "step": 428
    },
    {
      "epoch": 0.030050697417846543,
      "grad_norm": 3.9422085285186768,
      "learning_rate": 9.699775131348512e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 6912104,
      "step": 429
    },
    {
      "epoch": 0.03012074566357579,
      "grad_norm": 4.080913543701172,
      "learning_rate": 9.699075306479861e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 6928488,
      "step": 430
    },
    {
      "epoch": 0.030190793909305035,
      "grad_norm": 4.493942737579346,
      "learning_rate": 9.698375481611208e-05,
      "loss": 1.2274,
      "num_input_tokens_seen": 6944664,
      "step": 431
    },
    {
      "epoch": 0.03026084215503428,
      "grad_norm": 4.073723793029785,
      "learning_rate": 9.697675656742557e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 6960344,
      "step": 432
    },
    {
      "epoch": 0.030330890400763526,
      "grad_norm": 3.9672274589538574,
      "learning_rate": 9.696975831873906e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 6976720,
      "step": 433
    },
    {
      "epoch": 0.03040093864649277,
      "grad_norm": 4.497872829437256,
      "learning_rate": 9.696276007005255e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 6992552,
      "step": 434
    },
    {
      "epoch": 0.030470986892222017,
      "grad_norm": 4.422168731689453,
      "learning_rate": 9.695576182136604e-05,
      "loss": 1.34,
      "num_input_tokens_seen": 7008936,
      "step": 435
    },
    {
      "epoch": 0.030541035137951263,
      "grad_norm": 4.3009138107299805,
      "learning_rate": 9.694876357267951e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 7024512,
      "step": 436
    },
    {
      "epoch": 0.030611083383680508,
      "grad_norm": 4.04030704498291,
      "learning_rate": 9.6941765323993e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 7040896,
      "step": 437
    },
    {
      "epoch": 0.030681131629409757,
      "grad_norm": 3.877417802810669,
      "learning_rate": 9.693476707530649e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 7057280,
      "step": 438
    },
    {
      "epoch": 0.030751179875139003,
      "grad_norm": 3.8185505867004395,
      "learning_rate": 9.692776882661996e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 7072544,
      "step": 439
    },
    {
      "epoch": 0.03082122812086825,
      "grad_norm": 4.028950214385986,
      "learning_rate": 9.692077057793345e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 7088928,
      "step": 440
    },
    {
      "epoch": 0.030891276366597494,
      "grad_norm": 4.2786431312561035,
      "learning_rate": 9.691377232924694e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 7105248,
      "step": 441
    },
    {
      "epoch": 0.03096132461232674,
      "grad_norm": 4.193462371826172,
      "learning_rate": 9.690677408056043e-05,
      "loss": 1.1812,
      "num_input_tokens_seen": 7121008,
      "step": 442
    },
    {
      "epoch": 0.031031372858055985,
      "grad_norm": 3.93597412109375,
      "learning_rate": 9.68997758318739e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 7136944,
      "step": 443
    },
    {
      "epoch": 0.03110142110378523,
      "grad_norm": 4.3208537101745605,
      "learning_rate": 9.68927775831874e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 7152928,
      "step": 444
    },
    {
      "epoch": 0.031171469349514477,
      "grad_norm": 3.9743378162384033,
      "learning_rate": 9.688577933450088e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 7169312,
      "step": 445
    },
    {
      "epoch": 0.031241517595243726,
      "grad_norm": 4.226114273071289,
      "learning_rate": 9.687878108581436e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 7185696,
      "step": 446
    },
    {
      "epoch": 0.03131156584097297,
      "grad_norm": 4.210222244262695,
      "learning_rate": 9.687178283712785e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 7201784,
      "step": 447
    },
    {
      "epoch": 0.03138161408670222,
      "grad_norm": 4.311635971069336,
      "learning_rate": 9.686478458844133e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 7218168,
      "step": 448
    },
    {
      "epoch": 0.03145166233243146,
      "grad_norm": 4.56603479385376,
      "learning_rate": 9.685778633975482e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 7233360,
      "step": 449
    },
    {
      "epoch": 0.03152171057816071,
      "grad_norm": 4.232856750488281,
      "learning_rate": 9.685078809106831e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 7248280,
      "step": 450
    },
    {
      "epoch": 0.031591758823889954,
      "grad_norm": 4.512947082519531,
      "learning_rate": 9.684378984238179e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 7264664,
      "step": 451
    },
    {
      "epoch": 0.0316618070696192,
      "grad_norm": 4.273897171020508,
      "learning_rate": 9.683679159369528e-05,
      "loss": 1.2523,
      "num_input_tokens_seen": 7281048,
      "step": 452
    },
    {
      "epoch": 0.031731855315348445,
      "grad_norm": 4.288438320159912,
      "learning_rate": 9.682979334500875e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 7297424,
      "step": 453
    },
    {
      "epoch": 0.03180190356107769,
      "grad_norm": 4.27367639541626,
      "learning_rate": 9.682279509632225e-05,
      "loss": 1.1868,
      "num_input_tokens_seen": 7312792,
      "step": 454
    },
    {
      "epoch": 0.031871951806806936,
      "grad_norm": 3.978926181793213,
      "learning_rate": 9.681579684763574e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 7329176,
      "step": 455
    },
    {
      "epoch": 0.03194200005253618,
      "grad_norm": 4.4399919509887695,
      "learning_rate": 9.680879859894922e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 7345560,
      "step": 456
    },
    {
      "epoch": 0.03201204829826543,
      "grad_norm": 3.9786529541015625,
      "learning_rate": 9.68018003502627e-05,
      "loss": 1.1704,
      "num_input_tokens_seen": 7361944,
      "step": 457
    },
    {
      "epoch": 0.03208209654399467,
      "grad_norm": 4.171195030212402,
      "learning_rate": 9.679480210157618e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 7378328,
      "step": 458
    },
    {
      "epoch": 0.032152144789723926,
      "grad_norm": 3.9415268898010254,
      "learning_rate": 9.678780385288967e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 7394208,
      "step": 459
    },
    {
      "epoch": 0.03222219303545317,
      "grad_norm": 4.066036224365234,
      "learning_rate": 9.678080560420316e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 7410328,
      "step": 460
    },
    {
      "epoch": 0.03229224128118242,
      "grad_norm": 4.22513484954834,
      "learning_rate": 9.677380735551665e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 7426712,
      "step": 461
    },
    {
      "epoch": 0.03236228952691166,
      "grad_norm": 4.310954570770264,
      "learning_rate": 9.676680910683013e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 7442736,
      "step": 462
    },
    {
      "epoch": 0.03243233777264091,
      "grad_norm": 4.2868828773498535,
      "learning_rate": 9.675981085814361e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 7458560,
      "step": 463
    },
    {
      "epoch": 0.032502386018370154,
      "grad_norm": 4.318186283111572,
      "learning_rate": 9.67528126094571e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 7474944,
      "step": 464
    },
    {
      "epoch": 0.0325724342640994,
      "grad_norm": 4.040421009063721,
      "learning_rate": 9.674581436077059e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 7490344,
      "step": 465
    },
    {
      "epoch": 0.032642482509828645,
      "grad_norm": 3.914815902709961,
      "learning_rate": 9.673881611208406e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 7506728,
      "step": 466
    },
    {
      "epoch": 0.03271253075555789,
      "grad_norm": 4.054527282714844,
      "learning_rate": 9.673181786339755e-05,
      "loss": 1.2264,
      "num_input_tokens_seen": 7522912,
      "step": 467
    },
    {
      "epoch": 0.032782579001287136,
      "grad_norm": 4.295147895812988,
      "learning_rate": 9.672481961471104e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 7539040,
      "step": 468
    },
    {
      "epoch": 0.03285262724701638,
      "grad_norm": 4.109183311462402,
      "learning_rate": 9.671782136602453e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 7555424,
      "step": 469
    },
    {
      "epoch": 0.03292267549274563,
      "grad_norm": 4.131369590759277,
      "learning_rate": 9.6710823117338e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 7571808,
      "step": 470
    },
    {
      "epoch": 0.03299272373847487,
      "grad_norm": 3.998414993286133,
      "learning_rate": 9.670382486865149e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 7587528,
      "step": 471
    },
    {
      "epoch": 0.03306277198420412,
      "grad_norm": 4.1235551834106445,
      "learning_rate": 9.669682661996498e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 7603912,
      "step": 472
    },
    {
      "epoch": 0.033132820229933364,
      "grad_norm": 4.800798416137695,
      "learning_rate": 9.668982837127845e-05,
      "loss": 1.3634,
      "num_input_tokens_seen": 7617512,
      "step": 473
    },
    {
      "epoch": 0.03320286847566261,
      "grad_norm": 4.068000316619873,
      "learning_rate": 9.668283012259196e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 7633040,
      "step": 474
    },
    {
      "epoch": 0.033272916721391856,
      "grad_norm": 4.0715484619140625,
      "learning_rate": 9.667583187390543e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 7648416,
      "step": 475
    },
    {
      "epoch": 0.0333429649671211,
      "grad_norm": 3.937807321548462,
      "learning_rate": 9.666883362521892e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 7664624,
      "step": 476
    },
    {
      "epoch": 0.033413013212850354,
      "grad_norm": 4.195656776428223,
      "learning_rate": 9.666183537653241e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 7680480,
      "step": 477
    },
    {
      "epoch": 0.0334830614585796,
      "grad_norm": 4.227575778961182,
      "learning_rate": 9.665483712784588e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 7696632,
      "step": 478
    },
    {
      "epoch": 0.033553109704308845,
      "grad_norm": 4.328822135925293,
      "learning_rate": 9.664783887915937e-05,
      "loss": 1.221,
      "num_input_tokens_seen": 7713016,
      "step": 479
    },
    {
      "epoch": 0.03362315795003809,
      "grad_norm": 4.086736679077148,
      "learning_rate": 9.664084063047286e-05,
      "loss": 1.2817,
      "num_input_tokens_seen": 7729400,
      "step": 480
    },
    {
      "epoch": 0.033693206195767336,
      "grad_norm": 4.555233955383301,
      "learning_rate": 9.663384238178635e-05,
      "loss": 1.483,
      "num_input_tokens_seen": 7745784,
      "step": 481
    },
    {
      "epoch": 0.03376325444149658,
      "grad_norm": 4.118983745574951,
      "learning_rate": 9.662684413309984e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 7762168,
      "step": 482
    },
    {
      "epoch": 0.03383330268722583,
      "grad_norm": 4.232059001922607,
      "learning_rate": 9.661984588441331e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 7777920,
      "step": 483
    },
    {
      "epoch": 0.03390335093295507,
      "grad_norm": 6.288865089416504,
      "learning_rate": 9.66128476357268e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 7794304,
      "step": 484
    },
    {
      "epoch": 0.03397339917868432,
      "grad_norm": 4.133046627044678,
      "learning_rate": 9.660584938704028e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 7810200,
      "step": 485
    },
    {
      "epoch": 0.034043447424413564,
      "grad_norm": 4.147965431213379,
      "learning_rate": 9.659885113835377e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 7826384,
      "step": 486
    },
    {
      "epoch": 0.03411349567014281,
      "grad_norm": 4.1191020011901855,
      "learning_rate": 9.659185288966725e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 7841704,
      "step": 487
    },
    {
      "epoch": 0.034183543915872056,
      "grad_norm": 4.518441677093506,
      "learning_rate": 9.658485464098074e-05,
      "loss": 1.263,
      "num_input_tokens_seen": 7858088,
      "step": 488
    },
    {
      "epoch": 0.0342535921616013,
      "grad_norm": 4.321181297302246,
      "learning_rate": 9.657785639229423e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 7874472,
      "step": 489
    },
    {
      "epoch": 0.03432364040733055,
      "grad_norm": 4.366185665130615,
      "learning_rate": 9.65708581436077e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 7890856,
      "step": 490
    },
    {
      "epoch": 0.03439368865305979,
      "grad_norm": 4.042731761932373,
      "learning_rate": 9.65638598949212e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 7906776,
      "step": 491
    },
    {
      "epoch": 0.03446373689878904,
      "grad_norm": 3.743668556213379,
      "learning_rate": 9.655686164623468e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 7923160,
      "step": 492
    },
    {
      "epoch": 0.034533785144518284,
      "grad_norm": 3.8547139167785645,
      "learning_rate": 9.654986339754816e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 7939296,
      "step": 493
    },
    {
      "epoch": 0.034603833390247536,
      "grad_norm": 4.238414287567139,
      "learning_rate": 9.654286514886166e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 7955504,
      "step": 494
    },
    {
      "epoch": 0.03467388163597678,
      "grad_norm": 4.134857177734375,
      "learning_rate": 9.653586690017514e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 7971888,
      "step": 495
    },
    {
      "epoch": 0.03474392988170603,
      "grad_norm": 4.2501983642578125,
      "learning_rate": 9.652886865148862e-05,
      "loss": 1.1829,
      "num_input_tokens_seen": 7988272,
      "step": 496
    },
    {
      "epoch": 0.03481397812743527,
      "grad_norm": 7.4397053718566895,
      "learning_rate": 9.65218704028021e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 8003744,
      "step": 497
    },
    {
      "epoch": 0.03488402637316452,
      "grad_norm": 4.2750959396362305,
      "learning_rate": 9.651487215411559e-05,
      "loss": 1.2387,
      "num_input_tokens_seen": 8019184,
      "step": 498
    },
    {
      "epoch": 0.034954074618893764,
      "grad_norm": 4.156162261962891,
      "learning_rate": 9.650787390542908e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 8035176,
      "step": 499
    },
    {
      "epoch": 0.03502412286462301,
      "grad_norm": 4.178225040435791,
      "learning_rate": 9.650087565674257e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 8051560,
      "step": 500
    },
    {
      "epoch": 0.035094171110352256,
      "grad_norm": 4.147096157073975,
      "learning_rate": 9.649387740805605e-05,
      "loss": 1.2465,
      "num_input_tokens_seen": 8067944,
      "step": 501
    },
    {
      "epoch": 0.0351642193560815,
      "grad_norm": 4.329249858856201,
      "learning_rate": 9.648687915936953e-05,
      "loss": 1.2742,
      "num_input_tokens_seen": 8083824,
      "step": 502
    },
    {
      "epoch": 0.03523426760181075,
      "grad_norm": 4.404232978820801,
      "learning_rate": 9.647988091068302e-05,
      "loss": 1.1511,
      "num_input_tokens_seen": 8100208,
      "step": 503
    },
    {
      "epoch": 0.03530431584753999,
      "grad_norm": 4.190586090087891,
      "learning_rate": 9.64728826619965e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 8116048,
      "step": 504
    },
    {
      "epoch": 0.03537436409326924,
      "grad_norm": 4.262845516204834,
      "learning_rate": 9.646588441330998e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 8132432,
      "step": 505
    },
    {
      "epoch": 0.035444412338998484,
      "grad_norm": 4.452746391296387,
      "learning_rate": 9.645888616462347e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 8148816,
      "step": 506
    },
    {
      "epoch": 0.03551446058472773,
      "grad_norm": 4.111443042755127,
      "learning_rate": 9.645188791593696e-05,
      "loss": 1.0049,
      "num_input_tokens_seen": 8164856,
      "step": 507
    },
    {
      "epoch": 0.035584508830456975,
      "grad_norm": 4.292227268218994,
      "learning_rate": 9.644488966725045e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 8181240,
      "step": 508
    },
    {
      "epoch": 0.03565455707618622,
      "grad_norm": 4.295238971710205,
      "learning_rate": 9.643789141856394e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 8197624,
      "step": 509
    },
    {
      "epoch": 0.035724605321915466,
      "grad_norm": 3.930659294128418,
      "learning_rate": 9.643089316987741e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 8213816,
      "step": 510
    },
    {
      "epoch": 0.03579465356764472,
      "grad_norm": 4.092316150665283,
      "learning_rate": 9.64238949211909e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 8229632,
      "step": 511
    },
    {
      "epoch": 0.035864701813373964,
      "grad_norm": 4.2939252853393555,
      "learning_rate": 9.641689667250437e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 8245232,
      "step": 512
    },
    {
      "epoch": 0.03593475005910321,
      "grad_norm": 4.191503524780273,
      "learning_rate": 9.640989842381786e-05,
      "loss": 0.9399,
      "num_input_tokens_seen": 8260912,
      "step": 513
    },
    {
      "epoch": 0.036004798304832455,
      "grad_norm": 4.141485214233398,
      "learning_rate": 9.640290017513136e-05,
      "loss": 1.1334,
      "num_input_tokens_seen": 8276864,
      "step": 514
    },
    {
      "epoch": 0.0360748465505617,
      "grad_norm": 3.890547752380371,
      "learning_rate": 9.639590192644484e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 8292720,
      "step": 515
    },
    {
      "epoch": 0.03614489479629095,
      "grad_norm": 4.405922889709473,
      "learning_rate": 9.638890367775833e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 8309104,
      "step": 516
    },
    {
      "epoch": 0.03621494304202019,
      "grad_norm": 4.207942485809326,
      "learning_rate": 9.63819054290718e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 8325304,
      "step": 517
    },
    {
      "epoch": 0.03628499128774944,
      "grad_norm": 4.174366474151611,
      "learning_rate": 9.637490718038529e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 8341688,
      "step": 518
    },
    {
      "epoch": 0.036355039533478684,
      "grad_norm": 3.9641714096069336,
      "learning_rate": 9.636790893169878e-05,
      "loss": 1.2244,
      "num_input_tokens_seen": 8357760,
      "step": 519
    },
    {
      "epoch": 0.03642508777920793,
      "grad_norm": 5.832678318023682,
      "learning_rate": 9.636091068301227e-05,
      "loss": 1.0645,
      "num_input_tokens_seen": 8372712,
      "step": 520
    },
    {
      "epoch": 0.036495136024937175,
      "grad_norm": 3.7905161380767822,
      "learning_rate": 9.635391243432576e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 8389096,
      "step": 521
    },
    {
      "epoch": 0.03656518427066642,
      "grad_norm": 3.6744072437286377,
      "learning_rate": 9.634691418563923e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 8405216,
      "step": 522
    },
    {
      "epoch": 0.036635232516395666,
      "grad_norm": 4.897486209869385,
      "learning_rate": 9.633991593695272e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 8421600,
      "step": 523
    },
    {
      "epoch": 0.03670528076212491,
      "grad_norm": 3.821457862854004,
      "learning_rate": 9.63329176882662e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 8437984,
      "step": 524
    },
    {
      "epoch": 0.03677532900785416,
      "grad_norm": 3.873832941055298,
      "learning_rate": 9.632591943957969e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 8453760,
      "step": 525
    },
    {
      "epoch": 0.0368453772535834,
      "grad_norm": 4.139901161193848,
      "learning_rate": 9.631892119089317e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 8470144,
      "step": 526
    },
    {
      "epoch": 0.03691542549931265,
      "grad_norm": 3.9512782096862793,
      "learning_rate": 9.631192294220666e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 8486528,
      "step": 527
    },
    {
      "epoch": 0.0369854737450419,
      "grad_norm": 3.8937103748321533,
      "learning_rate": 9.630492469352015e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 8502912,
      "step": 528
    },
    {
      "epoch": 0.03705552199077115,
      "grad_norm": 4.482640743255615,
      "learning_rate": 9.629792644483363e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 8519296,
      "step": 529
    },
    {
      "epoch": 0.03712557023650039,
      "grad_norm": 4.127941131591797,
      "learning_rate": 9.629092819614711e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 8535160,
      "step": 530
    },
    {
      "epoch": 0.03719561848222964,
      "grad_norm": 3.973585844039917,
      "learning_rate": 9.62839299474606e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 8551256,
      "step": 531
    },
    {
      "epoch": 0.037265666727958884,
      "grad_norm": 4.22855281829834,
      "learning_rate": 9.627693169877408e-05,
      "loss": 1.134,
      "num_input_tokens_seen": 8567640,
      "step": 532
    },
    {
      "epoch": 0.03733571497368813,
      "grad_norm": 4.144021511077881,
      "learning_rate": 9.626993345008757e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 8583504,
      "step": 533
    },
    {
      "epoch": 0.037405763219417375,
      "grad_norm": 3.8666226863861084,
      "learning_rate": 9.626293520140106e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 8599888,
      "step": 534
    },
    {
      "epoch": 0.03747581146514662,
      "grad_norm": 4.215412616729736,
      "learning_rate": 9.625593695271454e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 8616256,
      "step": 535
    },
    {
      "epoch": 0.037545859710875866,
      "grad_norm": 4.353022575378418,
      "learning_rate": 9.624893870402803e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 8632640,
      "step": 536
    },
    {
      "epoch": 0.03761590795660511,
      "grad_norm": 3.778947591781616,
      "learning_rate": 9.624194045534151e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 8648624,
      "step": 537
    },
    {
      "epoch": 0.03768595620233436,
      "grad_norm": 4.481568336486816,
      "learning_rate": 9.6234942206655e-05,
      "loss": 1.3407,
      "num_input_tokens_seen": 8664200,
      "step": 538
    },
    {
      "epoch": 0.0377560044480636,
      "grad_norm": 4.066302299499512,
      "learning_rate": 9.622794395796847e-05,
      "loss": 0.995,
      "num_input_tokens_seen": 8680584,
      "step": 539
    },
    {
      "epoch": 0.03782605269379285,
      "grad_norm": 4.262768268585205,
      "learning_rate": 9.622094570928197e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 8696968,
      "step": 540
    },
    {
      "epoch": 0.037896100939522094,
      "grad_norm": 3.777597665786743,
      "learning_rate": 9.621394746059546e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 8713352,
      "step": 541
    },
    {
      "epoch": 0.03796614918525134,
      "grad_norm": 3.9732742309570312,
      "learning_rate": 9.620694921190894e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 8729048,
      "step": 542
    },
    {
      "epoch": 0.038036197430980585,
      "grad_norm": 4.543329238891602,
      "learning_rate": 9.619995096322243e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 8745432,
      "step": 543
    },
    {
      "epoch": 0.03810624567670983,
      "grad_norm": 4.903865814208984,
      "learning_rate": 9.61929527145359e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 8760296,
      "step": 544
    },
    {
      "epoch": 0.03817629392243908,
      "grad_norm": 4.197691917419434,
      "learning_rate": 9.618595446584939e-05,
      "loss": 1.1616,
      "num_input_tokens_seen": 8776680,
      "step": 545
    },
    {
      "epoch": 0.03824634216816833,
      "grad_norm": 3.912689208984375,
      "learning_rate": 9.617895621716288e-05,
      "loss": 0.9926,
      "num_input_tokens_seen": 8793064,
      "step": 546
    },
    {
      "epoch": 0.038316390413897575,
      "grad_norm": 4.291840076446533,
      "learning_rate": 9.617195796847637e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 8809448,
      "step": 547
    },
    {
      "epoch": 0.03838643865962682,
      "grad_norm": 3.9053072929382324,
      "learning_rate": 9.616495971978985e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 8825536,
      "step": 548
    },
    {
      "epoch": 0.038456486905356066,
      "grad_norm": 4.860696315765381,
      "learning_rate": 9.615796147110333e-05,
      "loss": 1.3045,
      "num_input_tokens_seen": 8841920,
      "step": 549
    },
    {
      "epoch": 0.03852653515108531,
      "grad_norm": 3.9394373893737793,
      "learning_rate": 9.615096322241682e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 8858304,
      "step": 550
    },
    {
      "epoch": 0.03859658339681456,
      "grad_norm": 3.8160409927368164,
      "learning_rate": 9.61439649737303e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 8874688,
      "step": 551
    },
    {
      "epoch": 0.0386666316425438,
      "grad_norm": 4.3792805671691895,
      "learning_rate": 9.613696672504378e-05,
      "loss": 1.2516,
      "num_input_tokens_seen": 8891072,
      "step": 552
    },
    {
      "epoch": 0.03873667988827305,
      "grad_norm": 4.103452682495117,
      "learning_rate": 9.612996847635727e-05,
      "loss": 0.9737,
      "num_input_tokens_seen": 8907456,
      "step": 553
    },
    {
      "epoch": 0.038806728134002294,
      "grad_norm": 4.117603302001953,
      "learning_rate": 9.612297022767076e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 8923816,
      "step": 554
    },
    {
      "epoch": 0.03887677637973154,
      "grad_norm": 4.272468566894531,
      "learning_rate": 9.611597197898425e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 8939344,
      "step": 555
    },
    {
      "epoch": 0.038946824625460785,
      "grad_norm": 4.323635578155518,
      "learning_rate": 9.610897373029772e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 8954920,
      "step": 556
    },
    {
      "epoch": 0.03901687287119003,
      "grad_norm": 3.783510684967041,
      "learning_rate": 9.610197548161121e-05,
      "loss": 1.0658,
      "num_input_tokens_seen": 8971304,
      "step": 557
    },
    {
      "epoch": 0.039086921116919277,
      "grad_norm": 4.3757548332214355,
      "learning_rate": 9.60949772329247e-05,
      "loss": 1.3186,
      "num_input_tokens_seen": 8987672,
      "step": 558
    },
    {
      "epoch": 0.03915696936264852,
      "grad_norm": 4.048824787139893,
      "learning_rate": 9.608797898423818e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 9003896,
      "step": 559
    },
    {
      "epoch": 0.03922701760837777,
      "grad_norm": 4.06865930557251,
      "learning_rate": 9.608098073555168e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 9020280,
      "step": 560
    },
    {
      "epoch": 0.03929706585410701,
      "grad_norm": 3.966737747192383,
      "learning_rate": 9.607398248686515e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 9036280,
      "step": 561
    },
    {
      "epoch": 0.03936711409983626,
      "grad_norm": 4.466656684875488,
      "learning_rate": 9.606698423817864e-05,
      "loss": 1.2462,
      "num_input_tokens_seen": 9052664,
      "step": 562
    },
    {
      "epoch": 0.03943716234556551,
      "grad_norm": 4.312132358551025,
      "learning_rate": 9.605998598949213e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 9068832,
      "step": 563
    },
    {
      "epoch": 0.03950721059129476,
      "grad_norm": 3.9202895164489746,
      "learning_rate": 9.60529877408056e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 9084680,
      "step": 564
    },
    {
      "epoch": 0.039577258837024,
      "grad_norm": 5.139899730682373,
      "learning_rate": 9.604598949211909e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 9099792,
      "step": 565
    },
    {
      "epoch": 0.03964730708275325,
      "grad_norm": 4.398557186126709,
      "learning_rate": 9.603899124343258e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 9116136,
      "step": 566
    },
    {
      "epoch": 0.039717355328482494,
      "grad_norm": 4.350982666015625,
      "learning_rate": 9.603199299474607e-05,
      "loss": 1.2174,
      "num_input_tokens_seen": 9132520,
      "step": 567
    },
    {
      "epoch": 0.03978740357421174,
      "grad_norm": 3.787644386291504,
      "learning_rate": 9.602499474605956e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 9148856,
      "step": 568
    },
    {
      "epoch": 0.039857451819940985,
      "grad_norm": 4.630245685577393,
      "learning_rate": 9.601799649737303e-05,
      "loss": 1.4135,
      "num_input_tokens_seen": 9164888,
      "step": 569
    },
    {
      "epoch": 0.03992750006567023,
      "grad_norm": 4.063969135284424,
      "learning_rate": 9.601099824868652e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 9181272,
      "step": 570
    },
    {
      "epoch": 0.039997548311399476,
      "grad_norm": 4.2443413734436035,
      "learning_rate": 9.6004e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 9197344,
      "step": 571
    },
    {
      "epoch": 0.04006759655712872,
      "grad_norm": 4.396352767944336,
      "learning_rate": 9.599700175131349e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 9212312,
      "step": 572
    },
    {
      "epoch": 0.04013764480285797,
      "grad_norm": 4.364585876464844,
      "learning_rate": 9.599000350262697e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 9228696,
      "step": 573
    },
    {
      "epoch": 0.04020769304858721,
      "grad_norm": 3.9348409175872803,
      "learning_rate": 9.598300525394046e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 9245080,
      "step": 574
    },
    {
      "epoch": 0.04027774129431646,
      "grad_norm": 4.051416873931885,
      "learning_rate": 9.597600700525395e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 9260752,
      "step": 575
    },
    {
      "epoch": 0.040347789540045705,
      "grad_norm": 4.661770820617676,
      "learning_rate": 9.596900875656743e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 9276792,
      "step": 576
    },
    {
      "epoch": 0.04041783778577495,
      "grad_norm": 4.378422260284424,
      "learning_rate": 9.596201050788092e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 9292768,
      "step": 577
    },
    {
      "epoch": 0.040487886031504196,
      "grad_norm": 4.4690399169921875,
      "learning_rate": 9.595501225919439e-05,
      "loss": 1.2398,
      "num_input_tokens_seen": 9309152,
      "step": 578
    },
    {
      "epoch": 0.04055793427723344,
      "grad_norm": 4.1711273193359375,
      "learning_rate": 9.594801401050788e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 9325536,
      "step": 579
    },
    {
      "epoch": 0.040627982522962694,
      "grad_norm": 3.8115949630737305,
      "learning_rate": 9.594101576182137e-05,
      "loss": 1.0317,
      "num_input_tokens_seen": 9341920,
      "step": 580
    },
    {
      "epoch": 0.04069803076869194,
      "grad_norm": 4.072190284729004,
      "learning_rate": 9.593401751313486e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 9357904,
      "step": 581
    },
    {
      "epoch": 0.040768079014421185,
      "grad_norm": 3.895766258239746,
      "learning_rate": 9.592701926444835e-05,
      "loss": 1.1906,
      "num_input_tokens_seen": 9373496,
      "step": 582
    },
    {
      "epoch": 0.04083812726015043,
      "grad_norm": 4.026490688323975,
      "learning_rate": 9.592002101576182e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 9389824,
      "step": 583
    },
    {
      "epoch": 0.040908175505879676,
      "grad_norm": 3.612987518310547,
      "learning_rate": 9.591302276707531e-05,
      "loss": 0.9376,
      "num_input_tokens_seen": 9406208,
      "step": 584
    },
    {
      "epoch": 0.04097822375160892,
      "grad_norm": 4.4619646072387695,
      "learning_rate": 9.59060245183888e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 9422592,
      "step": 585
    },
    {
      "epoch": 0.04104827199733817,
      "grad_norm": 3.990372896194458,
      "learning_rate": 9.589902626970229e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 9438816,
      "step": 586
    },
    {
      "epoch": 0.04111832024306741,
      "grad_norm": 3.7697947025299072,
      "learning_rate": 9.589202802101577e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 9455200,
      "step": 587
    },
    {
      "epoch": 0.04118836848879666,
      "grad_norm": 4.066056728363037,
      "learning_rate": 9.588502977232925e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 9471320,
      "step": 588
    },
    {
      "epoch": 0.041258416734525905,
      "grad_norm": 3.913506507873535,
      "learning_rate": 9.587803152364274e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 9487304,
      "step": 589
    },
    {
      "epoch": 0.04132846498025515,
      "grad_norm": 3.9049429893493652,
      "learning_rate": 9.587103327495623e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 9503688,
      "step": 590
    },
    {
      "epoch": 0.041398513225984396,
      "grad_norm": 4.316978454589844,
      "learning_rate": 9.58640350262697e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 9519488,
      "step": 591
    },
    {
      "epoch": 0.04146856147171364,
      "grad_norm": 3.7818517684936523,
      "learning_rate": 9.585703677758319e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 9535872,
      "step": 592
    },
    {
      "epoch": 0.04153860971744289,
      "grad_norm": 3.8751401901245117,
      "learning_rate": 9.585003852889668e-05,
      "loss": 1.1745,
      "num_input_tokens_seen": 9551928,
      "step": 593
    },
    {
      "epoch": 0.04160865796317213,
      "grad_norm": 4.357265949249268,
      "learning_rate": 9.584304028021017e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 9568312,
      "step": 594
    },
    {
      "epoch": 0.04167870620890138,
      "grad_norm": 4.184159755706787,
      "learning_rate": 9.583604203152366e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 9583968,
      "step": 595
    },
    {
      "epoch": 0.041748754454630624,
      "grad_norm": 3.9540369510650635,
      "learning_rate": 9.582904378283713e-05,
      "loss": 1.2032,
      "num_input_tokens_seen": 9600152,
      "step": 596
    },
    {
      "epoch": 0.04181880270035987,
      "grad_norm": 4.401122093200684,
      "learning_rate": 9.582204553415062e-05,
      "loss": 1.4808,
      "num_input_tokens_seen": 9615632,
      "step": 597
    },
    {
      "epoch": 0.04188885094608912,
      "grad_norm": 4.418131351470947,
      "learning_rate": 9.58150472854641e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 9631712,
      "step": 598
    },
    {
      "epoch": 0.04195889919181837,
      "grad_norm": 4.362226963043213,
      "learning_rate": 9.580804903677758e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 9648096,
      "step": 599
    },
    {
      "epoch": 0.04202894743754761,
      "grad_norm": 4.051177024841309,
      "learning_rate": 9.580105078809107e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 9663792,
      "step": 600
    },
    {
      "epoch": 0.04202894743754761,
      "eval_loss": 1.1809133291244507,
      "eval_runtime": 0.2062,
      "eval_samples_per_second": 4.849,
      "eval_steps_per_second": 4.849,
      "num_input_tokens_seen": 9663792,
      "step": 600
    },
    {
      "epoch": 0.04209899568327686,
      "grad_norm": 4.478739261627197,
      "learning_rate": 9.579405253940456e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 9680176,
      "step": 601
    },
    {
      "epoch": 0.042169043929006104,
      "grad_norm": 4.05004358291626,
      "learning_rate": 9.578705429071805e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 9696560,
      "step": 602
    },
    {
      "epoch": 0.04223909217473535,
      "grad_norm": 4.092396259307861,
      "learning_rate": 9.578005604203152e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 9712944,
      "step": 603
    },
    {
      "epoch": 0.042309140420464596,
      "grad_norm": 4.428014278411865,
      "learning_rate": 9.577305779334501e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 9729096,
      "step": 604
    },
    {
      "epoch": 0.04237918866619384,
      "grad_norm": 4.202315807342529,
      "learning_rate": 9.576605954465849e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 9745480,
      "step": 605
    },
    {
      "epoch": 0.04244923691192309,
      "grad_norm": 3.7633514404296875,
      "learning_rate": 9.575906129597198e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 9761272,
      "step": 606
    },
    {
      "epoch": 0.04251928515765233,
      "grad_norm": 4.170671463012695,
      "learning_rate": 9.575206304728548e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 9777656,
      "step": 607
    },
    {
      "epoch": 0.04258933340338158,
      "grad_norm": 4.20021915435791,
      "learning_rate": 9.574506479859895e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 9794032,
      "step": 608
    },
    {
      "epoch": 0.042659381649110824,
      "grad_norm": 4.437755107879639,
      "learning_rate": 9.573806654991244e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 9809936,
      "step": 609
    },
    {
      "epoch": 0.04272942989484007,
      "grad_norm": 4.417452335357666,
      "learning_rate": 9.573106830122592e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 9825232,
      "step": 610
    },
    {
      "epoch": 0.042799478140569315,
      "grad_norm": 4.144030570983887,
      "learning_rate": 9.57240700525394e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 9840648,
      "step": 611
    },
    {
      "epoch": 0.04286952638629856,
      "grad_norm": 3.991605043411255,
      "learning_rate": 9.57170718038529e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 9857032,
      "step": 612
    },
    {
      "epoch": 0.042939574632027806,
      "grad_norm": 4.516556262969971,
      "learning_rate": 9.571007355516638e-05,
      "loss": 1.3056,
      "num_input_tokens_seen": 9872328,
      "step": 613
    },
    {
      "epoch": 0.04300962287775705,
      "grad_norm": 4.030200481414795,
      "learning_rate": 9.570307530647987e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 9887832,
      "step": 614
    },
    {
      "epoch": 0.043079671123486304,
      "grad_norm": 4.345893859863281,
      "learning_rate": 9.569607705779335e-05,
      "loss": 1.2707,
      "num_input_tokens_seen": 9904216,
      "step": 615
    },
    {
      "epoch": 0.04314971936921555,
      "grad_norm": 4.158145427703857,
      "learning_rate": 9.568907880910684e-05,
      "loss": 1.0377,
      "num_input_tokens_seen": 9920072,
      "step": 616
    },
    {
      "epoch": 0.043219767614944796,
      "grad_norm": 4.155702590942383,
      "learning_rate": 9.568208056042032e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 9936416,
      "step": 617
    },
    {
      "epoch": 0.04328981586067404,
      "grad_norm": 3.76328444480896,
      "learning_rate": 9.56750823117338e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 9952456,
      "step": 618
    },
    {
      "epoch": 0.04335986410640329,
      "grad_norm": 4.252495765686035,
      "learning_rate": 9.566808406304729e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 9968608,
      "step": 619
    },
    {
      "epoch": 0.04342991235213253,
      "grad_norm": 9.254091262817383,
      "learning_rate": 9.566108581436078e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 9983016,
      "step": 620
    },
    {
      "epoch": 0.04349996059786178,
      "grad_norm": 4.028343200683594,
      "learning_rate": 9.565408756567426e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 9999400,
      "step": 621
    },
    {
      "epoch": 0.043570008843591024,
      "grad_norm": 4.051328659057617,
      "learning_rate": 9.564708931698775e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 10015384,
      "step": 622
    },
    {
      "epoch": 0.04364005708932027,
      "grad_norm": 4.495016098022461,
      "learning_rate": 9.564009106830123e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 10031152,
      "step": 623
    },
    {
      "epoch": 0.043710105335049515,
      "grad_norm": 4.876840114593506,
      "learning_rate": 9.563309281961472e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 10047536,
      "step": 624
    },
    {
      "epoch": 0.04378015358077876,
      "grad_norm": 4.407329559326172,
      "learning_rate": 9.562609457092819e-05,
      "loss": 1.2381,
      "num_input_tokens_seen": 10063920,
      "step": 625
    },
    {
      "epoch": 0.043850201826508006,
      "grad_norm": 4.161394119262695,
      "learning_rate": 9.561909632224168e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 10079024,
      "step": 626
    },
    {
      "epoch": 0.04392025007223725,
      "grad_norm": 4.382974624633789,
      "learning_rate": 9.561209807355518e-05,
      "loss": 1.3156,
      "num_input_tokens_seen": 10095408,
      "step": 627
    },
    {
      "epoch": 0.0439902983179665,
      "grad_norm": 4.004157543182373,
      "learning_rate": 9.560509982486866e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 10111792,
      "step": 628
    },
    {
      "epoch": 0.04406034656369574,
      "grad_norm": 3.9019265174865723,
      "learning_rate": 9.559810157618215e-05,
      "loss": 1.0948,
      "num_input_tokens_seen": 10128144,
      "step": 629
    },
    {
      "epoch": 0.04413039480942499,
      "grad_norm": 4.410470485687256,
      "learning_rate": 9.559110332749562e-05,
      "loss": 1.3219,
      "num_input_tokens_seen": 10144288,
      "step": 630
    },
    {
      "epoch": 0.044200443055154234,
      "grad_norm": 4.233544826507568,
      "learning_rate": 9.558410507880911e-05,
      "loss": 0.999,
      "num_input_tokens_seen": 10160296,
      "step": 631
    },
    {
      "epoch": 0.04427049130088349,
      "grad_norm": 4.120091438293457,
      "learning_rate": 9.557710683012258e-05,
      "loss": 1.0166,
      "num_input_tokens_seen": 10176680,
      "step": 632
    },
    {
      "epoch": 0.04434053954661273,
      "grad_norm": 5.061972618103027,
      "learning_rate": 9.557010858143609e-05,
      "loss": 1.251,
      "num_input_tokens_seen": 10192088,
      "step": 633
    },
    {
      "epoch": 0.04441058779234198,
      "grad_norm": 4.3690948486328125,
      "learning_rate": 9.556311033274958e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 10208472,
      "step": 634
    },
    {
      "epoch": 0.044480636038071224,
      "grad_norm": 3.798710346221924,
      "learning_rate": 9.555611208406305e-05,
      "loss": 1.0286,
      "num_input_tokens_seen": 10224856,
      "step": 635
    },
    {
      "epoch": 0.04455068428380047,
      "grad_norm": 4.41818380355835,
      "learning_rate": 9.554911383537654e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 10241200,
      "step": 636
    },
    {
      "epoch": 0.044620732529529715,
      "grad_norm": 4.256262302398682,
      "learning_rate": 9.554211558669001e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 10257584,
      "step": 637
    },
    {
      "epoch": 0.04469078077525896,
      "grad_norm": 4.176064968109131,
      "learning_rate": 9.55351173380035e-05,
      "loss": 1.1985,
      "num_input_tokens_seen": 10273760,
      "step": 638
    },
    {
      "epoch": 0.044760829020988206,
      "grad_norm": 3.9971530437469482,
      "learning_rate": 9.552811908931699e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 10290144,
      "step": 639
    },
    {
      "epoch": 0.04483087726671745,
      "grad_norm": 4.150514602661133,
      "learning_rate": 9.552112084063048e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 10306528,
      "step": 640
    },
    {
      "epoch": 0.0449009255124467,
      "grad_norm": 4.1868367195129395,
      "learning_rate": 9.551412259194397e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 10322480,
      "step": 641
    },
    {
      "epoch": 0.04497097375817594,
      "grad_norm": 4.409821510314941,
      "learning_rate": 9.550712434325744e-05,
      "loss": 1.2574,
      "num_input_tokens_seen": 10338864,
      "step": 642
    },
    {
      "epoch": 0.04504102200390519,
      "grad_norm": 4.500023365020752,
      "learning_rate": 9.550012609457093e-05,
      "loss": 1.35,
      "num_input_tokens_seen": 10355072,
      "step": 643
    },
    {
      "epoch": 0.045111070249634434,
      "grad_norm": 10.278129577636719,
      "learning_rate": 9.549312784588442e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 10371456,
      "step": 644
    },
    {
      "epoch": 0.04518111849536368,
      "grad_norm": 3.9800543785095215,
      "learning_rate": 9.54861295971979e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 10387720,
      "step": 645
    },
    {
      "epoch": 0.045251166741092926,
      "grad_norm": 3.855720281600952,
      "learning_rate": 9.547913134851138e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 10403936,
      "step": 646
    },
    {
      "epoch": 0.04532121498682217,
      "grad_norm": 4.719264984130859,
      "learning_rate": 9.547213309982487e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 10420320,
      "step": 647
    },
    {
      "epoch": 0.04539126323255142,
      "grad_norm": 4.6528167724609375,
      "learning_rate": 9.546513485113836e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 10436704,
      "step": 648
    },
    {
      "epoch": 0.04546131147828066,
      "grad_norm": 4.0597028732299805,
      "learning_rate": 9.545813660245185e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 10452592,
      "step": 649
    },
    {
      "epoch": 0.045531359724009915,
      "grad_norm": 4.161896705627441,
      "learning_rate": 9.545113835376533e-05,
      "loss": 1.1373,
      "num_input_tokens_seen": 10468976,
      "step": 650
    },
    {
      "epoch": 0.04560140796973916,
      "grad_norm": 4.125041961669922,
      "learning_rate": 9.544414010507881e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 10484584,
      "step": 651
    },
    {
      "epoch": 0.045671456215468406,
      "grad_norm": 4.278462886810303,
      "learning_rate": 9.543714185639229e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 10500504,
      "step": 652
    },
    {
      "epoch": 0.04574150446119765,
      "grad_norm": 4.766538619995117,
      "learning_rate": 9.543014360770579e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 10516472,
      "step": 653
    },
    {
      "epoch": 0.0458115527069269,
      "grad_norm": 4.457921504974365,
      "learning_rate": 9.542314535901928e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 10532272,
      "step": 654
    },
    {
      "epoch": 0.04588160095265614,
      "grad_norm": 5.021823883056641,
      "learning_rate": 9.541614711033275e-05,
      "loss": 1.1152,
      "num_input_tokens_seen": 10547696,
      "step": 655
    },
    {
      "epoch": 0.04595164919838539,
      "grad_norm": 4.407228469848633,
      "learning_rate": 9.540914886164624e-05,
      "loss": 1.0863,
      "num_input_tokens_seen": 10564080,
      "step": 656
    },
    {
      "epoch": 0.046021697444114634,
      "grad_norm": 3.9986062049865723,
      "learning_rate": 9.540215061295972e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 10580464,
      "step": 657
    },
    {
      "epoch": 0.04609174568984388,
      "grad_norm": 7.9165191650390625,
      "learning_rate": 9.539515236427321e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 10595336,
      "step": 658
    },
    {
      "epoch": 0.046161793935573125,
      "grad_norm": 4.357856273651123,
      "learning_rate": 9.53881541155867e-05,
      "loss": 1.0324,
      "num_input_tokens_seen": 10611720,
      "step": 659
    },
    {
      "epoch": 0.04623184218130237,
      "grad_norm": 3.8115761280059814,
      "learning_rate": 9.538115586690018e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 10628104,
      "step": 660
    },
    {
      "epoch": 0.04630189042703162,
      "grad_norm": 3.879671096801758,
      "learning_rate": 9.537415761821367e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 10644096,
      "step": 661
    },
    {
      "epoch": 0.04637193867276086,
      "grad_norm": 4.324586391448975,
      "learning_rate": 9.536715936952715e-05,
      "loss": 1.1904,
      "num_input_tokens_seen": 10659408,
      "step": 662
    },
    {
      "epoch": 0.04644198691849011,
      "grad_norm": 4.020029067993164,
      "learning_rate": 9.536016112084064e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 10675792,
      "step": 663
    },
    {
      "epoch": 0.046512035164219354,
      "grad_norm": 4.563455581665039,
      "learning_rate": 9.535316287215411e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 10691632,
      "step": 664
    },
    {
      "epoch": 0.0465820834099486,
      "grad_norm": 4.444424629211426,
      "learning_rate": 9.53461646234676e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 10708016,
      "step": 665
    },
    {
      "epoch": 0.046652131655677845,
      "grad_norm": 3.9864089488983154,
      "learning_rate": 9.533916637478109e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 10724176,
      "step": 666
    },
    {
      "epoch": 0.0467221799014071,
      "grad_norm": 4.860849857330322,
      "learning_rate": 9.533216812609458e-05,
      "loss": 1.2276,
      "num_input_tokens_seen": 10740560,
      "step": 667
    },
    {
      "epoch": 0.04679222814713634,
      "grad_norm": 3.9701120853424072,
      "learning_rate": 9.532516987740807e-05,
      "loss": 1.1406,
      "num_input_tokens_seen": 10756864,
      "step": 668
    },
    {
      "epoch": 0.04686227639286559,
      "grad_norm": 3.660257577896118,
      "learning_rate": 9.531817162872154e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 10773248,
      "step": 669
    },
    {
      "epoch": 0.046932324638594834,
      "grad_norm": 3.888510227203369,
      "learning_rate": 9.531117338003503e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 10789632,
      "step": 670
    },
    {
      "epoch": 0.04700237288432408,
      "grad_norm": 4.794105052947998,
      "learning_rate": 9.530417513134852e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 10804496,
      "step": 671
    },
    {
      "epoch": 0.047072421130053325,
      "grad_norm": 4.293116092681885,
      "learning_rate": 9.529717688266199e-05,
      "loss": 1.2509,
      "num_input_tokens_seen": 10819976,
      "step": 672
    },
    {
      "epoch": 0.04714246937578257,
      "grad_norm": 5.112069129943848,
      "learning_rate": 9.52901786339755e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 10836360,
      "step": 673
    },
    {
      "epoch": 0.04721251762151182,
      "grad_norm": 3.9091360569000244,
      "learning_rate": 9.528318038528897e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 10852744,
      "step": 674
    },
    {
      "epoch": 0.04728256586724106,
      "grad_norm": 4.032161235809326,
      "learning_rate": 9.527618213660246e-05,
      "loss": 1.2362,
      "num_input_tokens_seen": 10868928,
      "step": 675
    },
    {
      "epoch": 0.04735261411297031,
      "grad_norm": 3.931156635284424,
      "learning_rate": 9.526918388791595e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 10884776,
      "step": 676
    },
    {
      "epoch": 0.047422662358699554,
      "grad_norm": 3.9511048793792725,
      "learning_rate": 9.526218563922942e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 10901160,
      "step": 677
    },
    {
      "epoch": 0.0474927106044288,
      "grad_norm": 4.199029445648193,
      "learning_rate": 9.525518739054291e-05,
      "loss": 1.2813,
      "num_input_tokens_seen": 10917544,
      "step": 678
    },
    {
      "epoch": 0.047562758850158045,
      "grad_norm": 3.8590247631073,
      "learning_rate": 9.52481891418564e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 10933928,
      "step": 679
    },
    {
      "epoch": 0.04763280709588729,
      "grad_norm": 5.530341625213623,
      "learning_rate": 9.524119089316989e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 10949600,
      "step": 680
    },
    {
      "epoch": 0.047702855341616536,
      "grad_norm": 4.17647123336792,
      "learning_rate": 9.523419264448338e-05,
      "loss": 1.2985,
      "num_input_tokens_seen": 10965984,
      "step": 681
    },
    {
      "epoch": 0.04777290358734578,
      "grad_norm": 4.250451564788818,
      "learning_rate": 9.522719439579685e-05,
      "loss": 1.1638,
      "num_input_tokens_seen": 10982368,
      "step": 682
    },
    {
      "epoch": 0.04784295183307503,
      "grad_norm": 4.132594108581543,
      "learning_rate": 9.522019614711034e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 10998752,
      "step": 683
    },
    {
      "epoch": 0.04791300007880428,
      "grad_norm": 5.863363265991211,
      "learning_rate": 9.521319789842382e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 11014376,
      "step": 684
    },
    {
      "epoch": 0.047983048324533525,
      "grad_norm": 3.740323543548584,
      "learning_rate": 9.52061996497373e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 11030440,
      "step": 685
    },
    {
      "epoch": 0.04805309657026277,
      "grad_norm": 4.927120685577393,
      "learning_rate": 9.519920140105079e-05,
      "loss": 1.156,
      "num_input_tokens_seen": 11046824,
      "step": 686
    },
    {
      "epoch": 0.04812314481599202,
      "grad_norm": 4.708818435668945,
      "learning_rate": 9.519220315236428e-05,
      "loss": 1.2139,
      "num_input_tokens_seen": 11063208,
      "step": 687
    },
    {
      "epoch": 0.04819319306172126,
      "grad_norm": 3.7547767162323,
      "learning_rate": 9.518520490367777e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 11079592,
      "step": 688
    },
    {
      "epoch": 0.04826324130745051,
      "grad_norm": 4.038534641265869,
      "learning_rate": 9.517820665499124e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 11095976,
      "step": 689
    },
    {
      "epoch": 0.048333289553179754,
      "grad_norm": 4.159554481506348,
      "learning_rate": 9.517120840630473e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 11112360,
      "step": 690
    },
    {
      "epoch": 0.048403337798909,
      "grad_norm": 7.104836463928223,
      "learning_rate": 9.516421015761821e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 11127800,
      "step": 691
    },
    {
      "epoch": 0.048473386044638245,
      "grad_norm": 4.073885917663574,
      "learning_rate": 9.51572119089317e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 11144184,
      "step": 692
    },
    {
      "epoch": 0.04854343429036749,
      "grad_norm": 3.7190351486206055,
      "learning_rate": 9.51502136602452e-05,
      "loss": 1.1035,
      "num_input_tokens_seen": 11160568,
      "step": 693
    },
    {
      "epoch": 0.048613482536096736,
      "grad_norm": 4.252142429351807,
      "learning_rate": 9.514321541155867e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 11176952,
      "step": 694
    },
    {
      "epoch": 0.04868353078182598,
      "grad_norm": 4.418105125427246,
      "learning_rate": 9.513621716287216e-05,
      "loss": 1.2496,
      "num_input_tokens_seen": 11193336,
      "step": 695
    },
    {
      "epoch": 0.04875357902755523,
      "grad_norm": 4.195918560028076,
      "learning_rate": 9.512921891418564e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 11209720,
      "step": 696
    },
    {
      "epoch": 0.04882362727328447,
      "grad_norm": 5.138080596923828,
      "learning_rate": 9.512222066549913e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 11225888,
      "step": 697
    },
    {
      "epoch": 0.04889367551901372,
      "grad_norm": 4.489223003387451,
      "learning_rate": 9.511522241681261e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 11241744,
      "step": 698
    },
    {
      "epoch": 0.048963723764742964,
      "grad_norm": 3.972590208053589,
      "learning_rate": 9.51082241681261e-05,
      "loss": 1.2765,
      "num_input_tokens_seen": 11257768,
      "step": 699
    },
    {
      "epoch": 0.04903377201047221,
      "grad_norm": 13.274886131286621,
      "learning_rate": 9.510122591943959e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 11273216,
      "step": 700
    },
    {
      "epoch": 0.049103820256201455,
      "grad_norm": 3.7899255752563477,
      "learning_rate": 9.509422767075307e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 11289600,
      "step": 701
    },
    {
      "epoch": 0.04917386850193071,
      "grad_norm": 4.226947784423828,
      "learning_rate": 9.508722942206656e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 11305920,
      "step": 702
    },
    {
      "epoch": 0.049243916747659953,
      "grad_norm": 4.098162651062012,
      "learning_rate": 9.508023117338003e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 11322304,
      "step": 703
    },
    {
      "epoch": 0.0493139649933892,
      "grad_norm": 3.9205965995788574,
      "learning_rate": 9.507323292469352e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 11338688,
      "step": 704
    },
    {
      "epoch": 0.049384013239118445,
      "grad_norm": 4.06537389755249,
      "learning_rate": 9.506623467600701e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 11353544,
      "step": 705
    },
    {
      "epoch": 0.04945406148484769,
      "grad_norm": 4.309032440185547,
      "learning_rate": 9.50592364273205e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 11369928,
      "step": 706
    },
    {
      "epoch": 0.049524109730576936,
      "grad_norm": 4.320526599884033,
      "learning_rate": 9.505223817863399e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 11386312,
      "step": 707
    },
    {
      "epoch": 0.04959415797630618,
      "grad_norm": 5.025510787963867,
      "learning_rate": 9.504523992994747e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 11402696,
      "step": 708
    },
    {
      "epoch": 0.04966420622203543,
      "grad_norm": 3.9406464099884033,
      "learning_rate": 9.503824168126095e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 11419080,
      "step": 709
    },
    {
      "epoch": 0.04973425446776467,
      "grad_norm": 3.9148502349853516,
      "learning_rate": 9.503124343257444e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 11435464,
      "step": 710
    },
    {
      "epoch": 0.04980430271349392,
      "grad_norm": 3.9386026859283447,
      "learning_rate": 9.502424518388791e-05,
      "loss": 0.9516,
      "num_input_tokens_seen": 11451848,
      "step": 711
    },
    {
      "epoch": 0.049874350959223164,
      "grad_norm": 3.9537665843963623,
      "learning_rate": 9.50172469352014e-05,
      "loss": 1.1372,
      "num_input_tokens_seen": 11468216,
      "step": 712
    },
    {
      "epoch": 0.04994439920495241,
      "grad_norm": 3.97929310798645,
      "learning_rate": 9.501024868651489e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 11484192,
      "step": 713
    },
    {
      "epoch": 0.050014447450681655,
      "grad_norm": 3.9326419830322266,
      "learning_rate": 9.500325043782838e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 11500576,
      "step": 714
    },
    {
      "epoch": 0.0500844956964109,
      "grad_norm": 3.769347667694092,
      "learning_rate": 9.499625218914187e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 11516960,
      "step": 715
    },
    {
      "epoch": 0.050154543942140146,
      "grad_norm": 4.264547348022461,
      "learning_rate": 9.498925394045534e-05,
      "loss": 1.3166,
      "num_input_tokens_seen": 11532616,
      "step": 716
    },
    {
      "epoch": 0.05022459218786939,
      "grad_norm": 4.885791778564453,
      "learning_rate": 9.498225569176883e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 11548552,
      "step": 717
    },
    {
      "epoch": 0.05029464043359864,
      "grad_norm": 5.4089741706848145,
      "learning_rate": 9.49752574430823e-05,
      "loss": 1.3986,
      "num_input_tokens_seen": 11564936,
      "step": 718
    },
    {
      "epoch": 0.05036468867932789,
      "grad_norm": 4.503393173217773,
      "learning_rate": 9.496825919439581e-05,
      "loss": 0.9947,
      "num_input_tokens_seen": 11580720,
      "step": 719
    },
    {
      "epoch": 0.050434736925057136,
      "grad_norm": 4.364518165588379,
      "learning_rate": 9.49612609457093e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 11597104,
      "step": 720
    },
    {
      "epoch": 0.05050478517078638,
      "grad_norm": 4.229926109313965,
      "learning_rate": 9.495426269702277e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 11612120,
      "step": 721
    },
    {
      "epoch": 0.05057483341651563,
      "grad_norm": 4.477171897888184,
      "learning_rate": 9.494726444833626e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 11627000,
      "step": 722
    },
    {
      "epoch": 0.05064488166224487,
      "grad_norm": 4.071736812591553,
      "learning_rate": 9.494026619964973e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 11643256,
      "step": 723
    },
    {
      "epoch": 0.05071492990797412,
      "grad_norm": 4.219758033752441,
      "learning_rate": 9.493326795096322e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 11659424,
      "step": 724
    },
    {
      "epoch": 0.050784978153703364,
      "grad_norm": 4.108195781707764,
      "learning_rate": 9.492626970227671e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 11675808,
      "step": 725
    },
    {
      "epoch": 0.05085502639943261,
      "grad_norm": 3.964359760284424,
      "learning_rate": 9.49192714535902e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 11691760,
      "step": 726
    },
    {
      "epoch": 0.050925074645161855,
      "grad_norm": 4.585779190063477,
      "learning_rate": 9.491227320490369e-05,
      "loss": 1.1561,
      "num_input_tokens_seen": 11706600,
      "step": 727
    },
    {
      "epoch": 0.0509951228908911,
      "grad_norm": 3.8540141582489014,
      "learning_rate": 9.490527495621716e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 11722984,
      "step": 728
    },
    {
      "epoch": 0.051065171136620346,
      "grad_norm": 4.138955593109131,
      "learning_rate": 9.489827670753065e-05,
      "loss": 1.2842,
      "num_input_tokens_seen": 11738968,
      "step": 729
    },
    {
      "epoch": 0.05113521938234959,
      "grad_norm": 4.138274192810059,
      "learning_rate": 9.489127845884413e-05,
      "loss": 1.1452,
      "num_input_tokens_seen": 11754952,
      "step": 730
    },
    {
      "epoch": 0.05120526762807884,
      "grad_norm": 4.374305248260498,
      "learning_rate": 9.488428021015762e-05,
      "loss": 1.3622,
      "num_input_tokens_seen": 11770832,
      "step": 731
    },
    {
      "epoch": 0.05127531587380808,
      "grad_norm": 4.242674350738525,
      "learning_rate": 9.48772819614711e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 11786872,
      "step": 732
    },
    {
      "epoch": 0.05134536411953733,
      "grad_norm": 4.173389911651611,
      "learning_rate": 9.48702837127846e-05,
      "loss": 1.1853,
      "num_input_tokens_seen": 11803256,
      "step": 733
    },
    {
      "epoch": 0.051415412365266575,
      "grad_norm": 4.014588356018066,
      "learning_rate": 9.486328546409808e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 11819608,
      "step": 734
    },
    {
      "epoch": 0.05148546061099582,
      "grad_norm": 4.759418964385986,
      "learning_rate": 9.485628721541157e-05,
      "loss": 1.1605,
      "num_input_tokens_seen": 11834296,
      "step": 735
    },
    {
      "epoch": 0.05155550885672507,
      "grad_norm": 4.258687973022461,
      "learning_rate": 9.484928896672505e-05,
      "loss": 1.2993,
      "num_input_tokens_seen": 11849728,
      "step": 736
    },
    {
      "epoch": 0.05162555710245432,
      "grad_norm": 4.690395832061768,
      "learning_rate": 9.484229071803853e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 11866112,
      "step": 737
    },
    {
      "epoch": 0.051695605348183564,
      "grad_norm": 4.373327255249023,
      "learning_rate": 9.483529246935201e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 11881960,
      "step": 738
    },
    {
      "epoch": 0.05176565359391281,
      "grad_norm": 4.008789539337158,
      "learning_rate": 9.482829422066551e-05,
      "loss": 1.1174,
      "num_input_tokens_seen": 11897936,
      "step": 739
    },
    {
      "epoch": 0.051835701839642055,
      "grad_norm": 4.391345977783203,
      "learning_rate": 9.482129597197899e-05,
      "loss": 1.2045,
      "num_input_tokens_seen": 11914320,
      "step": 740
    },
    {
      "epoch": 0.0519057500853713,
      "grad_norm": 4.119503021240234,
      "learning_rate": 9.481429772329248e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 11930440,
      "step": 741
    },
    {
      "epoch": 0.051975798331100546,
      "grad_norm": 4.186014175415039,
      "learning_rate": 9.480729947460596e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 11946720,
      "step": 742
    },
    {
      "epoch": 0.05204584657682979,
      "grad_norm": 4.119131088256836,
      "learning_rate": 9.480030122591944e-05,
      "loss": 1.0792,
      "num_input_tokens_seen": 11962360,
      "step": 743
    },
    {
      "epoch": 0.05211589482255904,
      "grad_norm": 3.921030044555664,
      "learning_rate": 9.479330297723293e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 11978744,
      "step": 744
    },
    {
      "epoch": 0.05218594306828828,
      "grad_norm": 3.806251049041748,
      "learning_rate": 9.478630472854642e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 11994912,
      "step": 745
    },
    {
      "epoch": 0.05225599131401753,
      "grad_norm": 4.508687973022461,
      "learning_rate": 9.47793064798599e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 12011296,
      "step": 746
    },
    {
      "epoch": 0.052326039559746775,
      "grad_norm": 4.458346843719482,
      "learning_rate": 9.47723082311734e-05,
      "loss": 1.2878,
      "num_input_tokens_seen": 12027408,
      "step": 747
    },
    {
      "epoch": 0.05239608780547602,
      "grad_norm": 5.779678821563721,
      "learning_rate": 9.476530998248687e-05,
      "loss": 1.2722,
      "num_input_tokens_seen": 12043792,
      "step": 748
    },
    {
      "epoch": 0.052466136051205266,
      "grad_norm": 4.621145725250244,
      "learning_rate": 9.475831173380036e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 12059856,
      "step": 749
    },
    {
      "epoch": 0.05253618429693451,
      "grad_norm": 4.276626110076904,
      "learning_rate": 9.475131348511383e-05,
      "loss": 1.3378,
      "num_input_tokens_seen": 12076240,
      "step": 750
    },
    {
      "epoch": 0.05260623254266376,
      "grad_norm": 4.533468246459961,
      "learning_rate": 9.474431523642732e-05,
      "loss": 0.921,
      "num_input_tokens_seen": 12092416,
      "step": 751
    },
    {
      "epoch": 0.052676280788393,
      "grad_norm": 4.626596927642822,
      "learning_rate": 9.473731698774081e-05,
      "loss": 1.2807,
      "num_input_tokens_seen": 12108664,
      "step": 752
    },
    {
      "epoch": 0.052746329034122255,
      "grad_norm": 4.3372907638549805,
      "learning_rate": 9.47303187390543e-05,
      "loss": 1.2754,
      "num_input_tokens_seen": 12125048,
      "step": 753
    },
    {
      "epoch": 0.0528163772798515,
      "grad_norm": 3.6576266288757324,
      "learning_rate": 9.472332049036779e-05,
      "loss": 0.8487,
      "num_input_tokens_seen": 12141296,
      "step": 754
    },
    {
      "epoch": 0.052886425525580746,
      "grad_norm": 3.8973164558410645,
      "learning_rate": 9.471632224168126e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 12157544,
      "step": 755
    },
    {
      "epoch": 0.05295647377130999,
      "grad_norm": 3.9059019088745117,
      "learning_rate": 9.470932399299475e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 12173928,
      "step": 756
    },
    {
      "epoch": 0.05302652201703924,
      "grad_norm": 4.133029937744141,
      "learning_rate": 9.470232574430822e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 12189864,
      "step": 757
    },
    {
      "epoch": 0.05309657026276848,
      "grad_norm": 3.8380961418151855,
      "learning_rate": 9.469532749562171e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 12206248,
      "step": 758
    },
    {
      "epoch": 0.05316661850849773,
      "grad_norm": 4.753637790679932,
      "learning_rate": 9.468832924693522e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 12222632,
      "step": 759
    },
    {
      "epoch": 0.053236666754226974,
      "grad_norm": 4.704193592071533,
      "learning_rate": 9.468133099824869e-05,
      "loss": 1.2276,
      "num_input_tokens_seen": 12239016,
      "step": 760
    },
    {
      "epoch": 0.05330671499995622,
      "grad_norm": 3.870870351791382,
      "learning_rate": 9.467433274956218e-05,
      "loss": 0.916,
      "num_input_tokens_seen": 12254784,
      "step": 761
    },
    {
      "epoch": 0.053376763245685466,
      "grad_norm": 3.8597328662872314,
      "learning_rate": 9.466733450087567e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 12271160,
      "step": 762
    },
    {
      "epoch": 0.05344681149141471,
      "grad_norm": 3.7109553813934326,
      "learning_rate": 9.466033625218914e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 12286944,
      "step": 763
    },
    {
      "epoch": 0.05351685973714396,
      "grad_norm": 3.985595464706421,
      "learning_rate": 9.465333800350263e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 12303312,
      "step": 764
    },
    {
      "epoch": 0.0535869079828732,
      "grad_norm": 3.797247886657715,
      "learning_rate": 9.464633975481612e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 12319696,
      "step": 765
    },
    {
      "epoch": 0.05365695622860245,
      "grad_norm": 4.88303279876709,
      "learning_rate": 9.463934150612961e-05,
      "loss": 1.2865,
      "num_input_tokens_seen": 12335448,
      "step": 766
    },
    {
      "epoch": 0.053727004474331694,
      "grad_norm": 4.273831367492676,
      "learning_rate": 9.463234325744308e-05,
      "loss": 1.1724,
      "num_input_tokens_seen": 12351720,
      "step": 767
    },
    {
      "epoch": 0.05379705272006094,
      "grad_norm": 3.9505984783172607,
      "learning_rate": 9.462534500875657e-05,
      "loss": 1.1478,
      "num_input_tokens_seen": 12368104,
      "step": 768
    },
    {
      "epoch": 0.053867100965790185,
      "grad_norm": 4.20963191986084,
      "learning_rate": 9.461834676007006e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 12384488,
      "step": 769
    },
    {
      "epoch": 0.05393714921151943,
      "grad_norm": 4.106869220733643,
      "learning_rate": 9.461134851138354e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 12400128,
      "step": 770
    },
    {
      "epoch": 0.05400719745724868,
      "grad_norm": 4.28592586517334,
      "learning_rate": 9.460435026269702e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 12416512,
      "step": 771
    },
    {
      "epoch": 0.05407724570297793,
      "grad_norm": 3.821927070617676,
      "learning_rate": 9.459735201401051e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 12432896,
      "step": 772
    },
    {
      "epoch": 0.054147293948707174,
      "grad_norm": 4.14424467086792,
      "learning_rate": 9.4590353765324e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 12449208,
      "step": 773
    },
    {
      "epoch": 0.05421734219443642,
      "grad_norm": 4.610694885253906,
      "learning_rate": 9.458335551663749e-05,
      "loss": 1.2265,
      "num_input_tokens_seen": 12464128,
      "step": 774
    },
    {
      "epoch": 0.054287390440165666,
      "grad_norm": 4.410182952880859,
      "learning_rate": 9.457635726795097e-05,
      "loss": 1.1904,
      "num_input_tokens_seen": 12479728,
      "step": 775
    },
    {
      "epoch": 0.05435743868589491,
      "grad_norm": 4.096780300140381,
      "learning_rate": 9.456935901926445e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 12495720,
      "step": 776
    },
    {
      "epoch": 0.05442748693162416,
      "grad_norm": 4.028350830078125,
      "learning_rate": 9.456236077057793e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 12511480,
      "step": 777
    },
    {
      "epoch": 0.0544975351773534,
      "grad_norm": 5.264276504516602,
      "learning_rate": 9.455536252189142e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 12527864,
      "step": 778
    },
    {
      "epoch": 0.05456758342308265,
      "grad_norm": 4.371725082397461,
      "learning_rate": 9.454836427320492e-05,
      "loss": 1.1625,
      "num_input_tokens_seen": 12544168,
      "step": 779
    },
    {
      "epoch": 0.054637631668811894,
      "grad_norm": 4.692862510681152,
      "learning_rate": 9.45413660245184e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 12560552,
      "step": 780
    },
    {
      "epoch": 0.05470767991454114,
      "grad_norm": 3.7462823390960693,
      "learning_rate": 9.453436777583188e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 12576936,
      "step": 781
    },
    {
      "epoch": 0.054777728160270385,
      "grad_norm": 4.161571025848389,
      "learning_rate": 9.452736952714536e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 12593040,
      "step": 782
    },
    {
      "epoch": 0.05484777640599963,
      "grad_norm": 3.96793532371521,
      "learning_rate": 9.452037127845885e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 12609424,
      "step": 783
    },
    {
      "epoch": 0.054917824651728876,
      "grad_norm": 4.183755874633789,
      "learning_rate": 9.451337302977232e-05,
      "loss": 1.0868,
      "num_input_tokens_seen": 12625312,
      "step": 784
    },
    {
      "epoch": 0.05498787289745812,
      "grad_norm": 4.506673336029053,
      "learning_rate": 9.450637478108582e-05,
      "loss": 1.1112,
      "num_input_tokens_seen": 12641696,
      "step": 785
    },
    {
      "epoch": 0.05505792114318737,
      "grad_norm": 3.8601651191711426,
      "learning_rate": 9.449937653239931e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 12658080,
      "step": 786
    },
    {
      "epoch": 0.05512796938891661,
      "grad_norm": 5.190856456756592,
      "learning_rate": 9.449237828371279e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 12673032,
      "step": 787
    },
    {
      "epoch": 0.055198017634645866,
      "grad_norm": 4.323099136352539,
      "learning_rate": 9.448538003502628e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 12689064,
      "step": 788
    },
    {
      "epoch": 0.05526806588037511,
      "grad_norm": 4.271193981170654,
      "learning_rate": 9.447838178633976e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 12705448,
      "step": 789
    },
    {
      "epoch": 0.05533811412610436,
      "grad_norm": 3.793525218963623,
      "learning_rate": 9.447138353765324e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 12721832,
      "step": 790
    },
    {
      "epoch": 0.0554081623718336,
      "grad_norm": 3.747575283050537,
      "learning_rate": 9.446438528896673e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 12738216,
      "step": 791
    },
    {
      "epoch": 0.05547821061756285,
      "grad_norm": 4.222849369049072,
      "learning_rate": 9.445738704028022e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 12754600,
      "step": 792
    },
    {
      "epoch": 0.055548258863292094,
      "grad_norm": 9.102783203125,
      "learning_rate": 9.44503887915937e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 12770568,
      "step": 793
    },
    {
      "epoch": 0.05561830710902134,
      "grad_norm": 4.4447808265686035,
      "learning_rate": 9.444339054290718e-05,
      "loss": 1.2908,
      "num_input_tokens_seen": 12785768,
      "step": 794
    },
    {
      "epoch": 0.055688355354750585,
      "grad_norm": 4.038604259490967,
      "learning_rate": 9.443639229422067e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 12801704,
      "step": 795
    },
    {
      "epoch": 0.05575840360047983,
      "grad_norm": 4.492194652557373,
      "learning_rate": 9.442939404553416e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 12818088,
      "step": 796
    },
    {
      "epoch": 0.055828451846209076,
      "grad_norm": 3.978029489517212,
      "learning_rate": 9.442239579684763e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 12834432,
      "step": 797
    },
    {
      "epoch": 0.05589850009193832,
      "grad_norm": 4.014431476593018,
      "learning_rate": 9.441539754816112e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 12850816,
      "step": 798
    },
    {
      "epoch": 0.05596854833766757,
      "grad_norm": 4.0948638916015625,
      "learning_rate": 9.440839929947461e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 12867200,
      "step": 799
    },
    {
      "epoch": 0.05603859658339681,
      "grad_norm": 4.18120813369751,
      "learning_rate": 9.44014010507881e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 12883072,
      "step": 800
    },
    {
      "epoch": 0.05603859658339681,
      "eval_loss": 1.1718552112579346,
      "eval_runtime": 0.2039,
      "eval_samples_per_second": 4.905,
      "eval_steps_per_second": 4.905,
      "num_input_tokens_seen": 12883072,
      "step": 800
    },
    {
      "epoch": 0.05610864482912606,
      "grad_norm": 4.425891399383545,
      "learning_rate": 9.439440280210159e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 12899456,
      "step": 801
    },
    {
      "epoch": 0.056178693074855304,
      "grad_norm": 4.319190979003906,
      "learning_rate": 9.438740455341506e-05,
      "loss": 1.2612,
      "num_input_tokens_seen": 12915840,
      "step": 802
    },
    {
      "epoch": 0.05624874132058455,
      "grad_norm": 4.28010892868042,
      "learning_rate": 9.438040630472855e-05,
      "loss": 1.0853,
      "num_input_tokens_seen": 12932096,
      "step": 803
    },
    {
      "epoch": 0.056318789566313796,
      "grad_norm": 3.9454870223999023,
      "learning_rate": 9.437340805604203e-05,
      "loss": 1.055,
      "num_input_tokens_seen": 12948208,
      "step": 804
    },
    {
      "epoch": 0.05638883781204305,
      "grad_norm": 4.009400367736816,
      "learning_rate": 9.436640980735553e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 12964096,
      "step": 805
    },
    {
      "epoch": 0.056458886057772294,
      "grad_norm": 3.7949161529541016,
      "learning_rate": 9.435941155866902e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 12980480,
      "step": 806
    },
    {
      "epoch": 0.05652893430350154,
      "grad_norm": 3.910456418991089,
      "learning_rate": 9.435241330998249e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 12996864,
      "step": 807
    },
    {
      "epoch": 0.056598982549230785,
      "grad_norm": 4.744706630706787,
      "learning_rate": 9.434541506129598e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 13013248,
      "step": 808
    },
    {
      "epoch": 0.05666903079496003,
      "grad_norm": 4.4282732009887695,
      "learning_rate": 9.433841681260946e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 13029632,
      "step": 809
    },
    {
      "epoch": 0.056739079040689276,
      "grad_norm": 3.8422467708587646,
      "learning_rate": 9.433141856392294e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 13046016,
      "step": 810
    },
    {
      "epoch": 0.05680912728641852,
      "grad_norm": 4.1764445304870605,
      "learning_rate": 9.432442031523643e-05,
      "loss": 1.183,
      "num_input_tokens_seen": 13062400,
      "step": 811
    },
    {
      "epoch": 0.05687917553214777,
      "grad_norm": 4.713895320892334,
      "learning_rate": 9.431742206654992e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 13078584,
      "step": 812
    },
    {
      "epoch": 0.05694922377787701,
      "grad_norm": 4.265610694885254,
      "learning_rate": 9.431042381786341e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 13094968,
      "step": 813
    },
    {
      "epoch": 0.05701927202360626,
      "grad_norm": 3.9274330139160156,
      "learning_rate": 9.430342556917688e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 13111304,
      "step": 814
    },
    {
      "epoch": 0.057089320269335504,
      "grad_norm": 4.44935941696167,
      "learning_rate": 9.429642732049037e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 13127304,
      "step": 815
    },
    {
      "epoch": 0.05715936851506475,
      "grad_norm": 5.019375801086426,
      "learning_rate": 9.428942907180386e-05,
      "loss": 1.0535,
      "num_input_tokens_seen": 13143688,
      "step": 816
    },
    {
      "epoch": 0.057229416760793995,
      "grad_norm": 4.743424892425537,
      "learning_rate": 9.428243082311734e-05,
      "loss": 1.3912,
      "num_input_tokens_seen": 13160072,
      "step": 817
    },
    {
      "epoch": 0.05729946500652324,
      "grad_norm": 3.921475887298584,
      "learning_rate": 9.427543257443083e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 13176456,
      "step": 818
    },
    {
      "epoch": 0.05736951325225249,
      "grad_norm": 4.106019020080566,
      "learning_rate": 9.426843432574431e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 13192840,
      "step": 819
    },
    {
      "epoch": 0.05743956149798173,
      "grad_norm": 4.298704147338867,
      "learning_rate": 9.42614360770578e-05,
      "loss": 1.281,
      "num_input_tokens_seen": 13209144,
      "step": 820
    },
    {
      "epoch": 0.05750960974371098,
      "grad_norm": 4.29774284362793,
      "learning_rate": 9.425443782837128e-05,
      "loss": 1.2703,
      "num_input_tokens_seen": 13224752,
      "step": 821
    },
    {
      "epoch": 0.057579657989440224,
      "grad_norm": 4.6176838874816895,
      "learning_rate": 9.424743957968477e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 13240856,
      "step": 822
    },
    {
      "epoch": 0.057649706235169476,
      "grad_norm": 4.450786590576172,
      "learning_rate": 9.424044133099826e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 13256800,
      "step": 823
    },
    {
      "epoch": 0.05771975448089872,
      "grad_norm": 3.8302414417266846,
      "learning_rate": 9.423344308231173e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 13273032,
      "step": 824
    },
    {
      "epoch": 0.05778980272662797,
      "grad_norm": 4.641941070556641,
      "learning_rate": 9.422644483362523e-05,
      "loss": 1.2238,
      "num_input_tokens_seen": 13289104,
      "step": 825
    },
    {
      "epoch": 0.05785985097235721,
      "grad_norm": 4.369805335998535,
      "learning_rate": 9.421944658493871e-05,
      "loss": 1.2047,
      "num_input_tokens_seen": 13304752,
      "step": 826
    },
    {
      "epoch": 0.05792989921808646,
      "grad_norm": 3.863507032394409,
      "learning_rate": 9.42124483362522e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 13321088,
      "step": 827
    },
    {
      "epoch": 0.057999947463815704,
      "grad_norm": 5.323369979858398,
      "learning_rate": 9.420545008756568e-05,
      "loss": 1.1722,
      "num_input_tokens_seen": 13336912,
      "step": 828
    },
    {
      "epoch": 0.05806999570954495,
      "grad_norm": 4.006597995758057,
      "learning_rate": 9.419845183887916e-05,
      "loss": 1.0382,
      "num_input_tokens_seen": 13353280,
      "step": 829
    },
    {
      "epoch": 0.058140043955274195,
      "grad_norm": 4.1039886474609375,
      "learning_rate": 9.419145359019265e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 13369664,
      "step": 830
    },
    {
      "epoch": 0.05821009220100344,
      "grad_norm": 3.903517007827759,
      "learning_rate": 9.418445534150614e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 13386048,
      "step": 831
    },
    {
      "epoch": 0.05828014044673269,
      "grad_norm": 4.434885025024414,
      "learning_rate": 9.417745709281963e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 13402432,
      "step": 832
    },
    {
      "epoch": 0.05835018869246193,
      "grad_norm": 4.6121296882629395,
      "learning_rate": 9.417045884413311e-05,
      "loss": 1.2831,
      "num_input_tokens_seen": 13418816,
      "step": 833
    },
    {
      "epoch": 0.05842023693819118,
      "grad_norm": 3.6966841220855713,
      "learning_rate": 9.416346059544659e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 13435200,
      "step": 834
    },
    {
      "epoch": 0.058490285183920424,
      "grad_norm": 4.292221546173096,
      "learning_rate": 9.415646234676008e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 13451584,
      "step": 835
    },
    {
      "epoch": 0.05856033342964967,
      "grad_norm": 4.053999900817871,
      "learning_rate": 9.414946409807355e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 13467824,
      "step": 836
    },
    {
      "epoch": 0.058630381675378915,
      "grad_norm": 4.4411234855651855,
      "learning_rate": 9.414246584938704e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 13483200,
      "step": 837
    },
    {
      "epoch": 0.05870042992110816,
      "grad_norm": 3.956787347793579,
      "learning_rate": 9.413546760070053e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 13499584,
      "step": 838
    },
    {
      "epoch": 0.058770478166837406,
      "grad_norm": 5.050291061401367,
      "learning_rate": 9.412846935201402e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 13515448,
      "step": 839
    },
    {
      "epoch": 0.05884052641256666,
      "grad_norm": 3.8736393451690674,
      "learning_rate": 9.412147110332751e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 13531200,
      "step": 840
    },
    {
      "epoch": 0.058910574658295904,
      "grad_norm": 6.07747745513916,
      "learning_rate": 9.411447285464098e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 13547584,
      "step": 841
    },
    {
      "epoch": 0.05898062290402515,
      "grad_norm": 4.606445789337158,
      "learning_rate": 9.410747460595447e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 13563528,
      "step": 842
    },
    {
      "epoch": 0.059050671149754395,
      "grad_norm": 4.3981709480285645,
      "learning_rate": 9.410047635726796e-05,
      "loss": 1.3313,
      "num_input_tokens_seen": 13579912,
      "step": 843
    },
    {
      "epoch": 0.05912071939548364,
      "grad_norm": 3.64546799659729,
      "learning_rate": 9.409347810858143e-05,
      "loss": 0.8892,
      "num_input_tokens_seen": 13596296,
      "step": 844
    },
    {
      "epoch": 0.05919076764121289,
      "grad_norm": 4.15845251083374,
      "learning_rate": 9.408647985989494e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 13612680,
      "step": 845
    },
    {
      "epoch": 0.05926081588694213,
      "grad_norm": 6.049203872680664,
      "learning_rate": 9.407948161120841e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 13627832,
      "step": 846
    },
    {
      "epoch": 0.05933086413267138,
      "grad_norm": 3.7192461490631104,
      "learning_rate": 9.40724833625219e-05,
      "loss": 1.165,
      "num_input_tokens_seen": 13643824,
      "step": 847
    },
    {
      "epoch": 0.059400912378400623,
      "grad_norm": 4.183239936828613,
      "learning_rate": 9.406548511383537e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 13660208,
      "step": 848
    },
    {
      "epoch": 0.05947096062412987,
      "grad_norm": 4.126212120056152,
      "learning_rate": 9.405848686514886e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 13676592,
      "step": 849
    },
    {
      "epoch": 0.059541008869859115,
      "grad_norm": 4.033525466918945,
      "learning_rate": 9.405148861646235e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 13692600,
      "step": 850
    },
    {
      "epoch": 0.05961105711558836,
      "grad_norm": 4.162797451019287,
      "learning_rate": 9.404449036777584e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 13708984,
      "step": 851
    },
    {
      "epoch": 0.059681105361317606,
      "grad_norm": 4.057224750518799,
      "learning_rate": 9.403749211908933e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 13724656,
      "step": 852
    },
    {
      "epoch": 0.05975115360704685,
      "grad_norm": 4.201955318450928,
      "learning_rate": 9.40304938704028e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 13741040,
      "step": 853
    },
    {
      "epoch": 0.0598212018527761,
      "grad_norm": 3.8704352378845215,
      "learning_rate": 9.402349562171629e-05,
      "loss": 0.8946,
      "num_input_tokens_seen": 13757424,
      "step": 854
    },
    {
      "epoch": 0.05989125009850534,
      "grad_norm": 6.010958671569824,
      "learning_rate": 9.401649737302978e-05,
      "loss": 1.2095,
      "num_input_tokens_seen": 13773808,
      "step": 855
    },
    {
      "epoch": 0.05996129834423459,
      "grad_norm": 4.975742816925049,
      "learning_rate": 9.400949912434326e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 13789704,
      "step": 856
    },
    {
      "epoch": 0.06003134658996384,
      "grad_norm": 4.021739959716797,
      "learning_rate": 9.400250087565675e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 13806088,
      "step": 857
    },
    {
      "epoch": 0.06010139483569309,
      "grad_norm": 4.262394905090332,
      "learning_rate": 9.399550262697023e-05,
      "loss": 1.1053,
      "num_input_tokens_seen": 13821928,
      "step": 858
    },
    {
      "epoch": 0.06017144308142233,
      "grad_norm": 4.3033671379089355,
      "learning_rate": 9.398850437828372e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 13838232,
      "step": 859
    },
    {
      "epoch": 0.06024149132715158,
      "grad_norm": 4.066610336303711,
      "learning_rate": 9.398150612959721e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 13853912,
      "step": 860
    },
    {
      "epoch": 0.06031153957288082,
      "grad_norm": 4.308155059814453,
      "learning_rate": 9.397450788091069e-05,
      "loss": 1.3624,
      "num_input_tokens_seen": 13870224,
      "step": 861
    },
    {
      "epoch": 0.06038158781861007,
      "grad_norm": 4.307553291320801,
      "learning_rate": 9.396750963222417e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 13886608,
      "step": 862
    },
    {
      "epoch": 0.060451636064339315,
      "grad_norm": 3.8107142448425293,
      "learning_rate": 9.396051138353765e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 13902992,
      "step": 863
    },
    {
      "epoch": 0.06052168431006856,
      "grad_norm": 4.530765533447266,
      "learning_rate": 9.395351313485114e-05,
      "loss": 1.2028,
      "num_input_tokens_seen": 13919376,
      "step": 864
    },
    {
      "epoch": 0.060591732555797806,
      "grad_norm": 4.035069465637207,
      "learning_rate": 9.394651488616463e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 13935664,
      "step": 865
    },
    {
      "epoch": 0.06066178080152705,
      "grad_norm": 4.028316497802734,
      "learning_rate": 9.393951663747812e-05,
      "loss": 1.21,
      "num_input_tokens_seen": 13951096,
      "step": 866
    },
    {
      "epoch": 0.0607318290472563,
      "grad_norm": 4.039167881011963,
      "learning_rate": 9.39325183887916e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 13966272,
      "step": 867
    },
    {
      "epoch": 0.06080187729298554,
      "grad_norm": 4.139703273773193,
      "learning_rate": 9.392552014010508e-05,
      "loss": 1.2575,
      "num_input_tokens_seen": 13981848,
      "step": 868
    },
    {
      "epoch": 0.06087192553871479,
      "grad_norm": 4.222180366516113,
      "learning_rate": 9.391852189141857e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 13997920,
      "step": 869
    },
    {
      "epoch": 0.060941973784444034,
      "grad_norm": 3.7993030548095703,
      "learning_rate": 9.391152364273206e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 14014304,
      "step": 870
    },
    {
      "epoch": 0.06101202203017328,
      "grad_norm": 4.811493396759033,
      "learning_rate": 9.390452539404554e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 14030688,
      "step": 871
    },
    {
      "epoch": 0.061082070275902525,
      "grad_norm": 13.88792610168457,
      "learning_rate": 9.389752714535903e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 14045584,
      "step": 872
    },
    {
      "epoch": 0.06115211852163177,
      "grad_norm": 3.7678709030151367,
      "learning_rate": 9.389052889667251e-05,
      "loss": 1.1012,
      "num_input_tokens_seen": 14061968,
      "step": 873
    },
    {
      "epoch": 0.061222166767361016,
      "grad_norm": 4.252075672149658,
      "learning_rate": 9.3883530647986e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 14077584,
      "step": 874
    },
    {
      "epoch": 0.06129221501309027,
      "grad_norm": 3.555629253387451,
      "learning_rate": 9.387653239929947e-05,
      "loss": 0.8653,
      "num_input_tokens_seen": 14093704,
      "step": 875
    },
    {
      "epoch": 0.061362263258819515,
      "grad_norm": 4.122331619262695,
      "learning_rate": 9.386953415061296e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 14109624,
      "step": 876
    },
    {
      "epoch": 0.06143231150454876,
      "grad_norm": 3.6772518157958984,
      "learning_rate": 9.386253590192645e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 14126008,
      "step": 877
    },
    {
      "epoch": 0.061502359750278006,
      "grad_norm": 3.791351079940796,
      "learning_rate": 9.385553765323994e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 14142392,
      "step": 878
    },
    {
      "epoch": 0.06157240799600725,
      "grad_norm": 3.781759738922119,
      "learning_rate": 9.384853940455343e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 14158776,
      "step": 879
    },
    {
      "epoch": 0.0616424562417365,
      "grad_norm": 4.2420830726623535,
      "learning_rate": 9.38415411558669e-05,
      "loss": 1.268,
      "num_input_tokens_seen": 14173920,
      "step": 880
    },
    {
      "epoch": 0.06171250448746574,
      "grad_norm": 4.000860214233398,
      "learning_rate": 9.383454290718039e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 14190032,
      "step": 881
    },
    {
      "epoch": 0.06178255273319499,
      "grad_norm": 3.760969877243042,
      "learning_rate": 9.382754465849388e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 14206416,
      "step": 882
    },
    {
      "epoch": 0.061852600978924234,
      "grad_norm": 4.81919002532959,
      "learning_rate": 9.382054640980735e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 14222408,
      "step": 883
    },
    {
      "epoch": 0.06192264922465348,
      "grad_norm": 4.951950550079346,
      "learning_rate": 9.381354816112084e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 14238616,
      "step": 884
    },
    {
      "epoch": 0.061992697470382725,
      "grad_norm": 4.15132999420166,
      "learning_rate": 9.380654991243433e-05,
      "loss": 1.3171,
      "num_input_tokens_seen": 14254968,
      "step": 885
    },
    {
      "epoch": 0.06206274571611197,
      "grad_norm": 5.100244998931885,
      "learning_rate": 9.379955166374782e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 14271352,
      "step": 886
    },
    {
      "epoch": 0.062132793961841216,
      "grad_norm": 5.999105453491211,
      "learning_rate": 9.379255341506131e-05,
      "loss": 0.9824,
      "num_input_tokens_seen": 14287496,
      "step": 887
    },
    {
      "epoch": 0.06220284220757046,
      "grad_norm": 3.8826348781585693,
      "learning_rate": 9.378555516637478e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 14303880,
      "step": 888
    },
    {
      "epoch": 0.06227289045329971,
      "grad_norm": 5.308819770812988,
      "learning_rate": 9.377855691768827e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 14320264,
      "step": 889
    },
    {
      "epoch": 0.06234293869902895,
      "grad_norm": 4.383331775665283,
      "learning_rate": 9.377155866900175e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 14336232,
      "step": 890
    },
    {
      "epoch": 0.0624129869447582,
      "grad_norm": 4.335045337677002,
      "learning_rate": 9.376456042031524e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 14351704,
      "step": 891
    },
    {
      "epoch": 0.06248303519048745,
      "grad_norm": 3.6901326179504395,
      "learning_rate": 9.375756217162872e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 14368088,
      "step": 892
    },
    {
      "epoch": 0.0625530834362167,
      "grad_norm": 3.912727117538452,
      "learning_rate": 9.375056392294221e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 14383904,
      "step": 893
    },
    {
      "epoch": 0.06262313168194594,
      "grad_norm": 3.5688252449035645,
      "learning_rate": 9.37435656742557e-05,
      "loss": 0.833,
      "num_input_tokens_seen": 14399648,
      "step": 894
    },
    {
      "epoch": 0.06269317992767519,
      "grad_norm": 4.6460137367248535,
      "learning_rate": 9.373656742556918e-05,
      "loss": 1.2523,
      "num_input_tokens_seen": 14415640,
      "step": 895
    },
    {
      "epoch": 0.06276322817340443,
      "grad_norm": 3.8113012313842773,
      "learning_rate": 9.372956917688266e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 14432024,
      "step": 896
    },
    {
      "epoch": 0.06283327641913368,
      "grad_norm": 3.8755953311920166,
      "learning_rate": 9.372257092819615e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 14448152,
      "step": 897
    },
    {
      "epoch": 0.06290332466486293,
      "grad_norm": 4.225901126861572,
      "learning_rate": 9.371557267950964e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 14464536,
      "step": 898
    },
    {
      "epoch": 0.06297337291059217,
      "grad_norm": 3.9437992572784424,
      "learning_rate": 9.370857443082313e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 14480072,
      "step": 899
    },
    {
      "epoch": 0.06304342115632142,
      "grad_norm": 3.8961846828460693,
      "learning_rate": 9.37015761821366e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 14496456,
      "step": 900
    },
    {
      "epoch": 0.06311346940205066,
      "grad_norm": 4.844581604003906,
      "learning_rate": 9.36945779334501e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 14512520,
      "step": 901
    },
    {
      "epoch": 0.06318351764777991,
      "grad_norm": 4.89027214050293,
      "learning_rate": 9.368757968476357e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 14528904,
      "step": 902
    },
    {
      "epoch": 0.06325356589350915,
      "grad_norm": 4.303073883056641,
      "learning_rate": 9.368058143607706e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 14545288,
      "step": 903
    },
    {
      "epoch": 0.0633236141392384,
      "grad_norm": 5.145171165466309,
      "learning_rate": 9.367358318739055e-05,
      "loss": 1.3597,
      "num_input_tokens_seen": 14561672,
      "step": 904
    },
    {
      "epoch": 0.06339366238496764,
      "grad_norm": 5.7905964851379395,
      "learning_rate": 9.366658493870403e-05,
      "loss": 1.1075,
      "num_input_tokens_seen": 14575896,
      "step": 905
    },
    {
      "epoch": 0.06346371063069689,
      "grad_norm": 3.7394728660583496,
      "learning_rate": 9.365958669001752e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 14592280,
      "step": 906
    },
    {
      "epoch": 0.06353375887642614,
      "grad_norm": 3.916626453399658,
      "learning_rate": 9.3652588441331e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 14608072,
      "step": 907
    },
    {
      "epoch": 0.06360380712215538,
      "grad_norm": 5.088227272033691,
      "learning_rate": 9.364559019264449e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 14624360,
      "step": 908
    },
    {
      "epoch": 0.06367385536788463,
      "grad_norm": 3.8519606590270996,
      "learning_rate": 9.363859194395798e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 14640744,
      "step": 909
    },
    {
      "epoch": 0.06374390361361387,
      "grad_norm": 4.450200080871582,
      "learning_rate": 9.363159369527145e-05,
      "loss": 1.0145,
      "num_input_tokens_seen": 14657128,
      "step": 910
    },
    {
      "epoch": 0.06381395185934312,
      "grad_norm": 4.188115119934082,
      "learning_rate": 9.362459544658494e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 14673128,
      "step": 911
    },
    {
      "epoch": 0.06388400010507236,
      "grad_norm": 4.67346715927124,
      "learning_rate": 9.361759719789843e-05,
      "loss": 1.2841,
      "num_input_tokens_seen": 14689512,
      "step": 912
    },
    {
      "epoch": 0.06395404835080161,
      "grad_norm": 3.737790822982788,
      "learning_rate": 9.361059894921192e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 14705872,
      "step": 913
    },
    {
      "epoch": 0.06402409659653086,
      "grad_norm": 4.2486653327941895,
      "learning_rate": 9.36036007005254e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 14721816,
      "step": 914
    },
    {
      "epoch": 0.0640941448422601,
      "grad_norm": 4.120566368103027,
      "learning_rate": 9.359660245183888e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 14738200,
      "step": 915
    },
    {
      "epoch": 0.06416419308798935,
      "grad_norm": 5.259902477264404,
      "learning_rate": 9.358960420315237e-05,
      "loss": 1.3544,
      "num_input_tokens_seen": 14753920,
      "step": 916
    },
    {
      "epoch": 0.06423424133371859,
      "grad_norm": 3.900827646255493,
      "learning_rate": 9.358260595446584e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 14769640,
      "step": 917
    },
    {
      "epoch": 0.06430428957944785,
      "grad_norm": 4.103065490722656,
      "learning_rate": 9.357560770577935e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 14786024,
      "step": 918
    },
    {
      "epoch": 0.0643743378251771,
      "grad_norm": 3.9913623332977295,
      "learning_rate": 9.356860945709282e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 14802408,
      "step": 919
    },
    {
      "epoch": 0.06444438607090634,
      "grad_norm": 3.7369885444641113,
      "learning_rate": 9.356161120840631e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 14818792,
      "step": 920
    },
    {
      "epoch": 0.06451443431663559,
      "grad_norm": 4.029351711273193,
      "learning_rate": 9.35546129597198e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 14833792,
      "step": 921
    },
    {
      "epoch": 0.06458448256236483,
      "grad_norm": 4.043665885925293,
      "learning_rate": 9.354761471103327e-05,
      "loss": 1.179,
      "num_input_tokens_seen": 14850176,
      "step": 922
    },
    {
      "epoch": 0.06465453080809408,
      "grad_norm": 3.7803280353546143,
      "learning_rate": 9.354061646234676e-05,
      "loss": 0.9886,
      "num_input_tokens_seen": 14866096,
      "step": 923
    },
    {
      "epoch": 0.06472457905382333,
      "grad_norm": 5.537375450134277,
      "learning_rate": 9.353361821366025e-05,
      "loss": 1.2519,
      "num_input_tokens_seen": 14882480,
      "step": 924
    },
    {
      "epoch": 0.06479462729955257,
      "grad_norm": 4.944652557373047,
      "learning_rate": 9.352661996497374e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 14898864,
      "step": 925
    },
    {
      "epoch": 0.06486467554528182,
      "grad_norm": 4.3231611251831055,
      "learning_rate": 9.351962171628723e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 14913856,
      "step": 926
    },
    {
      "epoch": 0.06493472379101106,
      "grad_norm": 4.386692523956299,
      "learning_rate": 9.35126234676007e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 14929816,
      "step": 927
    },
    {
      "epoch": 0.06500477203674031,
      "grad_norm": 4.607088088989258,
      "learning_rate": 9.350562521891419e-05,
      "loss": 1.2197,
      "num_input_tokens_seen": 14946200,
      "step": 928
    },
    {
      "epoch": 0.06507482028246955,
      "grad_norm": 4.7108001708984375,
      "learning_rate": 9.349862697022767e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 14961816,
      "step": 929
    },
    {
      "epoch": 0.0651448685281988,
      "grad_norm": 3.844571352005005,
      "learning_rate": 9.349162872154115e-05,
      "loss": 1.2745,
      "num_input_tokens_seen": 14978200,
      "step": 930
    },
    {
      "epoch": 0.06521491677392804,
      "grad_norm": 4.078561782836914,
      "learning_rate": 9.348463047285464e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 14994440,
      "step": 931
    },
    {
      "epoch": 0.06528496501965729,
      "grad_norm": 4.317986011505127,
      "learning_rate": 9.347763222416813e-05,
      "loss": 1.3046,
      "num_input_tokens_seen": 15010824,
      "step": 932
    },
    {
      "epoch": 0.06535501326538654,
      "grad_norm": 4.459141254425049,
      "learning_rate": 9.347063397548162e-05,
      "loss": 1.2893,
      "num_input_tokens_seen": 15026608,
      "step": 933
    },
    {
      "epoch": 0.06542506151111578,
      "grad_norm": 4.251399993896484,
      "learning_rate": 9.34636357267951e-05,
      "loss": 1.2346,
      "num_input_tokens_seen": 15042328,
      "step": 934
    },
    {
      "epoch": 0.06549510975684503,
      "grad_norm": 4.568341255187988,
      "learning_rate": 9.345663747810858e-05,
      "loss": 1.4343,
      "num_input_tokens_seen": 15058712,
      "step": 935
    },
    {
      "epoch": 0.06556515800257427,
      "grad_norm": 4.7616424560546875,
      "learning_rate": 9.344963922942207e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 15075096,
      "step": 936
    },
    {
      "epoch": 0.06563520624830352,
      "grad_norm": 3.8224191665649414,
      "learning_rate": 9.344264098073555e-05,
      "loss": 1.0958,
      "num_input_tokens_seen": 15091480,
      "step": 937
    },
    {
      "epoch": 0.06570525449403276,
      "grad_norm": 4.985624313354492,
      "learning_rate": 9.343564273204905e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 15107864,
      "step": 938
    },
    {
      "epoch": 0.06577530273976201,
      "grad_norm": 4.3780975341796875,
      "learning_rate": 9.342864448336252e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 15123656,
      "step": 939
    },
    {
      "epoch": 0.06584535098549125,
      "grad_norm": 4.435183525085449,
      "learning_rate": 9.342164623467601e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 15140040,
      "step": 940
    },
    {
      "epoch": 0.0659153992312205,
      "grad_norm": 4.560804843902588,
      "learning_rate": 9.34146479859895e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 15156424,
      "step": 941
    },
    {
      "epoch": 0.06598544747694975,
      "grad_norm": 5.184841156005859,
      "learning_rate": 9.340764973730298e-05,
      "loss": 1.3124,
      "num_input_tokens_seen": 15172504,
      "step": 942
    },
    {
      "epoch": 0.06605549572267899,
      "grad_norm": 3.5243096351623535,
      "learning_rate": 9.340065148861647e-05,
      "loss": 0.8203,
      "num_input_tokens_seen": 15188888,
      "step": 943
    },
    {
      "epoch": 0.06612554396840824,
      "grad_norm": 4.041544437408447,
      "learning_rate": 9.339365323992995e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 15204672,
      "step": 944
    },
    {
      "epoch": 0.06619559221413748,
      "grad_norm": 3.720906972885132,
      "learning_rate": 9.338665499124344e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 15220688,
      "step": 945
    },
    {
      "epoch": 0.06626564045986673,
      "grad_norm": 3.9778380393981934,
      "learning_rate": 9.337965674255692e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 15236856,
      "step": 946
    },
    {
      "epoch": 0.06633568870559597,
      "grad_norm": 4.486488342285156,
      "learning_rate": 9.33726584938704e-05,
      "loss": 1.2408,
      "num_input_tokens_seen": 15253240,
      "step": 947
    },
    {
      "epoch": 0.06640573695132522,
      "grad_norm": 8.369994163513184,
      "learning_rate": 9.33656602451839e-05,
      "loss": 1.4841,
      "num_input_tokens_seen": 15267728,
      "step": 948
    },
    {
      "epoch": 0.06647578519705447,
      "grad_norm": 4.2056732177734375,
      "learning_rate": 9.335866199649737e-05,
      "loss": 1.4258,
      "num_input_tokens_seen": 15284112,
      "step": 949
    },
    {
      "epoch": 0.06654583344278371,
      "grad_norm": 4.396723747253418,
      "learning_rate": 9.335166374781086e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 15300496,
      "step": 950
    },
    {
      "epoch": 0.06661588168851296,
      "grad_norm": 3.7177491188049316,
      "learning_rate": 9.334466549912435e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 15316608,
      "step": 951
    },
    {
      "epoch": 0.0666859299342422,
      "grad_norm": 4.080933094024658,
      "learning_rate": 9.333766725043784e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 15332976,
      "step": 952
    },
    {
      "epoch": 0.06675597817997146,
      "grad_norm": 5.188856601715088,
      "learning_rate": 9.333066900175132e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 15349080,
      "step": 953
    },
    {
      "epoch": 0.06682602642570071,
      "grad_norm": 4.583539962768555,
      "learning_rate": 9.33236707530648e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 15365256,
      "step": 954
    },
    {
      "epoch": 0.06689607467142995,
      "grad_norm": 3.873830795288086,
      "learning_rate": 9.331667250437829e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 15381640,
      "step": 955
    },
    {
      "epoch": 0.0669661229171592,
      "grad_norm": 3.9574460983276367,
      "learning_rate": 9.330967425569176e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 15397800,
      "step": 956
    },
    {
      "epoch": 0.06703617116288844,
      "grad_norm": 3.8933448791503906,
      "learning_rate": 9.330267600700525e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 15414112,
      "step": 957
    },
    {
      "epoch": 0.06710621940861769,
      "grad_norm": 4.748478412628174,
      "learning_rate": 9.329567775831875e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 15430496,
      "step": 958
    },
    {
      "epoch": 0.06717626765434694,
      "grad_norm": 6.755379676818848,
      "learning_rate": 9.328867950963223e-05,
      "loss": 1.206,
      "num_input_tokens_seen": 15445072,
      "step": 959
    },
    {
      "epoch": 0.06724631590007618,
      "grad_norm": 4.382065773010254,
      "learning_rate": 9.328168126094572e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 15460336,
      "step": 960
    },
    {
      "epoch": 0.06731636414580543,
      "grad_norm": 5.037116527557373,
      "learning_rate": 9.327468301225919e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 15474752,
      "step": 961
    },
    {
      "epoch": 0.06738641239153467,
      "grad_norm": 5.838945388793945,
      "learning_rate": 9.326768476357268e-05,
      "loss": 1.314,
      "num_input_tokens_seen": 15491136,
      "step": 962
    },
    {
      "epoch": 0.06745646063726392,
      "grad_norm": 3.690436840057373,
      "learning_rate": 9.326068651488617e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 15507520,
      "step": 963
    },
    {
      "epoch": 0.06752650888299316,
      "grad_norm": 4.1123247146606445,
      "learning_rate": 9.325368826619966e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 15523904,
      "step": 964
    },
    {
      "epoch": 0.06759655712872241,
      "grad_norm": 4.120308876037598,
      "learning_rate": 9.324669001751315e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 15540136,
      "step": 965
    },
    {
      "epoch": 0.06766660537445165,
      "grad_norm": 3.9849514961242676,
      "learning_rate": 9.323969176882662e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 15556312,
      "step": 966
    },
    {
      "epoch": 0.0677366536201809,
      "grad_norm": 3.9164884090423584,
      "learning_rate": 9.323269352014011e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 15571864,
      "step": 967
    },
    {
      "epoch": 0.06780670186591015,
      "grad_norm": 4.282434940338135,
      "learning_rate": 9.32256952714536e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 15587800,
      "step": 968
    },
    {
      "epoch": 0.06787675011163939,
      "grad_norm": 4.118724346160889,
      "learning_rate": 9.321869702276707e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 15603128,
      "step": 969
    },
    {
      "epoch": 0.06794679835736864,
      "grad_norm": 4.233770847320557,
      "learning_rate": 9.321169877408056e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 15617864,
      "step": 970
    },
    {
      "epoch": 0.06801684660309788,
      "grad_norm": 3.933587074279785,
      "learning_rate": 9.320470052539405e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 15634248,
      "step": 971
    },
    {
      "epoch": 0.06808689484882713,
      "grad_norm": 4.641788482666016,
      "learning_rate": 9.319770227670754e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 15650304,
      "step": 972
    },
    {
      "epoch": 0.06815694309455637,
      "grad_norm": 4.138880729675293,
      "learning_rate": 9.319070402802102e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 15666688,
      "step": 973
    },
    {
      "epoch": 0.06822699134028562,
      "grad_norm": 4.823685169219971,
      "learning_rate": 9.31837057793345e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 15682936,
      "step": 974
    },
    {
      "epoch": 0.06829703958601487,
      "grad_norm": 4.432481288909912,
      "learning_rate": 9.317670753064799e-05,
      "loss": 0.9626,
      "num_input_tokens_seen": 15699320,
      "step": 975
    },
    {
      "epoch": 0.06836708783174411,
      "grad_norm": 4.115868091583252,
      "learning_rate": 9.316970928196147e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 15715296,
      "step": 976
    },
    {
      "epoch": 0.06843713607747336,
      "grad_norm": 3.964905023574829,
      "learning_rate": 9.316271103327496e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 15731680,
      "step": 977
    },
    {
      "epoch": 0.0685071843232026,
      "grad_norm": 3.686522960662842,
      "learning_rate": 9.315571278458846e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 15747808,
      "step": 978
    },
    {
      "epoch": 0.06857723256893185,
      "grad_norm": 4.0614423751831055,
      "learning_rate": 9.314871453590193e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 15764168,
      "step": 979
    },
    {
      "epoch": 0.0686472808146611,
      "grad_norm": 3.756350517272949,
      "learning_rate": 9.314171628721542e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 15780176,
      "step": 980
    },
    {
      "epoch": 0.06871732906039034,
      "grad_norm": 4.30344820022583,
      "learning_rate": 9.31347180385289e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 15795720,
      "step": 981
    },
    {
      "epoch": 0.06878737730611958,
      "grad_norm": 4.055768013000488,
      "learning_rate": 9.312771978984239e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 15811528,
      "step": 982
    },
    {
      "epoch": 0.06885742555184883,
      "grad_norm": 3.8779115676879883,
      "learning_rate": 9.312072154115586e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 15827392,
      "step": 983
    },
    {
      "epoch": 0.06892747379757808,
      "grad_norm": 5.014206886291504,
      "learning_rate": 9.311372329246936e-05,
      "loss": 1.3421,
      "num_input_tokens_seen": 15843776,
      "step": 984
    },
    {
      "epoch": 0.06899752204330732,
      "grad_norm": 4.548489570617676,
      "learning_rate": 9.310672504378285e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 15858880,
      "step": 985
    },
    {
      "epoch": 0.06906757028903657,
      "grad_norm": 4.312918186187744,
      "learning_rate": 9.309972679509633e-05,
      "loss": 1.2728,
      "num_input_tokens_seen": 15874840,
      "step": 986
    },
    {
      "epoch": 0.06913761853476583,
      "grad_norm": 3.9783735275268555,
      "learning_rate": 9.309272854640981e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 15890568,
      "step": 987
    },
    {
      "epoch": 0.06920766678049507,
      "grad_norm": 4.155986309051514,
      "learning_rate": 9.308573029772329e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 15906952,
      "step": 988
    },
    {
      "epoch": 0.06927771502622432,
      "grad_norm": 3.633018732070923,
      "learning_rate": 9.307873204903678e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 15923336,
      "step": 989
    },
    {
      "epoch": 0.06934776327195356,
      "grad_norm": 3.9513449668884277,
      "learning_rate": 9.307173380035027e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 15939720,
      "step": 990
    },
    {
      "epoch": 0.06941781151768281,
      "grad_norm": 4.296191692352295,
      "learning_rate": 9.306473555166376e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 15956104,
      "step": 991
    },
    {
      "epoch": 0.06948785976341205,
      "grad_norm": 5.266847133636475,
      "learning_rate": 9.305773730297724e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 15972488,
      "step": 992
    },
    {
      "epoch": 0.0695579080091413,
      "grad_norm": 4.321287155151367,
      "learning_rate": 9.305073905429072e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 15988408,
      "step": 993
    },
    {
      "epoch": 0.06962795625487055,
      "grad_norm": 4.1421613693237305,
      "learning_rate": 9.304374080560421e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 16002904,
      "step": 994
    },
    {
      "epoch": 0.06969800450059979,
      "grad_norm": 6.811270713806152,
      "learning_rate": 9.30367425569177e-05,
      "loss": 1.1012,
      "num_input_tokens_seen": 16017424,
      "step": 995
    },
    {
      "epoch": 0.06976805274632904,
      "grad_norm": 4.968684196472168,
      "learning_rate": 9.302974430823117e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 16033808,
      "step": 996
    },
    {
      "epoch": 0.06983810099205828,
      "grad_norm": 4.592737197875977,
      "learning_rate": 9.302274605954466e-05,
      "loss": 0.9698,
      "num_input_tokens_seen": 16050192,
      "step": 997
    },
    {
      "epoch": 0.06990814923778753,
      "grad_norm": 3.7984917163848877,
      "learning_rate": 9.301574781085815e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 16066192,
      "step": 998
    },
    {
      "epoch": 0.06997819748351677,
      "grad_norm": 4.594212055206299,
      "learning_rate": 9.300874956217164e-05,
      "loss": 1.3718,
      "num_input_tokens_seen": 16082576,
      "step": 999
    },
    {
      "epoch": 0.07004824572924602,
      "grad_norm": 5.062666893005371,
      "learning_rate": 9.300175131348511e-05,
      "loss": 1.3139,
      "num_input_tokens_seen": 16098960,
      "step": 1000
    },
    {
      "epoch": 0.07004824572924602,
      "eval_loss": 1.1650840044021606,
      "eval_runtime": 0.192,
      "eval_samples_per_second": 5.208,
      "eval_steps_per_second": 5.208,
      "num_input_tokens_seen": 16098960,
      "step": 1000
    },
    {
      "epoch": 0.07011829397497527,
      "grad_norm": 4.100902557373047,
      "learning_rate": 9.29947530647986e-05,
      "loss": 1.2711,
      "num_input_tokens_seen": 16115216,
      "step": 1001
    },
    {
      "epoch": 0.07018834222070451,
      "grad_norm": 4.24728536605835,
      "learning_rate": 9.298775481611209e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 16130080,
      "step": 1002
    },
    {
      "epoch": 0.07025839046643376,
      "grad_norm": 3.4653356075286865,
      "learning_rate": 9.298075656742556e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 16146400,
      "step": 1003
    },
    {
      "epoch": 0.070328438712163,
      "grad_norm": 5.548775672912598,
      "learning_rate": 9.297375831873907e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 16162784,
      "step": 1004
    },
    {
      "epoch": 0.07039848695789225,
      "grad_norm": 4.11661958694458,
      "learning_rate": 9.296676007005256e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 16179024,
      "step": 1005
    },
    {
      "epoch": 0.0704685352036215,
      "grad_norm": 4.006300449371338,
      "learning_rate": 9.295976182136603e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 16195408,
      "step": 1006
    },
    {
      "epoch": 0.07053858344935074,
      "grad_norm": 4.418802261352539,
      "learning_rate": 9.295276357267952e-05,
      "loss": 1.3845,
      "num_input_tokens_seen": 16211792,
      "step": 1007
    },
    {
      "epoch": 0.07060863169507998,
      "grad_norm": 5.625720024108887,
      "learning_rate": 9.2945765323993e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 16226584,
      "step": 1008
    },
    {
      "epoch": 0.07067867994080923,
      "grad_norm": 4.209630489349365,
      "learning_rate": 9.293876707530648e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 16242256,
      "step": 1009
    },
    {
      "epoch": 0.07074872818653848,
      "grad_norm": 4.0324788093566895,
      "learning_rate": 9.293176882661997e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 16258640,
      "step": 1010
    },
    {
      "epoch": 0.07081877643226772,
      "grad_norm": 4.0557684898376465,
      "learning_rate": 9.292477057793346e-05,
      "loss": 1.2831,
      "num_input_tokens_seen": 16275024,
      "step": 1011
    },
    {
      "epoch": 0.07088882467799697,
      "grad_norm": 4.511384010314941,
      "learning_rate": 9.291777232924695e-05,
      "loss": 1.1949,
      "num_input_tokens_seen": 16291112,
      "step": 1012
    },
    {
      "epoch": 0.07095887292372621,
      "grad_norm": 3.8120172023773193,
      "learning_rate": 9.291077408056042e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 16307496,
      "step": 1013
    },
    {
      "epoch": 0.07102892116945546,
      "grad_norm": 4.039558410644531,
      "learning_rate": 9.290377583187391e-05,
      "loss": 1.1575,
      "num_input_tokens_seen": 16323880,
      "step": 1014
    },
    {
      "epoch": 0.0710989694151847,
      "grad_norm": 3.9076366424560547,
      "learning_rate": 9.289677758318739e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 16339624,
      "step": 1015
    },
    {
      "epoch": 0.07116901766091395,
      "grad_norm": 3.8083527088165283,
      "learning_rate": 9.288977933450088e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 16356008,
      "step": 1016
    },
    {
      "epoch": 0.0712390659066432,
      "grad_norm": 4.5387282371521,
      "learning_rate": 9.288278108581436e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 16372392,
      "step": 1017
    },
    {
      "epoch": 0.07130911415237244,
      "grad_norm": 3.9228522777557373,
      "learning_rate": 9.287578283712785e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 16388776,
      "step": 1018
    },
    {
      "epoch": 0.07137916239810169,
      "grad_norm": 4.170912742614746,
      "learning_rate": 9.286878458844134e-05,
      "loss": 1.1324,
      "num_input_tokens_seen": 16405160,
      "step": 1019
    },
    {
      "epoch": 0.07144921064383093,
      "grad_norm": 4.426759719848633,
      "learning_rate": 9.286178633975482e-05,
      "loss": 1.2825,
      "num_input_tokens_seen": 16421544,
      "step": 1020
    },
    {
      "epoch": 0.07151925888956018,
      "grad_norm": 3.8606133460998535,
      "learning_rate": 9.28547880910683e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 16437736,
      "step": 1021
    },
    {
      "epoch": 0.07158930713528944,
      "grad_norm": 4.040006637573242,
      "learning_rate": 9.28477898423818e-05,
      "loss": 1.0824,
      "num_input_tokens_seen": 16453776,
      "step": 1022
    },
    {
      "epoch": 0.07165935538101868,
      "grad_norm": 3.7698042392730713,
      "learning_rate": 9.284079159369527e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 16470160,
      "step": 1023
    },
    {
      "epoch": 0.07172940362674793,
      "grad_norm": 4.180328369140625,
      "learning_rate": 9.283379334500877e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 16486280,
      "step": 1024
    },
    {
      "epoch": 0.07179945187247717,
      "grad_norm": 6.02299690246582,
      "learning_rate": 9.282679509632225e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 16501784,
      "step": 1025
    },
    {
      "epoch": 0.07186950011820642,
      "grad_norm": 4.239454746246338,
      "learning_rate": 9.281979684763573e-05,
      "loss": 1.3031,
      "num_input_tokens_seen": 16518096,
      "step": 1026
    },
    {
      "epoch": 0.07193954836393567,
      "grad_norm": 3.446030616760254,
      "learning_rate": 9.281279859894921e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 16534480,
      "step": 1027
    },
    {
      "epoch": 0.07200959660966491,
      "grad_norm": 4.2813568115234375,
      "learning_rate": 9.28058003502627e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 16550864,
      "step": 1028
    },
    {
      "epoch": 0.07207964485539416,
      "grad_norm": 5.289443016052246,
      "learning_rate": 9.279880210157619e-05,
      "loss": 1.3036,
      "num_input_tokens_seen": 16567248,
      "step": 1029
    },
    {
      "epoch": 0.0721496931011234,
      "grad_norm": 3.680283308029175,
      "learning_rate": 9.279180385288967e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 16583632,
      "step": 1030
    },
    {
      "epoch": 0.07221974134685265,
      "grad_norm": 4.283925533294678,
      "learning_rate": 9.278480560420316e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 16600016,
      "step": 1031
    },
    {
      "epoch": 0.0722897895925819,
      "grad_norm": 4.913532733917236,
      "learning_rate": 9.277780735551665e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 16616400,
      "step": 1032
    },
    {
      "epoch": 0.07235983783831114,
      "grad_norm": 4.344277381896973,
      "learning_rate": 9.277080910683013e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 16632024,
      "step": 1033
    },
    {
      "epoch": 0.07242988608404038,
      "grad_norm": 3.9231889247894287,
      "learning_rate": 9.276381085814362e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 16648408,
      "step": 1034
    },
    {
      "epoch": 0.07249993432976963,
      "grad_norm": 4.062288284301758,
      "learning_rate": 9.275681260945709e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 16664792,
      "step": 1035
    },
    {
      "epoch": 0.07256998257549888,
      "grad_norm": 4.163131237030029,
      "learning_rate": 9.274981436077058e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 16680216,
      "step": 1036
    },
    {
      "epoch": 0.07264003082122812,
      "grad_norm": 5.220231056213379,
      "learning_rate": 9.274281611208407e-05,
      "loss": 1.125,
      "num_input_tokens_seen": 16696160,
      "step": 1037
    },
    {
      "epoch": 0.07271007906695737,
      "grad_norm": 3.63785457611084,
      "learning_rate": 9.273581786339756e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 16712544,
      "step": 1038
    },
    {
      "epoch": 0.07278012731268661,
      "grad_norm": 4.612295627593994,
      "learning_rate": 9.272881961471105e-05,
      "loss": 1.3076,
      "num_input_tokens_seen": 16728928,
      "step": 1039
    },
    {
      "epoch": 0.07285017555841586,
      "grad_norm": 5.278262615203857,
      "learning_rate": 9.272182136602452e-05,
      "loss": 1.2682,
      "num_input_tokens_seen": 16744184,
      "step": 1040
    },
    {
      "epoch": 0.0729202238041451,
      "grad_norm": 4.3274455070495605,
      "learning_rate": 9.271482311733801e-05,
      "loss": 1.3517,
      "num_input_tokens_seen": 16760056,
      "step": 1041
    },
    {
      "epoch": 0.07299027204987435,
      "grad_norm": 4.1077375411987305,
      "learning_rate": 9.270782486865148e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 16776280,
      "step": 1042
    },
    {
      "epoch": 0.0730603202956036,
      "grad_norm": 3.954604148864746,
      "learning_rate": 9.270082661996497e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 16792456,
      "step": 1043
    },
    {
      "epoch": 0.07313036854133284,
      "grad_norm": 4.111297607421875,
      "learning_rate": 9.269382837127847e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 16808840,
      "step": 1044
    },
    {
      "epoch": 0.07320041678706209,
      "grad_norm": 3.56953501701355,
      "learning_rate": 9.268683012259195e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 16824720,
      "step": 1045
    },
    {
      "epoch": 0.07327046503279133,
      "grad_norm": 4.962648868560791,
      "learning_rate": 9.267983187390544e-05,
      "loss": 1.1714,
      "num_input_tokens_seen": 16841104,
      "step": 1046
    },
    {
      "epoch": 0.07334051327852058,
      "grad_norm": 3.7930710315704346,
      "learning_rate": 9.267283362521891e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 16857488,
      "step": 1047
    },
    {
      "epoch": 0.07341056152424982,
      "grad_norm": 4.158027172088623,
      "learning_rate": 9.26658353765324e-05,
      "loss": 1.1823,
      "num_input_tokens_seen": 16873856,
      "step": 1048
    },
    {
      "epoch": 0.07348060976997907,
      "grad_norm": 4.1571197509765625,
      "learning_rate": 9.265883712784589e-05,
      "loss": 1.2572,
      "num_input_tokens_seen": 16890240,
      "step": 1049
    },
    {
      "epoch": 0.07355065801570831,
      "grad_norm": 4.330874443054199,
      "learning_rate": 9.265183887915938e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 16906624,
      "step": 1050
    },
    {
      "epoch": 0.07362070626143756,
      "grad_norm": 6.105716705322266,
      "learning_rate": 9.264484063047287e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 16922864,
      "step": 1051
    },
    {
      "epoch": 0.0736907545071668,
      "grad_norm": 4.8344407081604,
      "learning_rate": 9.263784238178634e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 16939200,
      "step": 1052
    },
    {
      "epoch": 0.07376080275289605,
      "grad_norm": 3.553568124771118,
      "learning_rate": 9.263084413309983e-05,
      "loss": 0.7907,
      "num_input_tokens_seen": 16955584,
      "step": 1053
    },
    {
      "epoch": 0.0738308509986253,
      "grad_norm": 3.8178694248199463,
      "learning_rate": 9.26238458844133e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 16971968,
      "step": 1054
    },
    {
      "epoch": 0.07390089924435454,
      "grad_norm": 3.5509321689605713,
      "learning_rate": 9.26168476357268e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 16988352,
      "step": 1055
    },
    {
      "epoch": 0.0739709474900838,
      "grad_norm": 3.870811939239502,
      "learning_rate": 9.260984938704028e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 17004736,
      "step": 1056
    },
    {
      "epoch": 0.07404099573581305,
      "grad_norm": 11.86201286315918,
      "learning_rate": 9.260285113835377e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 17020544,
      "step": 1057
    },
    {
      "epoch": 0.0741110439815423,
      "grad_norm": 5.2176127433776855,
      "learning_rate": 9.259585288966726e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 17036472,
      "step": 1058
    },
    {
      "epoch": 0.07418109222727154,
      "grad_norm": 3.72566819190979,
      "learning_rate": 9.258885464098075e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 17052360,
      "step": 1059
    },
    {
      "epoch": 0.07425114047300078,
      "grad_norm": 4.323361396789551,
      "learning_rate": 9.258185639229422e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 17067672,
      "step": 1060
    },
    {
      "epoch": 0.07432118871873003,
      "grad_norm": 4.01705265045166,
      "learning_rate": 9.257485814360771e-05,
      "loss": 1.0402,
      "num_input_tokens_seen": 17084056,
      "step": 1061
    },
    {
      "epoch": 0.07439123696445928,
      "grad_norm": 4.4460039138793945,
      "learning_rate": 9.256785989492119e-05,
      "loss": 1.2294,
      "num_input_tokens_seen": 17100096,
      "step": 1062
    },
    {
      "epoch": 0.07446128521018852,
      "grad_norm": 4.634500503540039,
      "learning_rate": 9.256086164623468e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 17116440,
      "step": 1063
    },
    {
      "epoch": 0.07453133345591777,
      "grad_norm": 4.146971702575684,
      "learning_rate": 9.255386339754817e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 17132592,
      "step": 1064
    },
    {
      "epoch": 0.07460138170164701,
      "grad_norm": 6.171874523162842,
      "learning_rate": 9.254686514886165e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 17148704,
      "step": 1065
    },
    {
      "epoch": 0.07467142994737626,
      "grad_norm": 6.25461483001709,
      "learning_rate": 9.253986690017514e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 17164920,
      "step": 1066
    },
    {
      "epoch": 0.0747414781931055,
      "grad_norm": 3.886582851409912,
      "learning_rate": 9.253286865148862e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 17181304,
      "step": 1067
    },
    {
      "epoch": 0.07481152643883475,
      "grad_norm": 5.067885398864746,
      "learning_rate": 9.25258704028021e-05,
      "loss": 1.4475,
      "num_input_tokens_seen": 17197208,
      "step": 1068
    },
    {
      "epoch": 0.074881574684564,
      "grad_norm": 4.186190128326416,
      "learning_rate": 9.251887215411558e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 17212680,
      "step": 1069
    },
    {
      "epoch": 0.07495162293029324,
      "grad_norm": 4.059047698974609,
      "learning_rate": 9.251187390542908e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 17229064,
      "step": 1070
    },
    {
      "epoch": 0.07502167117602249,
      "grad_norm": 4.154530048370361,
      "learning_rate": 9.250487565674257e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 17245448,
      "step": 1071
    },
    {
      "epoch": 0.07509171942175173,
      "grad_norm": 3.760453701019287,
      "learning_rate": 9.249787740805605e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 17261832,
      "step": 1072
    },
    {
      "epoch": 0.07516176766748098,
      "grad_norm": 3.8155417442321777,
      "learning_rate": 9.249087915936954e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 17278216,
      "step": 1073
    },
    {
      "epoch": 0.07523181591321022,
      "grad_norm": 4.807973384857178,
      "learning_rate": 9.248388091068301e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 17294600,
      "step": 1074
    },
    {
      "epoch": 0.07530186415893947,
      "grad_norm": 11.421661376953125,
      "learning_rate": 9.24768826619965e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 17308960,
      "step": 1075
    },
    {
      "epoch": 0.07537191240466871,
      "grad_norm": 3.7491819858551025,
      "learning_rate": 9.246988441330999e-05,
      "loss": 1.1395,
      "num_input_tokens_seen": 17324536,
      "step": 1076
    },
    {
      "epoch": 0.07544196065039796,
      "grad_norm": 3.6289992332458496,
      "learning_rate": 9.246288616462348e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 17340920,
      "step": 1077
    },
    {
      "epoch": 0.0755120088961272,
      "grad_norm": 5.741896629333496,
      "learning_rate": 9.245588791593696e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 17357304,
      "step": 1078
    },
    {
      "epoch": 0.07558205714185645,
      "grad_norm": 3.5879697799682617,
      "learning_rate": 9.244888966725044e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 17373592,
      "step": 1079
    },
    {
      "epoch": 0.0756521053875857,
      "grad_norm": 7.3384504318237305,
      "learning_rate": 9.244189141856393e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 17387872,
      "step": 1080
    },
    {
      "epoch": 0.07572215363331494,
      "grad_norm": 3.6677255630493164,
      "learning_rate": 9.24348931698774e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 17403088,
      "step": 1081
    },
    {
      "epoch": 0.07579220187904419,
      "grad_norm": 3.953216075897217,
      "learning_rate": 9.242789492119089e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 17419392,
      "step": 1082
    },
    {
      "epoch": 0.07586225012477343,
      "grad_norm": 4.827987194061279,
      "learning_rate": 9.242089667250438e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 17435776,
      "step": 1083
    },
    {
      "epoch": 0.07593229837050268,
      "grad_norm": 4.416223526000977,
      "learning_rate": 9.241389842381787e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 17452080,
      "step": 1084
    },
    {
      "epoch": 0.07600234661623193,
      "grad_norm": 3.7776753902435303,
      "learning_rate": 9.240690017513136e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 17468160,
      "step": 1085
    },
    {
      "epoch": 0.07607239486196117,
      "grad_norm": 4.139477252960205,
      "learning_rate": 9.239990192644485e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 17484544,
      "step": 1086
    },
    {
      "epoch": 0.07614244310769042,
      "grad_norm": 5.218942642211914,
      "learning_rate": 9.239290367775832e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 17500928,
      "step": 1087
    },
    {
      "epoch": 0.07621249135341966,
      "grad_norm": 4.773080348968506,
      "learning_rate": 9.238590542907181e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 17517312,
      "step": 1088
    },
    {
      "epoch": 0.07628253959914891,
      "grad_norm": 3.840151309967041,
      "learning_rate": 9.237890718038528e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 17533696,
      "step": 1089
    },
    {
      "epoch": 0.07635258784487815,
      "grad_norm": 4.201962471008301,
      "learning_rate": 9.237190893169879e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 17549512,
      "step": 1090
    },
    {
      "epoch": 0.07642263609060741,
      "grad_norm": 4.4583001136779785,
      "learning_rate": 9.236491068301226e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 17565896,
      "step": 1091
    },
    {
      "epoch": 0.07649268433633666,
      "grad_norm": 4.013672351837158,
      "learning_rate": 9.235791243432575e-05,
      "loss": 1.2545,
      "num_input_tokens_seen": 17582264,
      "step": 1092
    },
    {
      "epoch": 0.0765627325820659,
      "grad_norm": 3.69555926322937,
      "learning_rate": 9.235091418563924e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 17597888,
      "step": 1093
    },
    {
      "epoch": 0.07663278082779515,
      "grad_norm": 4.341784954071045,
      "learning_rate": 9.234391593695271e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 17613392,
      "step": 1094
    },
    {
      "epoch": 0.0767028290735244,
      "grad_norm": 4.043522357940674,
      "learning_rate": 9.23369176882662e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 17629216,
      "step": 1095
    },
    {
      "epoch": 0.07677287731925364,
      "grad_norm": 4.330739498138428,
      "learning_rate": 9.232991943957969e-05,
      "loss": 1.2208,
      "num_input_tokens_seen": 17645600,
      "step": 1096
    },
    {
      "epoch": 0.07684292556498289,
      "grad_norm": 4.8433122634887695,
      "learning_rate": 9.232292119089318e-05,
      "loss": 0.9492,
      "num_input_tokens_seen": 17660952,
      "step": 1097
    },
    {
      "epoch": 0.07691297381071213,
      "grad_norm": 3.9039859771728516,
      "learning_rate": 9.231592294220667e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 17677336,
      "step": 1098
    },
    {
      "epoch": 0.07698302205644138,
      "grad_norm": 3.814103126525879,
      "learning_rate": 9.230892469352014e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 17693720,
      "step": 1099
    },
    {
      "epoch": 0.07705307030217062,
      "grad_norm": 3.9864039421081543,
      "learning_rate": 9.230192644483363e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 17710104,
      "step": 1100
    },
    {
      "epoch": 0.07712311854789987,
      "grad_norm": 4.469820499420166,
      "learning_rate": 9.229492819614711e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 17726488,
      "step": 1101
    },
    {
      "epoch": 0.07719316679362911,
      "grad_norm": 3.8044216632843018,
      "learning_rate": 9.22879299474606e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 17742648,
      "step": 1102
    },
    {
      "epoch": 0.07726321503935836,
      "grad_norm": 4.859435558319092,
      "learning_rate": 9.228093169877408e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 17759032,
      "step": 1103
    },
    {
      "epoch": 0.0773332632850876,
      "grad_norm": 3.830214023590088,
      "learning_rate": 9.227393345008757e-05,
      "loss": 1.1731,
      "num_input_tokens_seen": 17774872,
      "step": 1104
    },
    {
      "epoch": 0.07740331153081685,
      "grad_norm": 4.196676254272461,
      "learning_rate": 9.226693520140106e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 17790832,
      "step": 1105
    },
    {
      "epoch": 0.0774733597765461,
      "grad_norm": 4.50007438659668,
      "learning_rate": 9.225993695271454e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 17805024,
      "step": 1106
    },
    {
      "epoch": 0.07754340802227534,
      "grad_norm": 4.392070293426514,
      "learning_rate": 9.225293870402803e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 17820008,
      "step": 1107
    },
    {
      "epoch": 0.07761345626800459,
      "grad_norm": 4.09447717666626,
      "learning_rate": 9.22459404553415e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 17836392,
      "step": 1108
    },
    {
      "epoch": 0.07768350451373383,
      "grad_norm": 4.591554641723633,
      "learning_rate": 9.223894220665499e-05,
      "loss": 1.2772,
      "num_input_tokens_seen": 17852776,
      "step": 1109
    },
    {
      "epoch": 0.07775355275946308,
      "grad_norm": 5.629931926727295,
      "learning_rate": 9.223194395796849e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 17869160,
      "step": 1110
    },
    {
      "epoch": 0.07782360100519232,
      "grad_norm": 4.307553768157959,
      "learning_rate": 9.222494570928197e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 17885544,
      "step": 1111
    },
    {
      "epoch": 0.07789364925092157,
      "grad_norm": 4.599300384521484,
      "learning_rate": 9.221794746059545e-05,
      "loss": 1.1304,
      "num_input_tokens_seen": 17901848,
      "step": 1112
    },
    {
      "epoch": 0.07796369749665082,
      "grad_norm": 4.217408657073975,
      "learning_rate": 9.221094921190894e-05,
      "loss": 1.1611,
      "num_input_tokens_seen": 17918232,
      "step": 1113
    },
    {
      "epoch": 0.07803374574238006,
      "grad_norm": 3.885847568511963,
      "learning_rate": 9.220395096322242e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 17934504,
      "step": 1114
    },
    {
      "epoch": 0.07810379398810931,
      "grad_norm": 4.280134677886963,
      "learning_rate": 9.219695271453591e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 17950888,
      "step": 1115
    },
    {
      "epoch": 0.07817384223383855,
      "grad_norm": 4.081259727478027,
      "learning_rate": 9.21899544658494e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 17967088,
      "step": 1116
    },
    {
      "epoch": 0.0782438904795678,
      "grad_norm": 4.206293106079102,
      "learning_rate": 9.218295621716288e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 17983312,
      "step": 1117
    },
    {
      "epoch": 0.07831393872529704,
      "grad_norm": 4.837226390838623,
      "learning_rate": 9.217595796847636e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 17998768,
      "step": 1118
    },
    {
      "epoch": 0.07838398697102629,
      "grad_norm": 4.344440460205078,
      "learning_rate": 9.216895971978985e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 18014840,
      "step": 1119
    },
    {
      "epoch": 0.07845403521675554,
      "grad_norm": 4.357896327972412,
      "learning_rate": 9.216196147110334e-05,
      "loss": 1.2444,
      "num_input_tokens_seen": 18030696,
      "step": 1120
    },
    {
      "epoch": 0.07852408346248478,
      "grad_norm": 3.6449878215789795,
      "learning_rate": 9.215496322241681e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 18047024,
      "step": 1121
    },
    {
      "epoch": 0.07859413170821403,
      "grad_norm": 4.154385566711426,
      "learning_rate": 9.21479649737303e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 18063408,
      "step": 1122
    },
    {
      "epoch": 0.07866417995394327,
      "grad_norm": 3.5929031372070312,
      "learning_rate": 9.214096672504379e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 18079280,
      "step": 1123
    },
    {
      "epoch": 0.07873422819967252,
      "grad_norm": 3.5724170207977295,
      "learning_rate": 9.213396847635728e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 18094488,
      "step": 1124
    },
    {
      "epoch": 0.07880427644540176,
      "grad_norm": 4.100067615509033,
      "learning_rate": 9.212697022767077e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 18110872,
      "step": 1125
    },
    {
      "epoch": 0.07887432469113102,
      "grad_norm": 4.431338787078857,
      "learning_rate": 9.211997197898424e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 18127256,
      "step": 1126
    },
    {
      "epoch": 0.07894437293686027,
      "grad_norm": 3.9577043056488037,
      "learning_rate": 9.211297373029773e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 18143208,
      "step": 1127
    },
    {
      "epoch": 0.07901442118258951,
      "grad_norm": 4.753921985626221,
      "learning_rate": 9.21059754816112e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 18158888,
      "step": 1128
    },
    {
      "epoch": 0.07908446942831876,
      "grad_norm": 3.763982057571411,
      "learning_rate": 9.209897723292469e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 18175192,
      "step": 1129
    },
    {
      "epoch": 0.079154517674048,
      "grad_norm": 3.729553699493408,
      "learning_rate": 9.20919789842382e-05,
      "loss": 1.1152,
      "num_input_tokens_seen": 18191384,
      "step": 1130
    },
    {
      "epoch": 0.07922456591977725,
      "grad_norm": 3.7760956287384033,
      "learning_rate": 9.208498073555167e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 18207768,
      "step": 1131
    },
    {
      "epoch": 0.0792946141655065,
      "grad_norm": 4.64035177230835,
      "learning_rate": 9.207798248686516e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 18224152,
      "step": 1132
    },
    {
      "epoch": 0.07936466241123574,
      "grad_norm": 4.1443352699279785,
      "learning_rate": 9.207098423817863e-05,
      "loss": 1.2329,
      "num_input_tokens_seen": 18240536,
      "step": 1133
    },
    {
      "epoch": 0.07943471065696499,
      "grad_norm": 5.332706451416016,
      "learning_rate": 9.206398598949212e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 18255528,
      "step": 1134
    },
    {
      "epoch": 0.07950475890269423,
      "grad_norm": 3.914705514907837,
      "learning_rate": 9.20569877408056e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 18271768,
      "step": 1135
    },
    {
      "epoch": 0.07957480714842348,
      "grad_norm": 4.994162559509277,
      "learning_rate": 9.20499894921191e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 18288152,
      "step": 1136
    },
    {
      "epoch": 0.07964485539415272,
      "grad_norm": 4.132298946380615,
      "learning_rate": 9.204299124343259e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 18303784,
      "step": 1137
    },
    {
      "epoch": 0.07971490363988197,
      "grad_norm": 3.9048449993133545,
      "learning_rate": 9.203599299474606e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 18319968,
      "step": 1138
    },
    {
      "epoch": 0.07978495188561122,
      "grad_norm": 3.981844425201416,
      "learning_rate": 9.202899474605955e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 18335976,
      "step": 1139
    },
    {
      "epoch": 0.07985500013134046,
      "grad_norm": 4.491240501403809,
      "learning_rate": 9.202199649737304e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 18352360,
      "step": 1140
    },
    {
      "epoch": 0.07992504837706971,
      "grad_norm": 4.152430534362793,
      "learning_rate": 9.201499824868652e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 18368736,
      "step": 1141
    },
    {
      "epoch": 0.07999509662279895,
      "grad_norm": 4.337832450866699,
      "learning_rate": 9.2008e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 18385120,
      "step": 1142
    },
    {
      "epoch": 0.0800651448685282,
      "grad_norm": 4.865042209625244,
      "learning_rate": 9.200100175131349e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 18401504,
      "step": 1143
    },
    {
      "epoch": 0.08013519311425744,
      "grad_norm": 3.783113479614258,
      "learning_rate": 9.199400350262698e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 18417176,
      "step": 1144
    },
    {
      "epoch": 0.08020524135998669,
      "grad_norm": 4.98455286026001,
      "learning_rate": 9.198700525394046e-05,
      "loss": 1.2139,
      "num_input_tokens_seen": 18432584,
      "step": 1145
    },
    {
      "epoch": 0.08027528960571594,
      "grad_norm": 4.1859517097473145,
      "learning_rate": 9.198000700525394e-05,
      "loss": 1.1333,
      "num_input_tokens_seen": 18448968,
      "step": 1146
    },
    {
      "epoch": 0.08034533785144518,
      "grad_norm": 3.7193386554718018,
      "learning_rate": 9.197300875656743e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 18465352,
      "step": 1147
    },
    {
      "epoch": 0.08041538609717443,
      "grad_norm": 4.280893325805664,
      "learning_rate": 9.196601050788091e-05,
      "loss": 1.1261,
      "num_input_tokens_seen": 18481736,
      "step": 1148
    },
    {
      "epoch": 0.08048543434290367,
      "grad_norm": 3.9979352951049805,
      "learning_rate": 9.19590122591944e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 18498120,
      "step": 1149
    },
    {
      "epoch": 0.08055548258863292,
      "grad_norm": 5.594225883483887,
      "learning_rate": 9.195201401050789e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 18513944,
      "step": 1150
    },
    {
      "epoch": 0.08062553083436216,
      "grad_norm": 4.758842468261719,
      "learning_rate": 9.194501576182137e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 18530328,
      "step": 1151
    },
    {
      "epoch": 0.08069557908009141,
      "grad_norm": 5.597489356994629,
      "learning_rate": 9.193801751313486e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 18546632,
      "step": 1152
    },
    {
      "epoch": 0.08076562732582065,
      "grad_norm": 5.279472827911377,
      "learning_rate": 9.193101926444834e-05,
      "loss": 1.2897,
      "num_input_tokens_seen": 18561856,
      "step": 1153
    },
    {
      "epoch": 0.0808356755715499,
      "grad_norm": 4.672069072723389,
      "learning_rate": 9.192402101576183e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 18577944,
      "step": 1154
    },
    {
      "epoch": 0.08090572381727915,
      "grad_norm": 3.65533447265625,
      "learning_rate": 9.19170227670753e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 18593720,
      "step": 1155
    },
    {
      "epoch": 0.08097577206300839,
      "grad_norm": 4.212414741516113,
      "learning_rate": 9.19100245183888e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 18609864,
      "step": 1156
    },
    {
      "epoch": 0.08104582030873764,
      "grad_norm": 4.471503734588623,
      "learning_rate": 9.190302626970229e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 18626248,
      "step": 1157
    },
    {
      "epoch": 0.08111586855446688,
      "grad_norm": 4.952723979949951,
      "learning_rate": 9.189602802101577e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 18642632,
      "step": 1158
    },
    {
      "epoch": 0.08118591680019613,
      "grad_norm": 3.921449661254883,
      "learning_rate": 9.188902977232926e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 18659016,
      "step": 1159
    },
    {
      "epoch": 0.08125596504592539,
      "grad_norm": 3.728752374649048,
      "learning_rate": 9.188203152364273e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 18675400,
      "step": 1160
    },
    {
      "epoch": 0.08132601329165463,
      "grad_norm": 3.8742613792419434,
      "learning_rate": 9.187503327495622e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 18691232,
      "step": 1161
    },
    {
      "epoch": 0.08139606153738388,
      "grad_norm": 3.827157735824585,
      "learning_rate": 9.186803502626971e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 18707616,
      "step": 1162
    },
    {
      "epoch": 0.08146610978311312,
      "grad_norm": 3.8507778644561768,
      "learning_rate": 9.18610367775832e-05,
      "loss": 1.0317,
      "num_input_tokens_seen": 18724000,
      "step": 1163
    },
    {
      "epoch": 0.08153615802884237,
      "grad_norm": 5.328095436096191,
      "learning_rate": 9.185403852889669e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 18740384,
      "step": 1164
    },
    {
      "epoch": 0.08160620627457162,
      "grad_norm": 4.8900322914123535,
      "learning_rate": 9.184704028021016e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 18756768,
      "step": 1165
    },
    {
      "epoch": 0.08167625452030086,
      "grad_norm": 3.810084104537964,
      "learning_rate": 9.184004203152365e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 18772632,
      "step": 1166
    },
    {
      "epoch": 0.08174630276603011,
      "grad_norm": 4.318419456481934,
      "learning_rate": 9.183304378283714e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 18788272,
      "step": 1167
    },
    {
      "epoch": 0.08181635101175935,
      "grad_norm": 4.093379020690918,
      "learning_rate": 9.182604553415061e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 18803672,
      "step": 1168
    },
    {
      "epoch": 0.0818863992574886,
      "grad_norm": 4.630450248718262,
      "learning_rate": 9.18190472854641e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 18820056,
      "step": 1169
    },
    {
      "epoch": 0.08195644750321784,
      "grad_norm": 4.388457775115967,
      "learning_rate": 9.181204903677759e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 18836440,
      "step": 1170
    },
    {
      "epoch": 0.08202649574894709,
      "grad_norm": 3.6942262649536133,
      "learning_rate": 9.180505078809108e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 18852824,
      "step": 1171
    },
    {
      "epoch": 0.08209654399467634,
      "grad_norm": 3.937696933746338,
      "learning_rate": 9.179805253940455e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 18869208,
      "step": 1172
    },
    {
      "epoch": 0.08216659224040558,
      "grad_norm": 4.062703609466553,
      "learning_rate": 9.179105429071804e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 18885320,
      "step": 1173
    },
    {
      "epoch": 0.08223664048613483,
      "grad_norm": 7.794081211090088,
      "learning_rate": 9.178405604203153e-05,
      "loss": 1.2287,
      "num_input_tokens_seen": 18900224,
      "step": 1174
    },
    {
      "epoch": 0.08230668873186407,
      "grad_norm": 4.429391860961914,
      "learning_rate": 9.1777057793345e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 18916456,
      "step": 1175
    },
    {
      "epoch": 0.08237673697759332,
      "grad_norm": 3.954869508743286,
      "learning_rate": 9.17700595446585e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 18932840,
      "step": 1176
    },
    {
      "epoch": 0.08244678522332256,
      "grad_norm": 5.555337429046631,
      "learning_rate": 9.176306129597198e-05,
      "loss": 1.3628,
      "num_input_tokens_seen": 18949224,
      "step": 1177
    },
    {
      "epoch": 0.08251683346905181,
      "grad_norm": 3.575295925140381,
      "learning_rate": 9.175606304728547e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 18965552,
      "step": 1178
    },
    {
      "epoch": 0.08258688171478105,
      "grad_norm": 5.927703380584717,
      "learning_rate": 9.174906479859896e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 18981496,
      "step": 1179
    },
    {
      "epoch": 0.0826569299605103,
      "grad_norm": 6.553986549377441,
      "learning_rate": 9.174206654991243e-05,
      "loss": 1.4058,
      "num_input_tokens_seen": 18996808,
      "step": 1180
    },
    {
      "epoch": 0.08272697820623955,
      "grad_norm": 4.315832138061523,
      "learning_rate": 9.173506830122592e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 19013192,
      "step": 1181
    },
    {
      "epoch": 0.08279702645196879,
      "grad_norm": 3.818033218383789,
      "learning_rate": 9.172807005253941e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 19029464,
      "step": 1182
    },
    {
      "epoch": 0.08286707469769804,
      "grad_norm": 3.4207711219787598,
      "learning_rate": 9.17210718038529e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 19045592,
      "step": 1183
    },
    {
      "epoch": 0.08293712294342728,
      "grad_norm": 4.3305864334106445,
      "learning_rate": 9.171407355516639e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 19061864,
      "step": 1184
    },
    {
      "epoch": 0.08300717118915653,
      "grad_norm": 5.365218162536621,
      "learning_rate": 9.170707530647986e-05,
      "loss": 1.1669,
      "num_input_tokens_seen": 19075448,
      "step": 1185
    },
    {
      "epoch": 0.08307721943488577,
      "grad_norm": 3.9939708709716797,
      "learning_rate": 9.170007705779335e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 19091832,
      "step": 1186
    },
    {
      "epoch": 0.08314726768061502,
      "grad_norm": 3.8088884353637695,
      "learning_rate": 9.169307880910683e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 19107920,
      "step": 1187
    },
    {
      "epoch": 0.08321731592634427,
      "grad_norm": 3.858799457550049,
      "learning_rate": 9.168608056042032e-05,
      "loss": 0.9805,
      "num_input_tokens_seen": 19123776,
      "step": 1188
    },
    {
      "epoch": 0.08328736417207351,
      "grad_norm": 4.042770862579346,
      "learning_rate": 9.16790823117338e-05,
      "loss": 1.1668,
      "num_input_tokens_seen": 19139752,
      "step": 1189
    },
    {
      "epoch": 0.08335741241780276,
      "grad_norm": 4.2054762840271,
      "learning_rate": 9.16720840630473e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 19156136,
      "step": 1190
    },
    {
      "epoch": 0.083427460663532,
      "grad_norm": 4.450238227844238,
      "learning_rate": 9.166508581436078e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 19172240,
      "step": 1191
    },
    {
      "epoch": 0.08349750890926125,
      "grad_norm": 4.126129627227783,
      "learning_rate": 9.165808756567426e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 19188624,
      "step": 1192
    },
    {
      "epoch": 0.0835675571549905,
      "grad_norm": 4.131893157958984,
      "learning_rate": 9.165108931698775e-05,
      "loss": 1.2004,
      "num_input_tokens_seen": 19205008,
      "step": 1193
    },
    {
      "epoch": 0.08363760540071974,
      "grad_norm": 4.25187873840332,
      "learning_rate": 9.164409106830123e-05,
      "loss": 1.3571,
      "num_input_tokens_seen": 19220856,
      "step": 1194
    },
    {
      "epoch": 0.083707653646449,
      "grad_norm": 3.842498302459717,
      "learning_rate": 9.163709281961471e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 19237208,
      "step": 1195
    },
    {
      "epoch": 0.08377770189217824,
      "grad_norm": 3.694279432296753,
      "learning_rate": 9.16300945709282e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 19253592,
      "step": 1196
    },
    {
      "epoch": 0.08384775013790749,
      "grad_norm": 4.382254123687744,
      "learning_rate": 9.162309632224169e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 19269976,
      "step": 1197
    },
    {
      "epoch": 0.08391779838363674,
      "grad_norm": 4.267289161682129,
      "learning_rate": 9.161609807355518e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 19286360,
      "step": 1198
    },
    {
      "epoch": 0.08398784662936598,
      "grad_norm": 5.554534435272217,
      "learning_rate": 9.160909982486865e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 19301800,
      "step": 1199
    },
    {
      "epoch": 0.08405789487509523,
      "grad_norm": 4.1479668617248535,
      "learning_rate": 9.160210157618214e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 19317392,
      "step": 1200
    },
    {
      "epoch": 0.08405789487509523,
      "eval_loss": 1.1600490808486938,
      "eval_runtime": 0.2015,
      "eval_samples_per_second": 4.962,
      "eval_steps_per_second": 4.962,
      "num_input_tokens_seen": 19317392,
      "step": 1200
    },
    {
      "epoch": 0.08412794312082447,
      "grad_norm": 4.1876349449157715,
      "learning_rate": 9.159510332749563e-05,
      "loss": 1.2036,
      "num_input_tokens_seen": 19333776,
      "step": 1201
    },
    {
      "epoch": 0.08419799136655372,
      "grad_norm": 4.031203746795654,
      "learning_rate": 9.15881050788091e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 19349616,
      "step": 1202
    },
    {
      "epoch": 0.08426803961228296,
      "grad_norm": 4.013350963592529,
      "learning_rate": 9.15811068301226e-05,
      "loss": 1.2147,
      "num_input_tokens_seen": 19366000,
      "step": 1203
    },
    {
      "epoch": 0.08433808785801221,
      "grad_norm": 4.509790897369385,
      "learning_rate": 9.157410858143608e-05,
      "loss": 1.3484,
      "num_input_tokens_seen": 19381904,
      "step": 1204
    },
    {
      "epoch": 0.08440813610374145,
      "grad_norm": 4.630336761474609,
      "learning_rate": 9.156711033274957e-05,
      "loss": 1.0246,
      "num_input_tokens_seen": 19398288,
      "step": 1205
    },
    {
      "epoch": 0.0844781843494707,
      "grad_norm": 3.819884777069092,
      "learning_rate": 9.156011208406304e-05,
      "loss": 1.1242,
      "num_input_tokens_seen": 19414248,
      "step": 1206
    },
    {
      "epoch": 0.08454823259519995,
      "grad_norm": 3.7933132648468018,
      "learning_rate": 9.155311383537653e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 19430632,
      "step": 1207
    },
    {
      "epoch": 0.08461828084092919,
      "grad_norm": 5.7384934425354,
      "learning_rate": 9.154611558669002e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 19446248,
      "step": 1208
    },
    {
      "epoch": 0.08468832908665844,
      "grad_norm": 3.9594175815582275,
      "learning_rate": 9.153911733800351e-05,
      "loss": 1.2029,
      "num_input_tokens_seen": 19462632,
      "step": 1209
    },
    {
      "epoch": 0.08475837733238768,
      "grad_norm": 3.8251891136169434,
      "learning_rate": 9.1532119089317e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 19479016,
      "step": 1210
    },
    {
      "epoch": 0.08482842557811693,
      "grad_norm": 3.9750332832336426,
      "learning_rate": 9.152512084063049e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 19495112,
      "step": 1211
    },
    {
      "epoch": 0.08489847382384617,
      "grad_norm": 3.986170530319214,
      "learning_rate": 9.151812259194396e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 19511216,
      "step": 1212
    },
    {
      "epoch": 0.08496852206957542,
      "grad_norm": 3.914065361022949,
      "learning_rate": 9.151112434325745e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 19527600,
      "step": 1213
    },
    {
      "epoch": 0.08503857031530467,
      "grad_norm": 4.328094482421875,
      "learning_rate": 9.150412609457093e-05,
      "loss": 1.2076,
      "num_input_tokens_seen": 19543984,
      "step": 1214
    },
    {
      "epoch": 0.08510861856103391,
      "grad_norm": 4.112467288970947,
      "learning_rate": 9.149712784588441e-05,
      "loss": 1.1732,
      "num_input_tokens_seen": 19560368,
      "step": 1215
    },
    {
      "epoch": 0.08517866680676316,
      "grad_norm": 4.680009365081787,
      "learning_rate": 9.14901295971979e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 19575616,
      "step": 1216
    },
    {
      "epoch": 0.0852487150524924,
      "grad_norm": 4.4872660636901855,
      "learning_rate": 9.148313134851139e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 19592000,
      "step": 1217
    },
    {
      "epoch": 0.08531876329822165,
      "grad_norm": 3.7546637058258057,
      "learning_rate": 9.147613309982488e-05,
      "loss": 1.1989,
      "num_input_tokens_seen": 19608384,
      "step": 1218
    },
    {
      "epoch": 0.0853888115439509,
      "grad_norm": 5.590888500213623,
      "learning_rate": 9.146913485113835e-05,
      "loss": 1.1411,
      "num_input_tokens_seen": 19624768,
      "step": 1219
    },
    {
      "epoch": 0.08545885978968014,
      "grad_norm": 3.958021640777588,
      "learning_rate": 9.146213660245184e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 19641152,
      "step": 1220
    },
    {
      "epoch": 0.08552890803540938,
      "grad_norm": 3.7641196250915527,
      "learning_rate": 9.145513835376533e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 19657536,
      "step": 1221
    },
    {
      "epoch": 0.08559895628113863,
      "grad_norm": 4.395461559295654,
      "learning_rate": 9.14481401050788e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 19673712,
      "step": 1222
    },
    {
      "epoch": 0.08566900452686788,
      "grad_norm": 3.8162319660186768,
      "learning_rate": 9.144114185639231e-05,
      "loss": 1.1638,
      "num_input_tokens_seen": 19689336,
      "step": 1223
    },
    {
      "epoch": 0.08573905277259712,
      "grad_norm": 3.7025444507598877,
      "learning_rate": 9.143414360770578e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 19705464,
      "step": 1224
    },
    {
      "epoch": 0.08580910101832637,
      "grad_norm": 3.8621439933776855,
      "learning_rate": 9.142714535901927e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 19721848,
      "step": 1225
    },
    {
      "epoch": 0.08587914926405561,
      "grad_norm": 4.243250846862793,
      "learning_rate": 9.142014711033275e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 19738072,
      "step": 1226
    },
    {
      "epoch": 0.08594919750978486,
      "grad_norm": 4.05800724029541,
      "learning_rate": 9.141314886164624e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 19754456,
      "step": 1227
    },
    {
      "epoch": 0.0860192457555141,
      "grad_norm": 4.0894455909729,
      "learning_rate": 9.140615061295972e-05,
      "loss": 1.254,
      "num_input_tokens_seen": 19770840,
      "step": 1228
    },
    {
      "epoch": 0.08608929400124336,
      "grad_norm": 4.296894073486328,
      "learning_rate": 9.139915236427321e-05,
      "loss": 1.1298,
      "num_input_tokens_seen": 19786864,
      "step": 1229
    },
    {
      "epoch": 0.08615934224697261,
      "grad_norm": 4.0352888107299805,
      "learning_rate": 9.13921541155867e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 19801800,
      "step": 1230
    },
    {
      "epoch": 0.08622939049270185,
      "grad_norm": 4.087375640869141,
      "learning_rate": 9.138515586690018e-05,
      "loss": 0.9686,
      "num_input_tokens_seen": 19818184,
      "step": 1231
    },
    {
      "epoch": 0.0862994387384311,
      "grad_norm": 4.045078754425049,
      "learning_rate": 9.137815761821367e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 19833016,
      "step": 1232
    },
    {
      "epoch": 0.08636948698416035,
      "grad_norm": 4.399363040924072,
      "learning_rate": 9.137115936952714e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 19848912,
      "step": 1233
    },
    {
      "epoch": 0.08643953522988959,
      "grad_norm": 4.420406818389893,
      "learning_rate": 9.136416112084063e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 19865296,
      "step": 1234
    },
    {
      "epoch": 0.08650958347561884,
      "grad_norm": 4.131808280944824,
      "learning_rate": 9.135716287215412e-05,
      "loss": 1.1865,
      "num_input_tokens_seen": 19881376,
      "step": 1235
    },
    {
      "epoch": 0.08657963172134808,
      "grad_norm": 3.8256850242614746,
      "learning_rate": 9.13501646234676e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 19897704,
      "step": 1236
    },
    {
      "epoch": 0.08664967996707733,
      "grad_norm": 4.3497233390808105,
      "learning_rate": 9.13431663747811e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 19914088,
      "step": 1237
    },
    {
      "epoch": 0.08671972821280657,
      "grad_norm": 4.18136739730835,
      "learning_rate": 9.133616812609458e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 19930128,
      "step": 1238
    },
    {
      "epoch": 0.08678977645853582,
      "grad_norm": 4.782970905303955,
      "learning_rate": 9.132916987740806e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 19946512,
      "step": 1239
    },
    {
      "epoch": 0.08685982470426507,
      "grad_norm": 4.16589879989624,
      "learning_rate": 9.132217162872155e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 19962488,
      "step": 1240
    },
    {
      "epoch": 0.08692987294999431,
      "grad_norm": 3.73541522026062,
      "learning_rate": 9.131517338003502e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 19978584,
      "step": 1241
    },
    {
      "epoch": 0.08699992119572356,
      "grad_norm": 4.225815773010254,
      "learning_rate": 9.130817513134851e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 19994816,
      "step": 1242
    },
    {
      "epoch": 0.0870699694414528,
      "grad_norm": 7.807470321655273,
      "learning_rate": 9.1301176882662e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 20010576,
      "step": 1243
    },
    {
      "epoch": 0.08714001768718205,
      "grad_norm": 4.818174839019775,
      "learning_rate": 9.129417863397549e-05,
      "loss": 1.1892,
      "num_input_tokens_seen": 20025712,
      "step": 1244
    },
    {
      "epoch": 0.0872100659329113,
      "grad_norm": 3.8367979526519775,
      "learning_rate": 9.128718038528898e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 20041904,
      "step": 1245
    },
    {
      "epoch": 0.08728011417864054,
      "grad_norm": 3.9912586212158203,
      "learning_rate": 9.128018213660245e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 20058288,
      "step": 1246
    },
    {
      "epoch": 0.08735016242436978,
      "grad_norm": 4.842557907104492,
      "learning_rate": 9.127318388791594e-05,
      "loss": 1.2012,
      "num_input_tokens_seen": 20074672,
      "step": 1247
    },
    {
      "epoch": 0.08742021067009903,
      "grad_norm": 3.816938877105713,
      "learning_rate": 9.126618563922943e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 20090664,
      "step": 1248
    },
    {
      "epoch": 0.08749025891582828,
      "grad_norm": 3.712480306625366,
      "learning_rate": 9.125918739054292e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 20107048,
      "step": 1249
    },
    {
      "epoch": 0.08756030716155752,
      "grad_norm": 4.185492515563965,
      "learning_rate": 9.12521891418564e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 20123432,
      "step": 1250
    },
    {
      "epoch": 0.08763035540728677,
      "grad_norm": 5.510714530944824,
      "learning_rate": 9.124519089316988e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 20139112,
      "step": 1251
    },
    {
      "epoch": 0.08770040365301601,
      "grad_norm": 3.9170289039611816,
      "learning_rate": 9.123819264448337e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 20155496,
      "step": 1252
    },
    {
      "epoch": 0.08777045189874526,
      "grad_norm": 3.738008975982666,
      "learning_rate": 9.123119439579684e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 20171760,
      "step": 1253
    },
    {
      "epoch": 0.0878405001444745,
      "grad_norm": 4.845873832702637,
      "learning_rate": 9.122419614711033e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 20188056,
      "step": 1254
    },
    {
      "epoch": 0.08791054839020375,
      "grad_norm": 4.166906356811523,
      "learning_rate": 9.121719789842382e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 20204440,
      "step": 1255
    },
    {
      "epoch": 0.087980596635933,
      "grad_norm": 4.039194107055664,
      "learning_rate": 9.121019964973731e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 20220824,
      "step": 1256
    },
    {
      "epoch": 0.08805064488166224,
      "grad_norm": 3.545482635498047,
      "learning_rate": 9.12032014010508e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 20236888,
      "step": 1257
    },
    {
      "epoch": 0.08812069312739149,
      "grad_norm": 3.82114839553833,
      "learning_rate": 9.119620315236427e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 20253272,
      "step": 1258
    },
    {
      "epoch": 0.08819074137312073,
      "grad_norm": 4.770678997039795,
      "learning_rate": 9.118920490367776e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 20269656,
      "step": 1259
    },
    {
      "epoch": 0.08826078961884998,
      "grad_norm": 4.4319539070129395,
      "learning_rate": 9.118220665499124e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 20285456,
      "step": 1260
    },
    {
      "epoch": 0.08833083786457922,
      "grad_norm": 4.0923357009887695,
      "learning_rate": 9.117520840630473e-05,
      "loss": 1.2328,
      "num_input_tokens_seen": 20301232,
      "step": 1261
    },
    {
      "epoch": 0.08840088611030847,
      "grad_norm": 5.8347344398498535,
      "learning_rate": 9.116821015761821e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 20317224,
      "step": 1262
    },
    {
      "epoch": 0.08847093435603771,
      "grad_norm": 4.525367259979248,
      "learning_rate": 9.11612119089317e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 20332616,
      "step": 1263
    },
    {
      "epoch": 0.08854098260176697,
      "grad_norm": 3.9754436016082764,
      "learning_rate": 9.115421366024519e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 20348336,
      "step": 1264
    },
    {
      "epoch": 0.08861103084749622,
      "grad_norm": 4.40745735168457,
      "learning_rate": 9.114721541155868e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 20364312,
      "step": 1265
    },
    {
      "epoch": 0.08868107909322547,
      "grad_norm": 7.126221179962158,
      "learning_rate": 9.114021716287216e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 20380696,
      "step": 1266
    },
    {
      "epoch": 0.08875112733895471,
      "grad_norm": 4.306386947631836,
      "learning_rate": 9.113321891418564e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 20397080,
      "step": 1267
    },
    {
      "epoch": 0.08882117558468396,
      "grad_norm": 3.566943407058716,
      "learning_rate": 9.112622066549912e-05,
      "loss": 1.0463,
      "num_input_tokens_seen": 20413464,
      "step": 1268
    },
    {
      "epoch": 0.0888912238304132,
      "grad_norm": 3.975228786468506,
      "learning_rate": 9.111922241681262e-05,
      "loss": 1.2576,
      "num_input_tokens_seen": 20429848,
      "step": 1269
    },
    {
      "epoch": 0.08896127207614245,
      "grad_norm": 4.928854465484619,
      "learning_rate": 9.11122241681261e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 20446192,
      "step": 1270
    },
    {
      "epoch": 0.0890313203218717,
      "grad_norm": 4.288821697235107,
      "learning_rate": 9.110522591943958e-05,
      "loss": 1.2559,
      "num_input_tokens_seen": 20462576,
      "step": 1271
    },
    {
      "epoch": 0.08910136856760094,
      "grad_norm": 3.9346396923065186,
      "learning_rate": 9.109822767075307e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 20478520,
      "step": 1272
    },
    {
      "epoch": 0.08917141681333018,
      "grad_norm": 3.7976620197296143,
      "learning_rate": 9.109122942206655e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 20494408,
      "step": 1273
    },
    {
      "epoch": 0.08924146505905943,
      "grad_norm": 5.373577117919922,
      "learning_rate": 9.108423117338004e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 20510792,
      "step": 1274
    },
    {
      "epoch": 0.08931151330478868,
      "grad_norm": 4.248324394226074,
      "learning_rate": 9.107723292469353e-05,
      "loss": 1.3492,
      "num_input_tokens_seen": 20527064,
      "step": 1275
    },
    {
      "epoch": 0.08938156155051792,
      "grad_norm": 4.453672885894775,
      "learning_rate": 9.107023467600701e-05,
      "loss": 0.9763,
      "num_input_tokens_seen": 20543448,
      "step": 1276
    },
    {
      "epoch": 0.08945160979624717,
      "grad_norm": 4.8721184730529785,
      "learning_rate": 9.10632364273205e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 20559832,
      "step": 1277
    },
    {
      "epoch": 0.08952165804197641,
      "grad_norm": 5.0173540115356445,
      "learning_rate": 9.105623817863398e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 20576216,
      "step": 1278
    },
    {
      "epoch": 0.08959170628770566,
      "grad_norm": 5.00100040435791,
      "learning_rate": 9.104923992994747e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 20592600,
      "step": 1279
    },
    {
      "epoch": 0.0896617545334349,
      "grad_norm": 4.271099090576172,
      "learning_rate": 9.104224168126094e-05,
      "loss": 1.2307,
      "num_input_tokens_seen": 20608632,
      "step": 1280
    },
    {
      "epoch": 0.08973180277916415,
      "grad_norm": 4.246976852416992,
      "learning_rate": 9.103524343257443e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 20625016,
      "step": 1281
    },
    {
      "epoch": 0.0898018510248934,
      "grad_norm": 5.033923149108887,
      "learning_rate": 9.102824518388792e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 20641400,
      "step": 1282
    },
    {
      "epoch": 0.08987189927062264,
      "grad_norm": 4.4118571281433105,
      "learning_rate": 9.102124693520141e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 20657448,
      "step": 1283
    },
    {
      "epoch": 0.08994194751635189,
      "grad_norm": 4.150144577026367,
      "learning_rate": 9.10142486865149e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 20673080,
      "step": 1284
    },
    {
      "epoch": 0.09001199576208113,
      "grad_norm": 3.767683744430542,
      "learning_rate": 9.100725043782837e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 20689464,
      "step": 1285
    },
    {
      "epoch": 0.09008204400781038,
      "grad_norm": 4.816582202911377,
      "learning_rate": 9.100025218914186e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 20703896,
      "step": 1286
    },
    {
      "epoch": 0.09015209225353962,
      "grad_norm": 3.8913414478302,
      "learning_rate": 9.099325394045533e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 20720280,
      "step": 1287
    },
    {
      "epoch": 0.09022214049926887,
      "grad_norm": 4.305298328399658,
      "learning_rate": 9.098625569176882e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 20735944,
      "step": 1288
    },
    {
      "epoch": 0.09029218874499811,
      "grad_norm": 3.3120992183685303,
      "learning_rate": 9.097925744308233e-05,
      "loss": 0.8591,
      "num_input_tokens_seen": 20752128,
      "step": 1289
    },
    {
      "epoch": 0.09036223699072736,
      "grad_norm": 4.705013751983643,
      "learning_rate": 9.09722591943958e-05,
      "loss": 1.4579,
      "num_input_tokens_seen": 20768512,
      "step": 1290
    },
    {
      "epoch": 0.0904322852364566,
      "grad_norm": 5.08630895614624,
      "learning_rate": 9.096526094570929e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 20783976,
      "step": 1291
    },
    {
      "epoch": 0.09050233348218585,
      "grad_norm": 3.634686231613159,
      "learning_rate": 9.095826269702278e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 20800360,
      "step": 1292
    },
    {
      "epoch": 0.0905723817279151,
      "grad_norm": 4.220744609832764,
      "learning_rate": 9.095126444833625e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 20816744,
      "step": 1293
    },
    {
      "epoch": 0.09064242997364434,
      "grad_norm": 4.724472522735596,
      "learning_rate": 9.094426619964974e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 20833128,
      "step": 1294
    },
    {
      "epoch": 0.09071247821937359,
      "grad_norm": 3.9398090839385986,
      "learning_rate": 9.093726795096323e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 20849448,
      "step": 1295
    },
    {
      "epoch": 0.09078252646510283,
      "grad_norm": 4.260062217712402,
      "learning_rate": 9.093026970227672e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 20865832,
      "step": 1296
    },
    {
      "epoch": 0.09085257471083208,
      "grad_norm": 4.383310317993164,
      "learning_rate": 9.09232714535902e-05,
      "loss": 1.2542,
      "num_input_tokens_seen": 20881288,
      "step": 1297
    },
    {
      "epoch": 0.09092262295656132,
      "grad_norm": 4.479433059692383,
      "learning_rate": 9.091627320490368e-05,
      "loss": 0.9533,
      "num_input_tokens_seen": 20897328,
      "step": 1298
    },
    {
      "epoch": 0.09099267120229058,
      "grad_norm": 4.911858081817627,
      "learning_rate": 9.090927495621717e-05,
      "loss": 1.3399,
      "num_input_tokens_seen": 20913712,
      "step": 1299
    },
    {
      "epoch": 0.09106271944801983,
      "grad_norm": 4.015485763549805,
      "learning_rate": 9.090227670753065e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 20929984,
      "step": 1300
    },
    {
      "epoch": 0.09113276769374908,
      "grad_norm": 3.8690338134765625,
      "learning_rate": 9.089527845884413e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 20946368,
      "step": 1301
    },
    {
      "epoch": 0.09120281593947832,
      "grad_norm": 5.142012596130371,
      "learning_rate": 9.088828021015762e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 20962752,
      "step": 1302
    },
    {
      "epoch": 0.09127286418520757,
      "grad_norm": 3.954049587249756,
      "learning_rate": 9.088128196147111e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 20979136,
      "step": 1303
    },
    {
      "epoch": 0.09134291243093681,
      "grad_norm": 4.13312292098999,
      "learning_rate": 9.08742837127846e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 20995520,
      "step": 1304
    },
    {
      "epoch": 0.09141296067666606,
      "grad_norm": 4.24699592590332,
      "learning_rate": 9.086728546409808e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 21011904,
      "step": 1305
    },
    {
      "epoch": 0.0914830089223953,
      "grad_norm": 4.847048759460449,
      "learning_rate": 9.086028721541156e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 21027664,
      "step": 1306
    },
    {
      "epoch": 0.09155305716812455,
      "grad_norm": 4.573661804199219,
      "learning_rate": 9.085328896672504e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 21044048,
      "step": 1307
    },
    {
      "epoch": 0.0916231054138538,
      "grad_norm": 4.13530158996582,
      "learning_rate": 9.084629071803853e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 21060432,
      "step": 1308
    },
    {
      "epoch": 0.09169315365958304,
      "grad_norm": 4.017937183380127,
      "learning_rate": 9.083929246935203e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 21076816,
      "step": 1309
    },
    {
      "epoch": 0.09176320190531229,
      "grad_norm": 5.928586483001709,
      "learning_rate": 9.08322942206655e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 21093200,
      "step": 1310
    },
    {
      "epoch": 0.09183325015104153,
      "grad_norm": 4.2442169189453125,
      "learning_rate": 9.082529597197899e-05,
      "loss": 1.2794,
      "num_input_tokens_seen": 21109256,
      "step": 1311
    },
    {
      "epoch": 0.09190329839677078,
      "grad_norm": 4.891444683074951,
      "learning_rate": 9.081829772329247e-05,
      "loss": 1.1833,
      "num_input_tokens_seen": 21124848,
      "step": 1312
    },
    {
      "epoch": 0.09197334664250002,
      "grad_norm": 4.323850154876709,
      "learning_rate": 9.081129947460596e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 21141176,
      "step": 1313
    },
    {
      "epoch": 0.09204339488822927,
      "grad_norm": 4.239765644073486,
      "learning_rate": 9.080430122591943e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 21157240,
      "step": 1314
    },
    {
      "epoch": 0.09211344313395851,
      "grad_norm": 4.12881326675415,
      "learning_rate": 9.079730297723293e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 21173624,
      "step": 1315
    },
    {
      "epoch": 0.09218349137968776,
      "grad_norm": 4.238161087036133,
      "learning_rate": 9.079030472854642e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 21190008,
      "step": 1316
    },
    {
      "epoch": 0.092253539625417,
      "grad_norm": 4.124176502227783,
      "learning_rate": 9.07833064798599e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 21206392,
      "step": 1317
    },
    {
      "epoch": 0.09232358787114625,
      "grad_norm": 3.772136926651001,
      "learning_rate": 9.077630823117339e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 21222776,
      "step": 1318
    },
    {
      "epoch": 0.0923936361168755,
      "grad_norm": 4.628321170806885,
      "learning_rate": 9.076930998248687e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 21239160,
      "step": 1319
    },
    {
      "epoch": 0.09246368436260474,
      "grad_norm": 5.3034348487854,
      "learning_rate": 9.076231173380035e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 21255544,
      "step": 1320
    },
    {
      "epoch": 0.09253373260833399,
      "grad_norm": 3.6543760299682617,
      "learning_rate": 9.075531348511384e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 21271928,
      "step": 1321
    },
    {
      "epoch": 0.09260378085406323,
      "grad_norm": 4.1335062980651855,
      "learning_rate": 9.074831523642733e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 21288312,
      "step": 1322
    },
    {
      "epoch": 0.09267382909979248,
      "grad_norm": 3.6392204761505127,
      "learning_rate": 9.074131698774082e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 21304696,
      "step": 1323
    },
    {
      "epoch": 0.09274387734552172,
      "grad_norm": 4.035269737243652,
      "learning_rate": 9.073431873905429e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 21321080,
      "step": 1324
    },
    {
      "epoch": 0.09281392559125097,
      "grad_norm": 4.650269508361816,
      "learning_rate": 9.072732049036778e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 21337464,
      "step": 1325
    },
    {
      "epoch": 0.09288397383698022,
      "grad_norm": 5.850543022155762,
      "learning_rate": 9.072032224168127e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 21352968,
      "step": 1326
    },
    {
      "epoch": 0.09295402208270946,
      "grad_norm": 4.177901744842529,
      "learning_rate": 9.071332399299474e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 21368968,
      "step": 1327
    },
    {
      "epoch": 0.09302407032843871,
      "grad_norm": 4.582173824310303,
      "learning_rate": 9.070632574430823e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 21385352,
      "step": 1328
    },
    {
      "epoch": 0.09309411857416795,
      "grad_norm": 4.7911787033081055,
      "learning_rate": 9.069932749562173e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 21401144,
      "step": 1329
    },
    {
      "epoch": 0.0931641668198972,
      "grad_norm": 4.058457374572754,
      "learning_rate": 9.069232924693521e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 21416640,
      "step": 1330
    },
    {
      "epoch": 0.09323421506562644,
      "grad_norm": 4.972208499908447,
      "learning_rate": 9.06853309982487e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 21433024,
      "step": 1331
    },
    {
      "epoch": 0.09330426331135569,
      "grad_norm": 4.0875091552734375,
      "learning_rate": 9.067833274956217e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 21448888,
      "step": 1332
    },
    {
      "epoch": 0.09337431155708495,
      "grad_norm": 3.923112154006958,
      "learning_rate": 9.067133450087566e-05,
      "loss": 0.9824,
      "num_input_tokens_seen": 21465272,
      "step": 1333
    },
    {
      "epoch": 0.0934443598028142,
      "grad_norm": 4.067697525024414,
      "learning_rate": 9.066433625218914e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 21481656,
      "step": 1334
    },
    {
      "epoch": 0.09351440804854344,
      "grad_norm": 4.185417652130127,
      "learning_rate": 9.065733800350264e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 21498040,
      "step": 1335
    },
    {
      "epoch": 0.09358445629427269,
      "grad_norm": 7.31542444229126,
      "learning_rate": 9.065033975481613e-05,
      "loss": 1.4322,
      "num_input_tokens_seen": 21514088,
      "step": 1336
    },
    {
      "epoch": 0.09365450454000193,
      "grad_norm": 4.754745006561279,
      "learning_rate": 9.06433415061296e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 21530472,
      "step": 1337
    },
    {
      "epoch": 0.09372455278573118,
      "grad_norm": 5.81265926361084,
      "learning_rate": 9.063634325744309e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 21545728,
      "step": 1338
    },
    {
      "epoch": 0.09379460103146042,
      "grad_norm": 5.586238861083984,
      "learning_rate": 9.062934500875657e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 21562112,
      "step": 1339
    },
    {
      "epoch": 0.09386464927718967,
      "grad_norm": 4.096534729003906,
      "learning_rate": 9.062234676007005e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 21578496,
      "step": 1340
    },
    {
      "epoch": 0.09393469752291891,
      "grad_norm": 4.913814544677734,
      "learning_rate": 9.061534851138354e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 21594792,
      "step": 1341
    },
    {
      "epoch": 0.09400474576864816,
      "grad_norm": 3.8853912353515625,
      "learning_rate": 9.060835026269703e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 21611176,
      "step": 1342
    },
    {
      "epoch": 0.0940747940143774,
      "grad_norm": 4.187959671020508,
      "learning_rate": 9.060135201401052e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 21627560,
      "step": 1343
    },
    {
      "epoch": 0.09414484226010665,
      "grad_norm": 4.128627777099609,
      "learning_rate": 9.0594353765324e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 21643576,
      "step": 1344
    },
    {
      "epoch": 0.0942148905058359,
      "grad_norm": 4.7016825675964355,
      "learning_rate": 9.058735551663748e-05,
      "loss": 1.2425,
      "num_input_tokens_seen": 21658600,
      "step": 1345
    },
    {
      "epoch": 0.09428493875156514,
      "grad_norm": 3.970548391342163,
      "learning_rate": 9.058035726795097e-05,
      "loss": 1.0495,
      "num_input_tokens_seen": 21674264,
      "step": 1346
    },
    {
      "epoch": 0.09435498699729439,
      "grad_norm": 3.812196731567383,
      "learning_rate": 9.057335901926445e-05,
      "loss": 0.9558,
      "num_input_tokens_seen": 21690112,
      "step": 1347
    },
    {
      "epoch": 0.09442503524302363,
      "grad_norm": 3.6845176219940186,
      "learning_rate": 9.056636077057794e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 21705744,
      "step": 1348
    },
    {
      "epoch": 0.09449508348875288,
      "grad_norm": 4.119202136993408,
      "learning_rate": 9.055936252189142e-05,
      "loss": 1.0948,
      "num_input_tokens_seen": 21721776,
      "step": 1349
    },
    {
      "epoch": 0.09456513173448212,
      "grad_norm": 4.176985740661621,
      "learning_rate": 9.055236427320491e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 21737912,
      "step": 1350
    },
    {
      "epoch": 0.09463517998021137,
      "grad_norm": 4.057264804840088,
      "learning_rate": 9.054536602451839e-05,
      "loss": 1.1746,
      "num_input_tokens_seen": 21754296,
      "step": 1351
    },
    {
      "epoch": 0.09470522822594062,
      "grad_norm": 4.5631914138793945,
      "learning_rate": 9.053836777583188e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 21770680,
      "step": 1352
    },
    {
      "epoch": 0.09477527647166986,
      "grad_norm": 4.854849815368652,
      "learning_rate": 9.053136952714536e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 21787064,
      "step": 1353
    },
    {
      "epoch": 0.09484532471739911,
      "grad_norm": 5.326946258544922,
      "learning_rate": 9.052437127845884e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 21803448,
      "step": 1354
    },
    {
      "epoch": 0.09491537296312835,
      "grad_norm": 4.283742904663086,
      "learning_rate": 9.051737302977234e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 21819832,
      "step": 1355
    },
    {
      "epoch": 0.0949854212088576,
      "grad_norm": 4.165935039520264,
      "learning_rate": 9.051037478108582e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 21836216,
      "step": 1356
    },
    {
      "epoch": 0.09505546945458684,
      "grad_norm": 4.502480983734131,
      "learning_rate": 9.05033765323993e-05,
      "loss": 1.2854,
      "num_input_tokens_seen": 21852600,
      "step": 1357
    },
    {
      "epoch": 0.09512551770031609,
      "grad_norm": 4.185445308685303,
      "learning_rate": 9.04963782837128e-05,
      "loss": 1.2225,
      "num_input_tokens_seen": 21868984,
      "step": 1358
    },
    {
      "epoch": 0.09519556594604534,
      "grad_norm": 7.288909435272217,
      "learning_rate": 9.048938003502627e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 21884648,
      "step": 1359
    },
    {
      "epoch": 0.09526561419177458,
      "grad_norm": 4.038896560668945,
      "learning_rate": 9.048238178633976e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 21900704,
      "step": 1360
    },
    {
      "epoch": 0.09533566243750383,
      "grad_norm": 4.216241836547852,
      "learning_rate": 9.047538353765325e-05,
      "loss": 1.1379,
      "num_input_tokens_seen": 21916520,
      "step": 1361
    },
    {
      "epoch": 0.09540571068323307,
      "grad_norm": 4.2549147605896,
      "learning_rate": 9.046838528896673e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 21932904,
      "step": 1362
    },
    {
      "epoch": 0.09547575892896232,
      "grad_norm": 3.6919445991516113,
      "learning_rate": 9.046138704028022e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 21949288,
      "step": 1363
    },
    {
      "epoch": 0.09554580717469156,
      "grad_norm": 5.467876434326172,
      "learning_rate": 9.04543887915937e-05,
      "loss": 0.9735,
      "num_input_tokens_seen": 21965672,
      "step": 1364
    },
    {
      "epoch": 0.09561585542042081,
      "grad_norm": 4.036736011505127,
      "learning_rate": 9.044739054290719e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 21980792,
      "step": 1365
    },
    {
      "epoch": 0.09568590366615005,
      "grad_norm": 4.083346843719482,
      "learning_rate": 9.044039229422066e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 21996888,
      "step": 1366
    },
    {
      "epoch": 0.0957559519118793,
      "grad_norm": 3.553262948989868,
      "learning_rate": 9.043339404553415e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 22013160,
      "step": 1367
    },
    {
      "epoch": 0.09582600015760856,
      "grad_norm": 4.787721633911133,
      "learning_rate": 9.042639579684764e-05,
      "loss": 1.1444,
      "num_input_tokens_seen": 22029544,
      "step": 1368
    },
    {
      "epoch": 0.0958960484033378,
      "grad_norm": 3.8053700923919678,
      "learning_rate": 9.041939754816113e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 22045888,
      "step": 1369
    },
    {
      "epoch": 0.09596609664906705,
      "grad_norm": 3.7679660320281982,
      "learning_rate": 9.041239929947462e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 22062272,
      "step": 1370
    },
    {
      "epoch": 0.0960361448947963,
      "grad_norm": 5.086554527282715,
      "learning_rate": 9.040540105078809e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 22078080,
      "step": 1371
    },
    {
      "epoch": 0.09610619314052554,
      "grad_norm": 4.255527496337891,
      "learning_rate": 9.039840280210158e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 22093808,
      "step": 1372
    },
    {
      "epoch": 0.09617624138625479,
      "grad_norm": 6.081700325012207,
      "learning_rate": 9.039140455341507e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 22110192,
      "step": 1373
    },
    {
      "epoch": 0.09624628963198403,
      "grad_norm": 4.376565456390381,
      "learning_rate": 9.038440630472854e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 22126576,
      "step": 1374
    },
    {
      "epoch": 0.09631633787771328,
      "grad_norm": 4.051114559173584,
      "learning_rate": 9.037740805604205e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 22142768,
      "step": 1375
    },
    {
      "epoch": 0.09638638612344252,
      "grad_norm": 4.46164083480835,
      "learning_rate": 9.037040980735552e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 22158600,
      "step": 1376
    },
    {
      "epoch": 0.09645643436917177,
      "grad_norm": 4.242503643035889,
      "learning_rate": 9.036341155866901e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 22174984,
      "step": 1377
    },
    {
      "epoch": 0.09652648261490102,
      "grad_norm": 3.6338908672332764,
      "learning_rate": 9.035641330998248e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 22190880,
      "step": 1378
    },
    {
      "epoch": 0.09659653086063026,
      "grad_norm": 4.73402738571167,
      "learning_rate": 9.034941506129597e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 22206632,
      "step": 1379
    },
    {
      "epoch": 0.09666657910635951,
      "grad_norm": 4.450289726257324,
      "learning_rate": 9.034241681260946e-05,
      "loss": 1.0851,
      "num_input_tokens_seen": 22222896,
      "step": 1380
    },
    {
      "epoch": 0.09673662735208875,
      "grad_norm": 5.578179359436035,
      "learning_rate": 9.033541856392295e-05,
      "loss": 1.2856,
      "num_input_tokens_seen": 22238280,
      "step": 1381
    },
    {
      "epoch": 0.096806675597818,
      "grad_norm": 3.8745546340942383,
      "learning_rate": 9.032842031523644e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 22254664,
      "step": 1382
    },
    {
      "epoch": 0.09687672384354724,
      "grad_norm": 5.7268548011779785,
      "learning_rate": 9.032142206654991e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 22270000,
      "step": 1383
    },
    {
      "epoch": 0.09694677208927649,
      "grad_norm": 4.380898952484131,
      "learning_rate": 9.03144238178634e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 22286384,
      "step": 1384
    },
    {
      "epoch": 0.09701682033500574,
      "grad_norm": 5.762500762939453,
      "learning_rate": 9.030742556917689e-05,
      "loss": 1.2061,
      "num_input_tokens_seen": 22302272,
      "step": 1385
    },
    {
      "epoch": 0.09708686858073498,
      "grad_norm": 3.739488363265991,
      "learning_rate": 9.030042732049037e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 22318656,
      "step": 1386
    },
    {
      "epoch": 0.09715691682646423,
      "grad_norm": 4.584897994995117,
      "learning_rate": 9.029342907180385e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 22334704,
      "step": 1387
    },
    {
      "epoch": 0.09722696507219347,
      "grad_norm": 4.161139488220215,
      "learning_rate": 9.028643082311734e-05,
      "loss": 1.1638,
      "num_input_tokens_seen": 22349800,
      "step": 1388
    },
    {
      "epoch": 0.09729701331792272,
      "grad_norm": 4.115293979644775,
      "learning_rate": 9.027943257443083e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 22366184,
      "step": 1389
    },
    {
      "epoch": 0.09736706156365196,
      "grad_norm": 3.7355988025665283,
      "learning_rate": 9.027243432574432e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 22382568,
      "step": 1390
    },
    {
      "epoch": 0.09743710980938121,
      "grad_norm": 4.15507173538208,
      "learning_rate": 9.02654360770578e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 22398480,
      "step": 1391
    },
    {
      "epoch": 0.09750715805511045,
      "grad_norm": 3.770918607711792,
      "learning_rate": 9.025843782837128e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 22414864,
      "step": 1392
    },
    {
      "epoch": 0.0975772063008397,
      "grad_norm": 4.214321136474609,
      "learning_rate": 9.025143957968476e-05,
      "loss": 1.1738,
      "num_input_tokens_seen": 22429752,
      "step": 1393
    },
    {
      "epoch": 0.09764725454656895,
      "grad_norm": 3.9854986667633057,
      "learning_rate": 9.024444133099825e-05,
      "loss": 1.2832,
      "num_input_tokens_seen": 22446136,
      "step": 1394
    },
    {
      "epoch": 0.09771730279229819,
      "grad_norm": 4.996057510375977,
      "learning_rate": 9.023744308231174e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 22461160,
      "step": 1395
    },
    {
      "epoch": 0.09778735103802744,
      "grad_norm": 3.682765007019043,
      "learning_rate": 9.023044483362523e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 22477336,
      "step": 1396
    },
    {
      "epoch": 0.09785739928375668,
      "grad_norm": 4.367272853851318,
      "learning_rate": 9.022344658493871e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 22492952,
      "step": 1397
    },
    {
      "epoch": 0.09792744752948593,
      "grad_norm": 3.9716336727142334,
      "learning_rate": 9.021644833625219e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 22509336,
      "step": 1398
    },
    {
      "epoch": 0.09799749577521517,
      "grad_norm": 4.043631553649902,
      "learning_rate": 9.020945008756568e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 22525568,
      "step": 1399
    },
    {
      "epoch": 0.09806754402094442,
      "grad_norm": 4.343166351318359,
      "learning_rate": 9.020245183887917e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 22541328,
      "step": 1400
    },
    {
      "epoch": 0.09806754402094442,
      "eval_loss": 1.1561514139175415,
      "eval_runtime": 0.1977,
      "eval_samples_per_second": 5.058,
      "eval_steps_per_second": 5.058,
      "num_input_tokens_seen": 22541328,
      "step": 1400
    },
    {
      "epoch": 0.09813759226667366,
      "grad_norm": 4.709417819976807,
      "learning_rate": 9.019545359019265e-05,
      "loss": 1.1398,
      "num_input_tokens_seen": 22557304,
      "step": 1401
    },
    {
      "epoch": 0.09820764051240291,
      "grad_norm": 7.022638320922852,
      "learning_rate": 9.018845534150614e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 22573688,
      "step": 1402
    },
    {
      "epoch": 0.09827768875813217,
      "grad_norm": 3.7976694107055664,
      "learning_rate": 9.018145709281962e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 22589848,
      "step": 1403
    },
    {
      "epoch": 0.09834773700386142,
      "grad_norm": 3.70877742767334,
      "learning_rate": 9.01744588441331e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 22606232,
      "step": 1404
    },
    {
      "epoch": 0.09841778524959066,
      "grad_norm": 7.724960803985596,
      "learning_rate": 9.016746059544658e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 22621912,
      "step": 1405
    },
    {
      "epoch": 0.09848783349531991,
      "grad_norm": 3.9619522094726562,
      "learning_rate": 9.016046234676007e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 22638296,
      "step": 1406
    },
    {
      "epoch": 0.09855788174104915,
      "grad_norm": 3.8303041458129883,
      "learning_rate": 9.015346409807356e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 22654496,
      "step": 1407
    },
    {
      "epoch": 0.0986279299867784,
      "grad_norm": 4.029507637023926,
      "learning_rate": 9.014646584938705e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 22670544,
      "step": 1408
    },
    {
      "epoch": 0.09869797823250764,
      "grad_norm": 3.8487346172332764,
      "learning_rate": 9.013946760070054e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 22686592,
      "step": 1409
    },
    {
      "epoch": 0.09876802647823689,
      "grad_norm": 3.700751543045044,
      "learning_rate": 9.013246935201401e-05,
      "loss": 0.8698,
      "num_input_tokens_seen": 22702976,
      "step": 1410
    },
    {
      "epoch": 0.09883807472396614,
      "grad_norm": 3.686884641647339,
      "learning_rate": 9.01254711033275e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 22719360,
      "step": 1411
    },
    {
      "epoch": 0.09890812296969538,
      "grad_norm": 4.176409721374512,
      "learning_rate": 9.011847285464099e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 22735744,
      "step": 1412
    },
    {
      "epoch": 0.09897817121542463,
      "grad_norm": 4.331852912902832,
      "learning_rate": 9.011147460595446e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 22752128,
      "step": 1413
    },
    {
      "epoch": 0.09904821946115387,
      "grad_norm": 3.8534255027770996,
      "learning_rate": 9.010447635726795e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 22768512,
      "step": 1414
    },
    {
      "epoch": 0.09911826770688312,
      "grad_norm": 4.066548824310303,
      "learning_rate": 9.009747810858144e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 22784760,
      "step": 1415
    },
    {
      "epoch": 0.09918831595261236,
      "grad_norm": 4.076517581939697,
      "learning_rate": 9.009047985989493e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 22801144,
      "step": 1416
    },
    {
      "epoch": 0.09925836419834161,
      "grad_norm": 3.8858346939086914,
      "learning_rate": 9.008348161120842e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 22817320,
      "step": 1417
    },
    {
      "epoch": 0.09932841244407085,
      "grad_norm": 6.4605584144592285,
      "learning_rate": 9.007648336252189e-05,
      "loss": 1.2701,
      "num_input_tokens_seen": 22833704,
      "step": 1418
    },
    {
      "epoch": 0.0993984606898001,
      "grad_norm": 4.157481670379639,
      "learning_rate": 9.006948511383538e-05,
      "loss": 1.0169,
      "num_input_tokens_seen": 22850088,
      "step": 1419
    },
    {
      "epoch": 0.09946850893552935,
      "grad_norm": 3.725755214691162,
      "learning_rate": 9.006248686514886e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 22866472,
      "step": 1420
    },
    {
      "epoch": 0.09953855718125859,
      "grad_norm": 4.012838363647461,
      "learning_rate": 9.005548861646236e-05,
      "loss": 0.8425,
      "num_input_tokens_seen": 22882856,
      "step": 1421
    },
    {
      "epoch": 0.09960860542698784,
      "grad_norm": 3.8754239082336426,
      "learning_rate": 9.004849036777583e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 22899240,
      "step": 1422
    },
    {
      "epoch": 0.09967865367271708,
      "grad_norm": 3.90873384475708,
      "learning_rate": 9.004149211908932e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 22915160,
      "step": 1423
    },
    {
      "epoch": 0.09974870191844633,
      "grad_norm": 5.698948860168457,
      "learning_rate": 9.003449387040281e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 22930592,
      "step": 1424
    },
    {
      "epoch": 0.09981875016417557,
      "grad_norm": 4.103662014007568,
      "learning_rate": 9.002749562171629e-05,
      "loss": 1.2384,
      "num_input_tokens_seen": 22946976,
      "step": 1425
    },
    {
      "epoch": 0.09988879840990482,
      "grad_norm": 4.404048442840576,
      "learning_rate": 9.002049737302977e-05,
      "loss": 1.3855,
      "num_input_tokens_seen": 22963360,
      "step": 1426
    },
    {
      "epoch": 0.09995884665563406,
      "grad_norm": 4.043710708618164,
      "learning_rate": 9.001349912434326e-05,
      "loss": 1.2713,
      "num_input_tokens_seen": 22979544,
      "step": 1427
    },
    {
      "epoch": 0.10002889490136331,
      "grad_norm": 4.169802188873291,
      "learning_rate": 9.000650087565675e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 22995072,
      "step": 1428
    },
    {
      "epoch": 0.10009894314709256,
      "grad_norm": 4.010350227355957,
      "learning_rate": 8.999950262697024e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 23010904,
      "step": 1429
    },
    {
      "epoch": 0.1001689913928218,
      "grad_norm": 4.496591567993164,
      "learning_rate": 8.999250437828372e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 23027288,
      "step": 1430
    },
    {
      "epoch": 0.10023903963855105,
      "grad_norm": 4.2428765296936035,
      "learning_rate": 8.99855061295972e-05,
      "loss": 1.0258,
      "num_input_tokens_seen": 23043352,
      "step": 1431
    },
    {
      "epoch": 0.10030908788428029,
      "grad_norm": 4.083342552185059,
      "learning_rate": 8.997850788091068e-05,
      "loss": 1.227,
      "num_input_tokens_seen": 23059736,
      "step": 1432
    },
    {
      "epoch": 0.10037913613000954,
      "grad_norm": 3.860734462738037,
      "learning_rate": 8.997150963222417e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 23075400,
      "step": 1433
    },
    {
      "epoch": 0.10044918437573878,
      "grad_norm": 3.985151767730713,
      "learning_rate": 8.996451138353766e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 23091704,
      "step": 1434
    },
    {
      "epoch": 0.10051923262146803,
      "grad_norm": 4.039731502532959,
      "learning_rate": 8.995751313485114e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 23108088,
      "step": 1435
    },
    {
      "epoch": 0.10058928086719728,
      "grad_norm": 6.1780619621276855,
      "learning_rate": 8.995051488616463e-05,
      "loss": 1.0645,
      "num_input_tokens_seen": 23123128,
      "step": 1436
    },
    {
      "epoch": 0.10065932911292653,
      "grad_norm": 4.5783586502075195,
      "learning_rate": 8.994351663747811e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 23139168,
      "step": 1437
    },
    {
      "epoch": 0.10072937735865578,
      "grad_norm": 3.889927864074707,
      "learning_rate": 8.99365183887916e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 23154952,
      "step": 1438
    },
    {
      "epoch": 0.10079942560438503,
      "grad_norm": 3.927945852279663,
      "learning_rate": 8.992952014010509e-05,
      "loss": 1.2428,
      "num_input_tokens_seen": 23170688,
      "step": 1439
    },
    {
      "epoch": 0.10086947385011427,
      "grad_norm": 3.8991434574127197,
      "learning_rate": 8.992252189141856e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 23186432,
      "step": 1440
    },
    {
      "epoch": 0.10093952209584352,
      "grad_norm": 3.6479310989379883,
      "learning_rate": 8.991552364273206e-05,
      "loss": 0.9656,
      "num_input_tokens_seen": 23202816,
      "step": 1441
    },
    {
      "epoch": 0.10100957034157276,
      "grad_norm": 4.637960910797119,
      "learning_rate": 8.990852539404554e-05,
      "loss": 1.2853,
      "num_input_tokens_seen": 23218304,
      "step": 1442
    },
    {
      "epoch": 0.10107961858730201,
      "grad_norm": 4.000091552734375,
      "learning_rate": 8.990152714535903e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 23234688,
      "step": 1443
    },
    {
      "epoch": 0.10114966683303125,
      "grad_norm": 4.959738731384277,
      "learning_rate": 8.989452889667251e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 23250656,
      "step": 1444
    },
    {
      "epoch": 0.1012197150787605,
      "grad_norm": 3.9251675605773926,
      "learning_rate": 8.988753064798599e-05,
      "loss": 0.9219,
      "num_input_tokens_seen": 23266984,
      "step": 1445
    },
    {
      "epoch": 0.10128976332448975,
      "grad_norm": 4.28665828704834,
      "learning_rate": 8.988053239929948e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 23283368,
      "step": 1446
    },
    {
      "epoch": 0.10135981157021899,
      "grad_norm": 4.421731472015381,
      "learning_rate": 8.987353415061297e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 23298728,
      "step": 1447
    },
    {
      "epoch": 0.10142985981594824,
      "grad_norm": 5.080065727233887,
      "learning_rate": 8.986653590192646e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 23315112,
      "step": 1448
    },
    {
      "epoch": 0.10149990806167748,
      "grad_norm": 5.618803977966309,
      "learning_rate": 8.985953765323993e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 23329864,
      "step": 1449
    },
    {
      "epoch": 0.10156995630740673,
      "grad_norm": 3.756836414337158,
      "learning_rate": 8.985253940455342e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 23345672,
      "step": 1450
    },
    {
      "epoch": 0.10164000455313597,
      "grad_norm": 4.461424827575684,
      "learning_rate": 8.984554115586691e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 23362056,
      "step": 1451
    },
    {
      "epoch": 0.10171005279886522,
      "grad_norm": 5.267919063568115,
      "learning_rate": 8.983854290718038e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 23378440,
      "step": 1452
    },
    {
      "epoch": 0.10178010104459446,
      "grad_norm": 5.513551235198975,
      "learning_rate": 8.983154465849387e-05,
      "loss": 1.2405,
      "num_input_tokens_seen": 23394824,
      "step": 1453
    },
    {
      "epoch": 0.10185014929032371,
      "grad_norm": 4.46366548538208,
      "learning_rate": 8.982454640980736e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 23410568,
      "step": 1454
    },
    {
      "epoch": 0.10192019753605296,
      "grad_norm": 5.066692352294922,
      "learning_rate": 8.981754816112085e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 23426952,
      "step": 1455
    },
    {
      "epoch": 0.1019902457817822,
      "grad_norm": 3.980743169784546,
      "learning_rate": 8.981054991243434e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 23443256,
      "step": 1456
    },
    {
      "epoch": 0.10206029402751145,
      "grad_norm": 4.088611125946045,
      "learning_rate": 8.980355166374781e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 23459640,
      "step": 1457
    },
    {
      "epoch": 0.10213034227324069,
      "grad_norm": 3.9585626125335693,
      "learning_rate": 8.97965534150613e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 23475176,
      "step": 1458
    },
    {
      "epoch": 0.10220039051896994,
      "grad_norm": 3.7923290729522705,
      "learning_rate": 8.978955516637478e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 23490704,
      "step": 1459
    },
    {
      "epoch": 0.10227043876469918,
      "grad_norm": 3.9089629650115967,
      "learning_rate": 8.978255691768826e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 23506168,
      "step": 1460
    },
    {
      "epoch": 0.10234048701042843,
      "grad_norm": 6.2259039878845215,
      "learning_rate": 8.977555866900175e-05,
      "loss": 1.2854,
      "num_input_tokens_seen": 23522552,
      "step": 1461
    },
    {
      "epoch": 0.10241053525615768,
      "grad_norm": 4.071867942810059,
      "learning_rate": 8.976856042031524e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 23538936,
      "step": 1462
    },
    {
      "epoch": 0.10248058350188692,
      "grad_norm": 4.587897777557373,
      "learning_rate": 8.976156217162873e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 23554536,
      "step": 1463
    },
    {
      "epoch": 0.10255063174761617,
      "grad_norm": 3.944937229156494,
      "learning_rate": 8.97545639229422e-05,
      "loss": 1.1503,
      "num_input_tokens_seen": 23570888,
      "step": 1464
    },
    {
      "epoch": 0.10262067999334541,
      "grad_norm": 3.7418766021728516,
      "learning_rate": 8.97475656742557e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 23587272,
      "step": 1465
    },
    {
      "epoch": 0.10269072823907466,
      "grad_norm": 3.9055676460266113,
      "learning_rate": 8.974056742556918e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 23603640,
      "step": 1466
    },
    {
      "epoch": 0.1027607764848039,
      "grad_norm": 3.9338066577911377,
      "learning_rate": 8.973356917688267e-05,
      "loss": 1.2389,
      "num_input_tokens_seen": 23620024,
      "step": 1467
    },
    {
      "epoch": 0.10283082473053315,
      "grad_norm": 4.024602890014648,
      "learning_rate": 8.972657092819616e-05,
      "loss": 1.1358,
      "num_input_tokens_seen": 23636408,
      "step": 1468
    },
    {
      "epoch": 0.1029008729762624,
      "grad_norm": 4.09812068939209,
      "learning_rate": 8.971957267950963e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 23652480,
      "step": 1469
    },
    {
      "epoch": 0.10297092122199164,
      "grad_norm": 4.7382025718688965,
      "learning_rate": 8.971257443082312e-05,
      "loss": 1.4112,
      "num_input_tokens_seen": 23668424,
      "step": 1470
    },
    {
      "epoch": 0.10304096946772089,
      "grad_norm": 4.518669605255127,
      "learning_rate": 8.970557618213661e-05,
      "loss": 1.3466,
      "num_input_tokens_seen": 23684808,
      "step": 1471
    },
    {
      "epoch": 0.10311101771345015,
      "grad_norm": 4.023036003112793,
      "learning_rate": 8.969857793345009e-05,
      "loss": 1.0246,
      "num_input_tokens_seen": 23701192,
      "step": 1472
    },
    {
      "epoch": 0.10318106595917939,
      "grad_norm": 4.6244215965271,
      "learning_rate": 8.969157968476358e-05,
      "loss": 1.2391,
      "num_input_tokens_seen": 23717576,
      "step": 1473
    },
    {
      "epoch": 0.10325111420490864,
      "grad_norm": 4.517683506011963,
      "learning_rate": 8.968458143607706e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 23733960,
      "step": 1474
    },
    {
      "epoch": 0.10332116245063788,
      "grad_norm": 4.048764705657959,
      "learning_rate": 8.967758318739055e-05,
      "loss": 1.0453,
      "num_input_tokens_seen": 23750344,
      "step": 1475
    },
    {
      "epoch": 0.10339121069636713,
      "grad_norm": 4.248376369476318,
      "learning_rate": 8.967058493870403e-05,
      "loss": 1.176,
      "num_input_tokens_seen": 23766160,
      "step": 1476
    },
    {
      "epoch": 0.10346125894209637,
      "grad_norm": 3.780548095703125,
      "learning_rate": 8.966358669001752e-05,
      "loss": 0.9048,
      "num_input_tokens_seen": 23782544,
      "step": 1477
    },
    {
      "epoch": 0.10353130718782562,
      "grad_norm": 4.26375675201416,
      "learning_rate": 8.9656588441331e-05,
      "loss": 0.8651,
      "num_input_tokens_seen": 23798928,
      "step": 1478
    },
    {
      "epoch": 0.10360135543355486,
      "grad_norm": 3.9202687740325928,
      "learning_rate": 8.964959019264448e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 23815312,
      "step": 1479
    },
    {
      "epoch": 0.10367140367928411,
      "grad_norm": 3.983797788619995,
      "learning_rate": 8.964259194395797e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 23831696,
      "step": 1480
    },
    {
      "epoch": 0.10374145192501336,
      "grad_norm": 4.471195220947266,
      "learning_rate": 8.963559369527146e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 23847768,
      "step": 1481
    },
    {
      "epoch": 0.1038115001707426,
      "grad_norm": 3.560317039489746,
      "learning_rate": 8.962859544658495e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 23864152,
      "step": 1482
    },
    {
      "epoch": 0.10388154841647185,
      "grad_norm": 3.8699846267700195,
      "learning_rate": 8.962159719789843e-05,
      "loss": 1.1438,
      "num_input_tokens_seen": 23880536,
      "step": 1483
    },
    {
      "epoch": 0.10395159666220109,
      "grad_norm": 4.547862529754639,
      "learning_rate": 8.961459894921191e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 23896704,
      "step": 1484
    },
    {
      "epoch": 0.10402164490793034,
      "grad_norm": 4.669456481933594,
      "learning_rate": 8.96076007005254e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 23913088,
      "step": 1485
    },
    {
      "epoch": 0.10409169315365958,
      "grad_norm": 4.346285343170166,
      "learning_rate": 8.960060245183887e-05,
      "loss": 1.2677,
      "num_input_tokens_seen": 23929472,
      "step": 1486
    },
    {
      "epoch": 0.10416174139938883,
      "grad_norm": 6.5028581619262695,
      "learning_rate": 8.959360420315236e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 23945216,
      "step": 1487
    },
    {
      "epoch": 0.10423178964511808,
      "grad_norm": 3.935488224029541,
      "learning_rate": 8.958660595446586e-05,
      "loss": 1.2657,
      "num_input_tokens_seen": 23961600,
      "step": 1488
    },
    {
      "epoch": 0.10430183789084732,
      "grad_norm": 3.772397518157959,
      "learning_rate": 8.957960770577934e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 23977984,
      "step": 1489
    },
    {
      "epoch": 0.10437188613657657,
      "grad_norm": 4.508286476135254,
      "learning_rate": 8.957260945709283e-05,
      "loss": 1.2694,
      "num_input_tokens_seen": 23993752,
      "step": 1490
    },
    {
      "epoch": 0.10444193438230581,
      "grad_norm": 4.667380332946777,
      "learning_rate": 8.95656112084063e-05,
      "loss": 1.2837,
      "num_input_tokens_seen": 24009832,
      "step": 1491
    },
    {
      "epoch": 0.10451198262803506,
      "grad_norm": 7.675503730773926,
      "learning_rate": 8.955861295971979e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 24025784,
      "step": 1492
    },
    {
      "epoch": 0.1045820308737643,
      "grad_norm": 4.427548408508301,
      "learning_rate": 8.955161471103328e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 24041568,
      "step": 1493
    },
    {
      "epoch": 0.10465207911949355,
      "grad_norm": 3.9065396785736084,
      "learning_rate": 8.954461646234677e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 24057952,
      "step": 1494
    },
    {
      "epoch": 0.1047221273652228,
      "grad_norm": 4.052605628967285,
      "learning_rate": 8.953761821366026e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 24074336,
      "step": 1495
    },
    {
      "epoch": 0.10479217561095204,
      "grad_norm": 3.758476734161377,
      "learning_rate": 8.953061996497373e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 24090720,
      "step": 1496
    },
    {
      "epoch": 0.10486222385668129,
      "grad_norm": 4.4470014572143555,
      "learning_rate": 8.952362171628722e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 24107024,
      "step": 1497
    },
    {
      "epoch": 0.10493227210241053,
      "grad_norm": 4.222001075744629,
      "learning_rate": 8.951662346760071e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 24123408,
      "step": 1498
    },
    {
      "epoch": 0.10500232034813978,
      "grad_norm": 4.72997522354126,
      "learning_rate": 8.950962521891418e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 24137672,
      "step": 1499
    },
    {
      "epoch": 0.10507236859386902,
      "grad_norm": 4.342312812805176,
      "learning_rate": 8.950262697022767e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 24153248,
      "step": 1500
    },
    {
      "epoch": 0.10514241683959827,
      "grad_norm": 4.723706245422363,
      "learning_rate": 8.949562872154116e-05,
      "loss": 1.1075,
      "num_input_tokens_seen": 24169240,
      "step": 1501
    },
    {
      "epoch": 0.10521246508532751,
      "grad_norm": 4.244345188140869,
      "learning_rate": 8.948863047285465e-05,
      "loss": 1.1839,
      "num_input_tokens_seen": 24184608,
      "step": 1502
    },
    {
      "epoch": 0.10528251333105676,
      "grad_norm": 3.6271615028381348,
      "learning_rate": 8.948163222416812e-05,
      "loss": 1.0755,
      "num_input_tokens_seen": 24200992,
      "step": 1503
    },
    {
      "epoch": 0.105352561576786,
      "grad_norm": 3.858696937561035,
      "learning_rate": 8.947463397548161e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 24217376,
      "step": 1504
    },
    {
      "epoch": 0.10542260982251525,
      "grad_norm": 7.14077091217041,
      "learning_rate": 8.94676357267951e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 24232368,
      "step": 1505
    },
    {
      "epoch": 0.10549265806824451,
      "grad_norm": 4.203495979309082,
      "learning_rate": 8.946063747810858e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 24248520,
      "step": 1506
    },
    {
      "epoch": 0.10556270631397376,
      "grad_norm": 4.344188213348389,
      "learning_rate": 8.945363922942207e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 24264824,
      "step": 1507
    },
    {
      "epoch": 0.105632754559703,
      "grad_norm": 6.156280517578125,
      "learning_rate": 8.944664098073557e-05,
      "loss": 1.3046,
      "num_input_tokens_seen": 24281208,
      "step": 1508
    },
    {
      "epoch": 0.10570280280543225,
      "grad_norm": 4.687212944030762,
      "learning_rate": 8.943964273204904e-05,
      "loss": 1.1898,
      "num_input_tokens_seen": 24297592,
      "step": 1509
    },
    {
      "epoch": 0.10577285105116149,
      "grad_norm": 3.9128546714782715,
      "learning_rate": 8.943264448336253e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 24313976,
      "step": 1510
    },
    {
      "epoch": 0.10584289929689074,
      "grad_norm": 5.766979694366455,
      "learning_rate": 8.9425646234676e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 24330296,
      "step": 1511
    },
    {
      "epoch": 0.10591294754261998,
      "grad_norm": 3.9610238075256348,
      "learning_rate": 8.94186479859895e-05,
      "loss": 1.279,
      "num_input_tokens_seen": 24346680,
      "step": 1512
    },
    {
      "epoch": 0.10598299578834923,
      "grad_norm": 4.262688636779785,
      "learning_rate": 8.941164973730297e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 24362408,
      "step": 1513
    },
    {
      "epoch": 0.10605304403407848,
      "grad_norm": 5.190121173858643,
      "learning_rate": 8.940465148861647e-05,
      "loss": 1.0615,
      "num_input_tokens_seen": 24378248,
      "step": 1514
    },
    {
      "epoch": 0.10612309227980772,
      "grad_norm": 4.5859479904174805,
      "learning_rate": 8.939765323992996e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 24394632,
      "step": 1515
    },
    {
      "epoch": 0.10619314052553697,
      "grad_norm": 4.021294593811035,
      "learning_rate": 8.939065499124344e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 24411016,
      "step": 1516
    },
    {
      "epoch": 0.10626318877126621,
      "grad_norm": 4.296265602111816,
      "learning_rate": 8.938365674255692e-05,
      "loss": 1.1753,
      "num_input_tokens_seen": 24426792,
      "step": 1517
    },
    {
      "epoch": 0.10633323701699546,
      "grad_norm": 3.4397289752960205,
      "learning_rate": 8.93766584938704e-05,
      "loss": 0.8159,
      "num_input_tokens_seen": 24443176,
      "step": 1518
    },
    {
      "epoch": 0.1064032852627247,
      "grad_norm": 4.009952545166016,
      "learning_rate": 8.936966024518389e-05,
      "loss": 1.0728,
      "num_input_tokens_seen": 24459384,
      "step": 1519
    },
    {
      "epoch": 0.10647333350845395,
      "grad_norm": 4.786280632019043,
      "learning_rate": 8.936266199649738e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 24474904,
      "step": 1520
    },
    {
      "epoch": 0.1065433817541832,
      "grad_norm": 3.869297981262207,
      "learning_rate": 8.935566374781087e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 24490456,
      "step": 1521
    },
    {
      "epoch": 0.10661342999991244,
      "grad_norm": 3.995553731918335,
      "learning_rate": 8.934866549912435e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 24506840,
      "step": 1522
    },
    {
      "epoch": 0.10668347824564169,
      "grad_norm": 4.195018291473389,
      "learning_rate": 8.934166725043783e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 24522440,
      "step": 1523
    },
    {
      "epoch": 0.10675352649137093,
      "grad_norm": 4.0432515144348145,
      "learning_rate": 8.933466900175132e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 24538824,
      "step": 1524
    },
    {
      "epoch": 0.10682357473710018,
      "grad_norm": 5.120638847351074,
      "learning_rate": 8.93276707530648e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 24555208,
      "step": 1525
    },
    {
      "epoch": 0.10689362298282942,
      "grad_norm": 3.925096035003662,
      "learning_rate": 8.932067250437828e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 24571544,
      "step": 1526
    },
    {
      "epoch": 0.10696367122855867,
      "grad_norm": 3.930328130722046,
      "learning_rate": 8.931367425569177e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 24587736,
      "step": 1527
    },
    {
      "epoch": 0.10703371947428791,
      "grad_norm": 3.7056055068969727,
      "learning_rate": 8.930667600700526e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 24604120,
      "step": 1528
    },
    {
      "epoch": 0.10710376772001716,
      "grad_norm": 5.945568561553955,
      "learning_rate": 8.929967775831875e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 24620504,
      "step": 1529
    },
    {
      "epoch": 0.1071738159657464,
      "grad_norm": 3.7765939235687256,
      "learning_rate": 8.929267950963222e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 24635440,
      "step": 1530
    },
    {
      "epoch": 0.10724386421147565,
      "grad_norm": 4.229284286499023,
      "learning_rate": 8.928568126094571e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 24651824,
      "step": 1531
    },
    {
      "epoch": 0.1073139124572049,
      "grad_norm": 4.198834419250488,
      "learning_rate": 8.92786830122592e-05,
      "loss": 1.118,
      "num_input_tokens_seen": 24668208,
      "step": 1532
    },
    {
      "epoch": 0.10738396070293414,
      "grad_norm": 8.091620445251465,
      "learning_rate": 8.927168476357267e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 24684248,
      "step": 1533
    },
    {
      "epoch": 0.10745400894866339,
      "grad_norm": 4.091879844665527,
      "learning_rate": 8.926468651488618e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 24700632,
      "step": 1534
    },
    {
      "epoch": 0.10752405719439263,
      "grad_norm": 3.90326189994812,
      "learning_rate": 8.925768826619966e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 24717016,
      "step": 1535
    },
    {
      "epoch": 0.10759410544012188,
      "grad_norm": 4.097111225128174,
      "learning_rate": 8.925069001751314e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 24732776,
      "step": 1536
    },
    {
      "epoch": 0.10766415368585112,
      "grad_norm": 3.5537095069885254,
      "learning_rate": 8.924369176882663e-05,
      "loss": 0.989,
      "num_input_tokens_seen": 24749064,
      "step": 1537
    },
    {
      "epoch": 0.10773420193158037,
      "grad_norm": 4.3086256980896,
      "learning_rate": 8.92366935201401e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 24765448,
      "step": 1538
    },
    {
      "epoch": 0.10780425017730962,
      "grad_norm": 4.177425861358643,
      "learning_rate": 8.922969527145359e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 24780816,
      "step": 1539
    },
    {
      "epoch": 0.10787429842303886,
      "grad_norm": 3.6013338565826416,
      "learning_rate": 8.922269702276708e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 24796600,
      "step": 1540
    },
    {
      "epoch": 0.10794434666876812,
      "grad_norm": 4.05686092376709,
      "learning_rate": 8.921569877408057e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 24812984,
      "step": 1541
    },
    {
      "epoch": 0.10801439491449737,
      "grad_norm": 4.245424747467041,
      "learning_rate": 8.920870052539406e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 24829368,
      "step": 1542
    },
    {
      "epoch": 0.10808444316022661,
      "grad_norm": 3.9563350677490234,
      "learning_rate": 8.920170227670753e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 24845752,
      "step": 1543
    },
    {
      "epoch": 0.10815449140595586,
      "grad_norm": 4.209373474121094,
      "learning_rate": 8.919470402802102e-05,
      "loss": 1.2394,
      "num_input_tokens_seen": 24862136,
      "step": 1544
    },
    {
      "epoch": 0.1082245396516851,
      "grad_norm": 3.6590163707733154,
      "learning_rate": 8.91877057793345e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 24878520,
      "step": 1545
    },
    {
      "epoch": 0.10829458789741435,
      "grad_norm": 3.937568187713623,
      "learning_rate": 8.918070753064799e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 24894696,
      "step": 1546
    },
    {
      "epoch": 0.1083646361431436,
      "grad_norm": 3.948453426361084,
      "learning_rate": 8.917370928196147e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 24910208,
      "step": 1547
    },
    {
      "epoch": 0.10843468438887284,
      "grad_norm": 3.61549711227417,
      "learning_rate": 8.916671103327496e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 24926592,
      "step": 1548
    },
    {
      "epoch": 0.10850473263460209,
      "grad_norm": 4.091664791107178,
      "learning_rate": 8.915971278458845e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 24942976,
      "step": 1549
    },
    {
      "epoch": 0.10857478088033133,
      "grad_norm": 5.494830131530762,
      "learning_rate": 8.915271453590193e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 24957984,
      "step": 1550
    },
    {
      "epoch": 0.10864482912606058,
      "grad_norm": 4.546476364135742,
      "learning_rate": 8.914571628721541e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 24974368,
      "step": 1551
    },
    {
      "epoch": 0.10871487737178982,
      "grad_norm": 3.775996446609497,
      "learning_rate": 8.91387180385289e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 24990200,
      "step": 1552
    },
    {
      "epoch": 0.10878492561751907,
      "grad_norm": 3.9989728927612305,
      "learning_rate": 8.913171978984238e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 25006584,
      "step": 1553
    },
    {
      "epoch": 0.10885497386324831,
      "grad_norm": 4.417224884033203,
      "learning_rate": 8.912472154115588e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 25022464,
      "step": 1554
    },
    {
      "epoch": 0.10892502210897756,
      "grad_norm": 4.604903697967529,
      "learning_rate": 8.911772329246936e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 25038848,
      "step": 1555
    },
    {
      "epoch": 0.1089950703547068,
      "grad_norm": 4.823176860809326,
      "learning_rate": 8.911072504378284e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 25053776,
      "step": 1556
    },
    {
      "epoch": 0.10906511860043605,
      "grad_norm": 3.6778531074523926,
      "learning_rate": 8.910372679509632e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 25069872,
      "step": 1557
    },
    {
      "epoch": 0.1091351668461653,
      "grad_norm": 4.344213485717773,
      "learning_rate": 8.909672854640981e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 25086256,
      "step": 1558
    },
    {
      "epoch": 0.10920521509189454,
      "grad_norm": 4.592464923858643,
      "learning_rate": 8.90897302977233e-05,
      "loss": 1.502,
      "num_input_tokens_seen": 25102640,
      "step": 1559
    },
    {
      "epoch": 0.10927526333762379,
      "grad_norm": 4.103248119354248,
      "learning_rate": 8.908273204903678e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 25118328,
      "step": 1560
    },
    {
      "epoch": 0.10934531158335303,
      "grad_norm": 4.637456893920898,
      "learning_rate": 8.907573380035027e-05,
      "loss": 1.3611,
      "num_input_tokens_seen": 25134712,
      "step": 1561
    },
    {
      "epoch": 0.10941535982908228,
      "grad_norm": 4.4709930419921875,
      "learning_rate": 8.906873555166376e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 25149304,
      "step": 1562
    },
    {
      "epoch": 0.10948540807481152,
      "grad_norm": 4.154660701751709,
      "learning_rate": 8.906173730297724e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 25165360,
      "step": 1563
    },
    {
      "epoch": 0.10955545632054077,
      "grad_norm": 4.1212334632873535,
      "learning_rate": 8.905473905429073e-05,
      "loss": 1.2015,
      "num_input_tokens_seen": 25181744,
      "step": 1564
    },
    {
      "epoch": 0.10962550456627002,
      "grad_norm": 3.8060882091522217,
      "learning_rate": 8.90477408056042e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 25197800,
      "step": 1565
    },
    {
      "epoch": 0.10969555281199926,
      "grad_norm": 3.4948956966400146,
      "learning_rate": 8.904074255691769e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 25214008,
      "step": 1566
    },
    {
      "epoch": 0.1097656010577285,
      "grad_norm": 4.181606292724609,
      "learning_rate": 8.903374430823118e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 25229496,
      "step": 1567
    },
    {
      "epoch": 0.10983564930345775,
      "grad_norm": 4.206098556518555,
      "learning_rate": 8.902674605954467e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 25244864,
      "step": 1568
    },
    {
      "epoch": 0.109905697549187,
      "grad_norm": 3.797475576400757,
      "learning_rate": 8.901974781085815e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 25261248,
      "step": 1569
    },
    {
      "epoch": 0.10997574579491624,
      "grad_norm": 4.131814479827881,
      "learning_rate": 8.901274956217163e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 25277632,
      "step": 1570
    },
    {
      "epoch": 0.11004579404064549,
      "grad_norm": 3.9447309970855713,
      "learning_rate": 8.900575131348512e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 25294016,
      "step": 1571
    },
    {
      "epoch": 0.11011584228637473,
      "grad_norm": 3.916949510574341,
      "learning_rate": 8.89987530647986e-05,
      "loss": 1.1793,
      "num_input_tokens_seen": 25309912,
      "step": 1572
    },
    {
      "epoch": 0.11018589053210398,
      "grad_norm": 3.7132885456085205,
      "learning_rate": 8.899175481611208e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 25326296,
      "step": 1573
    },
    {
      "epoch": 0.11025593877783323,
      "grad_norm": 4.5842390060424805,
      "learning_rate": 8.898475656742558e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 25342328,
      "step": 1574
    },
    {
      "epoch": 0.11032598702356247,
      "grad_norm": 3.578962802886963,
      "learning_rate": 8.897775831873906e-05,
      "loss": 1.0599,
      "num_input_tokens_seen": 25357640,
      "step": 1575
    },
    {
      "epoch": 0.11039603526929173,
      "grad_norm": 3.5823471546173096,
      "learning_rate": 8.897076007005255e-05,
      "loss": 0.9519,
      "num_input_tokens_seen": 25373424,
      "step": 1576
    },
    {
      "epoch": 0.11046608351502098,
      "grad_norm": 3.721482515335083,
      "learning_rate": 8.896376182136602e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 25389808,
      "step": 1577
    },
    {
      "epoch": 0.11053613176075022,
      "grad_norm": 4.874295711517334,
      "learning_rate": 8.895676357267951e-05,
      "loss": 1.3507,
      "num_input_tokens_seen": 25406192,
      "step": 1578
    },
    {
      "epoch": 0.11060618000647947,
      "grad_norm": 3.8547258377075195,
      "learning_rate": 8.8949765323993e-05,
      "loss": 0.9444,
      "num_input_tokens_seen": 25421632,
      "step": 1579
    },
    {
      "epoch": 0.11067622825220871,
      "grad_norm": 4.847586631774902,
      "learning_rate": 8.894276707530649e-05,
      "loss": 1.0526,
      "num_input_tokens_seen": 25438016,
      "step": 1580
    },
    {
      "epoch": 0.11074627649793796,
      "grad_norm": 3.950594425201416,
      "learning_rate": 8.893576882661998e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 25454400,
      "step": 1581
    },
    {
      "epoch": 0.1108163247436672,
      "grad_norm": 3.7372758388519287,
      "learning_rate": 8.892877057793345e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 25470304,
      "step": 1582
    },
    {
      "epoch": 0.11088637298939645,
      "grad_norm": 3.8695788383483887,
      "learning_rate": 8.892177232924694e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 25486688,
      "step": 1583
    },
    {
      "epoch": 0.1109564212351257,
      "grad_norm": 4.623810768127441,
      "learning_rate": 8.891477408056042e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 25503072,
      "step": 1584
    },
    {
      "epoch": 0.11102646948085494,
      "grad_norm": 4.03538179397583,
      "learning_rate": 8.89077758318739e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 25519008,
      "step": 1585
    },
    {
      "epoch": 0.11109651772658419,
      "grad_norm": 7.486603736877441,
      "learning_rate": 8.890077758318739e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 25533808,
      "step": 1586
    },
    {
      "epoch": 0.11116656597231343,
      "grad_norm": 4.660414218902588,
      "learning_rate": 8.889377933450088e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 25549784,
      "step": 1587
    },
    {
      "epoch": 0.11123661421804268,
      "grad_norm": 3.9375548362731934,
      "learning_rate": 8.888678108581437e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 25566168,
      "step": 1588
    },
    {
      "epoch": 0.11130666246377192,
      "grad_norm": 4.275035858154297,
      "learning_rate": 8.887978283712786e-05,
      "loss": 1.1802,
      "num_input_tokens_seen": 25582552,
      "step": 1589
    },
    {
      "epoch": 0.11137671070950117,
      "grad_norm": 4.799124240875244,
      "learning_rate": 8.887278458844133e-05,
      "loss": 1.2702,
      "num_input_tokens_seen": 25598936,
      "step": 1590
    },
    {
      "epoch": 0.11144675895523042,
      "grad_norm": 4.143614768981934,
      "learning_rate": 8.886578633975482e-05,
      "loss": 1.1797,
      "num_input_tokens_seen": 25615320,
      "step": 1591
    },
    {
      "epoch": 0.11151680720095966,
      "grad_norm": 4.490556716918945,
      "learning_rate": 8.88587880910683e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 25630624,
      "step": 1592
    },
    {
      "epoch": 0.1115868554466889,
      "grad_norm": 6.010688781738281,
      "learning_rate": 8.885178984238179e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 25646048,
      "step": 1593
    },
    {
      "epoch": 0.11165690369241815,
      "grad_norm": 3.7447726726531982,
      "learning_rate": 8.884479159369527e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 25661528,
      "step": 1594
    },
    {
      "epoch": 0.1117269519381474,
      "grad_norm": 4.77920389175415,
      "learning_rate": 8.883779334500876e-05,
      "loss": 1.1158,
      "num_input_tokens_seen": 25677912,
      "step": 1595
    },
    {
      "epoch": 0.11179700018387664,
      "grad_norm": 3.9812231063842773,
      "learning_rate": 8.883079509632225e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 25694296,
      "step": 1596
    },
    {
      "epoch": 0.11186704842960589,
      "grad_norm": 3.7404634952545166,
      "learning_rate": 8.882379684763573e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 25710448,
      "step": 1597
    },
    {
      "epoch": 0.11193709667533513,
      "grad_norm": 4.466211318969727,
      "learning_rate": 8.881679859894922e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 25726624,
      "step": 1598
    },
    {
      "epoch": 0.11200714492106438,
      "grad_norm": 3.6850225925445557,
      "learning_rate": 8.880980035026269e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 25742456,
      "step": 1599
    },
    {
      "epoch": 0.11207719316679363,
      "grad_norm": 4.128363609313965,
      "learning_rate": 8.880280210157619e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 25758840,
      "step": 1600
    },
    {
      "epoch": 0.11207719316679363,
      "eval_loss": 1.1512293815612793,
      "eval_runtime": 0.1988,
      "eval_samples_per_second": 5.031,
      "eval_steps_per_second": 5.031,
      "num_input_tokens_seen": 25758840,
      "step": 1600
    },
    {
      "epoch": 0.11214724141252287,
      "grad_norm": 4.852661609649658,
      "learning_rate": 8.879580385288968e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 25774312,
      "step": 1601
    },
    {
      "epoch": 0.11221728965825212,
      "grad_norm": 4.501857280731201,
      "learning_rate": 8.878880560420316e-05,
      "loss": 1.302,
      "num_input_tokens_seen": 25790696,
      "step": 1602
    },
    {
      "epoch": 0.11228733790398136,
      "grad_norm": 4.142490863800049,
      "learning_rate": 8.878180735551665e-05,
      "loss": 1.0375,
      "num_input_tokens_seen": 25807080,
      "step": 1603
    },
    {
      "epoch": 0.11235738614971061,
      "grad_norm": 3.606905698776245,
      "learning_rate": 8.877480910683012e-05,
      "loss": 0.9254,
      "num_input_tokens_seen": 25822552,
      "step": 1604
    },
    {
      "epoch": 0.11242743439543985,
      "grad_norm": 3.837010145187378,
      "learning_rate": 8.876781085814361e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 25838088,
      "step": 1605
    },
    {
      "epoch": 0.1124974826411691,
      "grad_norm": 3.9082963466644287,
      "learning_rate": 8.87608126094571e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 25854240,
      "step": 1606
    },
    {
      "epoch": 0.11256753088689835,
      "grad_norm": 4.062923908233643,
      "learning_rate": 8.875381436077059e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 25870624,
      "step": 1607
    },
    {
      "epoch": 0.11263757913262759,
      "grad_norm": 4.331594944000244,
      "learning_rate": 8.874681611208407e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 25886656,
      "step": 1608
    },
    {
      "epoch": 0.11270762737835684,
      "grad_norm": 3.77466082572937,
      "learning_rate": 8.873981786339755e-05,
      "loss": 0.936,
      "num_input_tokens_seen": 25902704,
      "step": 1609
    },
    {
      "epoch": 0.1127776756240861,
      "grad_norm": 3.3747365474700928,
      "learning_rate": 8.873281961471104e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 25919088,
      "step": 1610
    },
    {
      "epoch": 0.11284772386981534,
      "grad_norm": 5.377493381500244,
      "learning_rate": 8.872582136602451e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 25935472,
      "step": 1611
    },
    {
      "epoch": 0.11291777211554459,
      "grad_norm": 5.506969451904297,
      "learning_rate": 8.8718823117338e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 25951664,
      "step": 1612
    },
    {
      "epoch": 0.11298782036127383,
      "grad_norm": 4.874104976654053,
      "learning_rate": 8.871182486865149e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 25968048,
      "step": 1613
    },
    {
      "epoch": 0.11305786860700308,
      "grad_norm": 4.666824817657471,
      "learning_rate": 8.870482661996498e-05,
      "loss": 1.2155,
      "num_input_tokens_seen": 25983784,
      "step": 1614
    },
    {
      "epoch": 0.11312791685273232,
      "grad_norm": 3.949862241744995,
      "learning_rate": 8.869782837127847e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 26000168,
      "step": 1615
    },
    {
      "epoch": 0.11319796509846157,
      "grad_norm": 3.866542339324951,
      "learning_rate": 8.869083012259196e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 26015456,
      "step": 1616
    },
    {
      "epoch": 0.11326801334419082,
      "grad_norm": 3.8679909706115723,
      "learning_rate": 8.868383187390543e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 26031224,
      "step": 1617
    },
    {
      "epoch": 0.11333806158992006,
      "grad_norm": 4.7508087158203125,
      "learning_rate": 8.867683362521892e-05,
      "loss": 1.2837,
      "num_input_tokens_seen": 26046952,
      "step": 1618
    },
    {
      "epoch": 0.1134081098356493,
      "grad_norm": 3.878549337387085,
      "learning_rate": 8.86698353765324e-05,
      "loss": 0.99,
      "num_input_tokens_seen": 26063280,
      "step": 1619
    },
    {
      "epoch": 0.11347815808137855,
      "grad_norm": 3.8016276359558105,
      "learning_rate": 8.86628371278459e-05,
      "loss": 1.1682,
      "num_input_tokens_seen": 26079616,
      "step": 1620
    },
    {
      "epoch": 0.1135482063271078,
      "grad_norm": 4.040102481842041,
      "learning_rate": 8.865583887915937e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 26095232,
      "step": 1621
    },
    {
      "epoch": 0.11361825457283704,
      "grad_norm": 3.932529926300049,
      "learning_rate": 8.864884063047286e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 26111616,
      "step": 1622
    },
    {
      "epoch": 0.11368830281856629,
      "grad_norm": 4.568112373352051,
      "learning_rate": 8.864184238178635e-05,
      "loss": 1.1932,
      "num_input_tokens_seen": 26128000,
      "step": 1623
    },
    {
      "epoch": 0.11375835106429553,
      "grad_norm": 4.23036527633667,
      "learning_rate": 8.863484413309982e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 26144384,
      "step": 1624
    },
    {
      "epoch": 0.11382839931002478,
      "grad_norm": 4.209012031555176,
      "learning_rate": 8.862784588441331e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 26160768,
      "step": 1625
    },
    {
      "epoch": 0.11389844755575403,
      "grad_norm": 3.865983724594116,
      "learning_rate": 8.86208476357268e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 26177152,
      "step": 1626
    },
    {
      "epoch": 0.11396849580148327,
      "grad_norm": 3.781083822250366,
      "learning_rate": 8.861384938704029e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 26193536,
      "step": 1627
    },
    {
      "epoch": 0.11403854404721252,
      "grad_norm": 4.330471038818359,
      "learning_rate": 8.860685113835378e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 26208976,
      "step": 1628
    },
    {
      "epoch": 0.11410859229294176,
      "grad_norm": 4.772238254547119,
      "learning_rate": 8.859985288966725e-05,
      "loss": 1.2796,
      "num_input_tokens_seen": 26225360,
      "step": 1629
    },
    {
      "epoch": 0.11417864053867101,
      "grad_norm": 4.0468668937683105,
      "learning_rate": 8.859285464098074e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 26241744,
      "step": 1630
    },
    {
      "epoch": 0.11424868878440025,
      "grad_norm": 3.9648735523223877,
      "learning_rate": 8.858585639229422e-05,
      "loss": 1.2185,
      "num_input_tokens_seen": 26258128,
      "step": 1631
    },
    {
      "epoch": 0.1143187370301295,
      "grad_norm": 4.7014079093933105,
      "learning_rate": 8.85788581436077e-05,
      "loss": 1.1795,
      "num_input_tokens_seen": 26274512,
      "step": 1632
    },
    {
      "epoch": 0.11438878527585875,
      "grad_norm": 4.6375627517700195,
      "learning_rate": 8.85718598949212e-05,
      "loss": 1.0074,
      "num_input_tokens_seen": 26290008,
      "step": 1633
    },
    {
      "epoch": 0.11445883352158799,
      "grad_norm": 4.427719593048096,
      "learning_rate": 8.856486164623468e-05,
      "loss": 1.2769,
      "num_input_tokens_seen": 26305512,
      "step": 1634
    },
    {
      "epoch": 0.11452888176731724,
      "grad_norm": 6.001821994781494,
      "learning_rate": 8.855786339754817e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 26319504,
      "step": 1635
    },
    {
      "epoch": 0.11459893001304648,
      "grad_norm": 3.970672369003296,
      "learning_rate": 8.855086514886165e-05,
      "loss": 1.1944,
      "num_input_tokens_seen": 26335888,
      "step": 1636
    },
    {
      "epoch": 0.11466897825877573,
      "grad_norm": 3.924450635910034,
      "learning_rate": 8.854386690017514e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 26351536,
      "step": 1637
    },
    {
      "epoch": 0.11473902650450497,
      "grad_norm": 4.400977611541748,
      "learning_rate": 8.853686865148861e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 26367808,
      "step": 1638
    },
    {
      "epoch": 0.11480907475023422,
      "grad_norm": 3.9734365940093994,
      "learning_rate": 8.85298704028021e-05,
      "loss": 1.2258,
      "num_input_tokens_seen": 26383864,
      "step": 1639
    },
    {
      "epoch": 0.11487912299596346,
      "grad_norm": 3.792949914932251,
      "learning_rate": 8.85228721541156e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 26400248,
      "step": 1640
    },
    {
      "epoch": 0.11494917124169271,
      "grad_norm": 5.14591121673584,
      "learning_rate": 8.851587390542908e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 26416056,
      "step": 1641
    },
    {
      "epoch": 0.11501921948742196,
      "grad_norm": 5.0158162117004395,
      "learning_rate": 8.850887565674256e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 26431400,
      "step": 1642
    },
    {
      "epoch": 0.1150892677331512,
      "grad_norm": 4.459201812744141,
      "learning_rate": 8.850187740805605e-05,
      "loss": 1.2371,
      "num_input_tokens_seen": 26446920,
      "step": 1643
    },
    {
      "epoch": 0.11515931597888045,
      "grad_norm": 3.717949867248535,
      "learning_rate": 8.849487915936953e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 26463304,
      "step": 1644
    },
    {
      "epoch": 0.1152293642246097,
      "grad_norm": 3.7555253505706787,
      "learning_rate": 8.848788091068302e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 26479296,
      "step": 1645
    },
    {
      "epoch": 0.11529941247033895,
      "grad_norm": 4.3726325035095215,
      "learning_rate": 8.84808826619965e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 26495024,
      "step": 1646
    },
    {
      "epoch": 0.1153694607160682,
      "grad_norm": 3.728700876235962,
      "learning_rate": 8.847388441331e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 26511408,
      "step": 1647
    },
    {
      "epoch": 0.11543950896179744,
      "grad_norm": 4.276855945587158,
      "learning_rate": 8.846688616462347e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 26527688,
      "step": 1648
    },
    {
      "epoch": 0.11550955720752669,
      "grad_norm": 5.386009693145752,
      "learning_rate": 8.845988791593696e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 26544072,
      "step": 1649
    },
    {
      "epoch": 0.11557960545325593,
      "grad_norm": 4.978610992431641,
      "learning_rate": 8.845288966725045e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 26560456,
      "step": 1650
    },
    {
      "epoch": 0.11564965369898518,
      "grad_norm": 5.325594425201416,
      "learning_rate": 8.844589141856392e-05,
      "loss": 0.9983,
      "num_input_tokens_seen": 26576840,
      "step": 1651
    },
    {
      "epoch": 0.11571970194471443,
      "grad_norm": 4.359868049621582,
      "learning_rate": 8.843889316987741e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 26593224,
      "step": 1652
    },
    {
      "epoch": 0.11578975019044367,
      "grad_norm": 7.921500205993652,
      "learning_rate": 8.84318949211909e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 26607352,
      "step": 1653
    },
    {
      "epoch": 0.11585979843617292,
      "grad_norm": 3.51788330078125,
      "learning_rate": 8.842489667250439e-05,
      "loss": 1.0677,
      "num_input_tokens_seen": 26623696,
      "step": 1654
    },
    {
      "epoch": 0.11592984668190216,
      "grad_norm": 4.120747089385986,
      "learning_rate": 8.841789842381788e-05,
      "loss": 1.2139,
      "num_input_tokens_seen": 26639832,
      "step": 1655
    },
    {
      "epoch": 0.11599989492763141,
      "grad_norm": 4.077361106872559,
      "learning_rate": 8.841090017513135e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 26655432,
      "step": 1656
    },
    {
      "epoch": 0.11606994317336065,
      "grad_norm": 3.9629955291748047,
      "learning_rate": 8.840390192644484e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 26671816,
      "step": 1657
    },
    {
      "epoch": 0.1161399914190899,
      "grad_norm": 3.933544635772705,
      "learning_rate": 8.839690367775831e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 26688096,
      "step": 1658
    },
    {
      "epoch": 0.11621003966481915,
      "grad_norm": 4.702983379364014,
      "learning_rate": 8.83899054290718e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 26704480,
      "step": 1659
    },
    {
      "epoch": 0.11628008791054839,
      "grad_norm": 4.536739826202393,
      "learning_rate": 8.83829071803853e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 26720864,
      "step": 1660
    },
    {
      "epoch": 0.11635013615627764,
      "grad_norm": 4.419711589813232,
      "learning_rate": 8.837590893169878e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 26737248,
      "step": 1661
    },
    {
      "epoch": 0.11642018440200688,
      "grad_norm": 4.106175899505615,
      "learning_rate": 8.836891068301227e-05,
      "loss": 1.0682,
      "num_input_tokens_seen": 26753632,
      "step": 1662
    },
    {
      "epoch": 0.11649023264773613,
      "grad_norm": 3.469658374786377,
      "learning_rate": 8.836191243432574e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 26769944,
      "step": 1663
    },
    {
      "epoch": 0.11656028089346537,
      "grad_norm": 7.273227691650391,
      "learning_rate": 8.835491418563923e-05,
      "loss": 1.1699,
      "num_input_tokens_seen": 26784520,
      "step": 1664
    },
    {
      "epoch": 0.11663032913919462,
      "grad_norm": 3.611165761947632,
      "learning_rate": 8.834791593695271e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 26800360,
      "step": 1665
    },
    {
      "epoch": 0.11670037738492386,
      "grad_norm": 4.405304908752441,
      "learning_rate": 8.834091768826621e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 26816744,
      "step": 1666
    },
    {
      "epoch": 0.11677042563065311,
      "grad_norm": 3.897247791290283,
      "learning_rate": 8.83339194395797e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 26832520,
      "step": 1667
    },
    {
      "epoch": 0.11684047387638236,
      "grad_norm": 3.898019313812256,
      "learning_rate": 8.832692119089317e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 26848080,
      "step": 1668
    },
    {
      "epoch": 0.1169105221221116,
      "grad_norm": 4.6351542472839355,
      "learning_rate": 8.831992294220666e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 26864464,
      "step": 1669
    },
    {
      "epoch": 0.11698057036784085,
      "grad_norm": 4.020237922668457,
      "learning_rate": 8.831292469352015e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 26880848,
      "step": 1670
    },
    {
      "epoch": 0.11705061861357009,
      "grad_norm": 5.813192367553711,
      "learning_rate": 8.830592644483363e-05,
      "loss": 1.2867,
      "num_input_tokens_seen": 26897232,
      "step": 1671
    },
    {
      "epoch": 0.11712066685929934,
      "grad_norm": 4.058423042297363,
      "learning_rate": 8.829892819614711e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 26912872,
      "step": 1672
    },
    {
      "epoch": 0.11719071510502858,
      "grad_norm": 4.76987361907959,
      "learning_rate": 8.82919299474606e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 26929256,
      "step": 1673
    },
    {
      "epoch": 0.11726076335075783,
      "grad_norm": 3.8400967121124268,
      "learning_rate": 8.828493169877409e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 26945624,
      "step": 1674
    },
    {
      "epoch": 0.11733081159648708,
      "grad_norm": 4.49709415435791,
      "learning_rate": 8.827793345008757e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 26961464,
      "step": 1675
    },
    {
      "epoch": 0.11740085984221632,
      "grad_norm": 4.143093109130859,
      "learning_rate": 8.827093520140105e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 26976720,
      "step": 1676
    },
    {
      "epoch": 0.11747090808794557,
      "grad_norm": 4.138030529022217,
      "learning_rate": 8.826393695271454e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 26993056,
      "step": 1677
    },
    {
      "epoch": 0.11754095633367481,
      "grad_norm": 3.8191847801208496,
      "learning_rate": 8.825693870402802e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 27009440,
      "step": 1678
    },
    {
      "epoch": 0.11761100457940406,
      "grad_norm": 3.8392176628112793,
      "learning_rate": 8.824994045534151e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 27024880,
      "step": 1679
    },
    {
      "epoch": 0.11768105282513332,
      "grad_norm": 4.468568801879883,
      "learning_rate": 8.8242942206655e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 27040672,
      "step": 1680
    },
    {
      "epoch": 0.11775110107086256,
      "grad_norm": 3.6515510082244873,
      "learning_rate": 8.823594395796848e-05,
      "loss": 1.0659,
      "num_input_tokens_seen": 27057056,
      "step": 1681
    },
    {
      "epoch": 0.11782114931659181,
      "grad_norm": 4.479739189147949,
      "learning_rate": 8.822894570928197e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 27073440,
      "step": 1682
    },
    {
      "epoch": 0.11789119756232105,
      "grad_norm": 3.762479782104492,
      "learning_rate": 8.822194746059545e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 27089824,
      "step": 1683
    },
    {
      "epoch": 0.1179612458080503,
      "grad_norm": 4.694389343261719,
      "learning_rate": 8.821494921190894e-05,
      "loss": 1.2785,
      "num_input_tokens_seen": 27106208,
      "step": 1684
    },
    {
      "epoch": 0.11803129405377955,
      "grad_norm": 3.738931179046631,
      "learning_rate": 8.820795096322241e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 27122352,
      "step": 1685
    },
    {
      "epoch": 0.11810134229950879,
      "grad_norm": 4.065624237060547,
      "learning_rate": 8.820095271453591e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 27138160,
      "step": 1686
    },
    {
      "epoch": 0.11817139054523804,
      "grad_norm": 3.5373826026916504,
      "learning_rate": 8.81939544658494e-05,
      "loss": 0.8786,
      "num_input_tokens_seen": 27154544,
      "step": 1687
    },
    {
      "epoch": 0.11824143879096728,
      "grad_norm": 3.773066282272339,
      "learning_rate": 8.818695621716288e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 27170928,
      "step": 1688
    },
    {
      "epoch": 0.11831148703669653,
      "grad_norm": 3.3876242637634277,
      "learning_rate": 8.817995796847637e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 27187312,
      "step": 1689
    },
    {
      "epoch": 0.11838153528242577,
      "grad_norm": 4.526343822479248,
      "learning_rate": 8.817295971978984e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 27202208,
      "step": 1690
    },
    {
      "epoch": 0.11845158352815502,
      "grad_norm": 4.691114902496338,
      "learning_rate": 8.816596147110333e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 27218592,
      "step": 1691
    },
    {
      "epoch": 0.11852163177388426,
      "grad_norm": 3.90531849861145,
      "learning_rate": 8.815896322241682e-05,
      "loss": 1.1438,
      "num_input_tokens_seen": 27234976,
      "step": 1692
    },
    {
      "epoch": 0.11859168001961351,
      "grad_norm": 3.5546317100524902,
      "learning_rate": 8.81519649737303e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 27251360,
      "step": 1693
    },
    {
      "epoch": 0.11866172826534276,
      "grad_norm": 5.117360591888428,
      "learning_rate": 8.81449667250438e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 27267744,
      "step": 1694
    },
    {
      "epoch": 0.118731776511072,
      "grad_norm": 4.055267810821533,
      "learning_rate": 8.813796847635727e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 27283688,
      "step": 1695
    },
    {
      "epoch": 0.11880182475680125,
      "grad_norm": 4.04268741607666,
      "learning_rate": 8.813097022767076e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 27300072,
      "step": 1696
    },
    {
      "epoch": 0.11887187300253049,
      "grad_norm": 4.048800945281982,
      "learning_rate": 8.812397197898425e-05,
      "loss": 0.9231,
      "num_input_tokens_seen": 27316456,
      "step": 1697
    },
    {
      "epoch": 0.11894192124825974,
      "grad_norm": 4.445494174957275,
      "learning_rate": 8.811697373029772e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 27332464,
      "step": 1698
    },
    {
      "epoch": 0.11901196949398898,
      "grad_norm": 4.522054672241211,
      "learning_rate": 8.810997548161121e-05,
      "loss": 1.3945,
      "num_input_tokens_seen": 27348848,
      "step": 1699
    },
    {
      "epoch": 0.11908201773971823,
      "grad_norm": 4.106349468231201,
      "learning_rate": 8.81029772329247e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 27365232,
      "step": 1700
    },
    {
      "epoch": 0.11915206598544748,
      "grad_norm": 6.059356689453125,
      "learning_rate": 8.809597898423819e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 27380448,
      "step": 1701
    },
    {
      "epoch": 0.11922211423117672,
      "grad_norm": 3.8089959621429443,
      "learning_rate": 8.808898073555166e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 27396832,
      "step": 1702
    },
    {
      "epoch": 0.11929216247690597,
      "grad_norm": 4.21024227142334,
      "learning_rate": 8.808198248686515e-05,
      "loss": 1.306,
      "num_input_tokens_seen": 27413096,
      "step": 1703
    },
    {
      "epoch": 0.11936221072263521,
      "grad_norm": 4.286004066467285,
      "learning_rate": 8.807498423817864e-05,
      "loss": 1.2325,
      "num_input_tokens_seen": 27429480,
      "step": 1704
    },
    {
      "epoch": 0.11943225896836446,
      "grad_norm": 3.512561559677124,
      "learning_rate": 8.806798598949212e-05,
      "loss": 0.8804,
      "num_input_tokens_seen": 27445864,
      "step": 1705
    },
    {
      "epoch": 0.1195023072140937,
      "grad_norm": 4.096526145935059,
      "learning_rate": 8.806098774080562e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 27462248,
      "step": 1706
    },
    {
      "epoch": 0.11957235545982295,
      "grad_norm": 5.032350063323975,
      "learning_rate": 8.805398949211909e-05,
      "loss": 0.8948,
      "num_input_tokens_seen": 27478312,
      "step": 1707
    },
    {
      "epoch": 0.1196424037055522,
      "grad_norm": 4.756420612335205,
      "learning_rate": 8.804699124343258e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 27494696,
      "step": 1708
    },
    {
      "epoch": 0.11971245195128144,
      "grad_norm": 4.869518756866455,
      "learning_rate": 8.803999299474607e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 27511080,
      "step": 1709
    },
    {
      "epoch": 0.11978250019701069,
      "grad_norm": 3.451759099960327,
      "learning_rate": 8.803299474605954e-05,
      "loss": 0.9171,
      "num_input_tokens_seen": 27527328,
      "step": 1710
    },
    {
      "epoch": 0.11985254844273993,
      "grad_norm": 4.247021675109863,
      "learning_rate": 8.802599649737303e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 27543712,
      "step": 1711
    },
    {
      "epoch": 0.11992259668846918,
      "grad_norm": 4.597024917602539,
      "learning_rate": 8.801899824868652e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 27560096,
      "step": 1712
    },
    {
      "epoch": 0.11999264493419842,
      "grad_norm": 4.242952823638916,
      "learning_rate": 8.801200000000001e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 27576320,
      "step": 1713
    },
    {
      "epoch": 0.12006269317992768,
      "grad_norm": 5.1166486740112305,
      "learning_rate": 8.80050017513135e-05,
      "loss": 1.4222,
      "num_input_tokens_seen": 27591024,
      "step": 1714
    },
    {
      "epoch": 0.12013274142565693,
      "grad_norm": 4.6713714599609375,
      "learning_rate": 8.799800350262697e-05,
      "loss": 1.1869,
      "num_input_tokens_seen": 27606352,
      "step": 1715
    },
    {
      "epoch": 0.12020278967138617,
      "grad_norm": 4.62678861618042,
      "learning_rate": 8.799100525394046e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 27622736,
      "step": 1716
    },
    {
      "epoch": 0.12027283791711542,
      "grad_norm": 3.611985206604004,
      "learning_rate": 8.798400700525394e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 27639120,
      "step": 1717
    },
    {
      "epoch": 0.12034288616284466,
      "grad_norm": 4.165099143981934,
      "learning_rate": 8.797700875656743e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 27654024,
      "step": 1718
    },
    {
      "epoch": 0.12041293440857391,
      "grad_norm": 4.532061576843262,
      "learning_rate": 8.797001050788091e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 27670408,
      "step": 1719
    },
    {
      "epoch": 0.12048298265430316,
      "grad_norm": 4.880197048187256,
      "learning_rate": 8.79630122591944e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 27686792,
      "step": 1720
    },
    {
      "epoch": 0.1205530309000324,
      "grad_norm": 3.521052360534668,
      "learning_rate": 8.795601401050789e-05,
      "loss": 0.9048,
      "num_input_tokens_seen": 27703176,
      "step": 1721
    },
    {
      "epoch": 0.12062307914576165,
      "grad_norm": 3.965725898742676,
      "learning_rate": 8.794901576182137e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 27719024,
      "step": 1722
    },
    {
      "epoch": 0.12069312739149089,
      "grad_norm": 3.936962842941284,
      "learning_rate": 8.794201751313486e-05,
      "loss": 1.1531,
      "num_input_tokens_seen": 27734736,
      "step": 1723
    },
    {
      "epoch": 0.12076317563722014,
      "grad_norm": 5.225526332855225,
      "learning_rate": 8.793501926444834e-05,
      "loss": 1.2784,
      "num_input_tokens_seen": 27751120,
      "step": 1724
    },
    {
      "epoch": 0.12083322388294938,
      "grad_norm": 4.125289440155029,
      "learning_rate": 8.792802101576182e-05,
      "loss": 1.1893,
      "num_input_tokens_seen": 27767288,
      "step": 1725
    },
    {
      "epoch": 0.12090327212867863,
      "grad_norm": 3.9352405071258545,
      "learning_rate": 8.792102276707532e-05,
      "loss": 1.1867,
      "num_input_tokens_seen": 27783672,
      "step": 1726
    },
    {
      "epoch": 0.12097332037440787,
      "grad_norm": 3.908578634262085,
      "learning_rate": 8.79140245183888e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 27799640,
      "step": 1727
    },
    {
      "epoch": 0.12104336862013712,
      "grad_norm": 3.694387435913086,
      "learning_rate": 8.790702626970229e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 27816024,
      "step": 1728
    },
    {
      "epoch": 0.12111341686586637,
      "grad_norm": 4.0100016593933105,
      "learning_rate": 8.790002802101576e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 27832408,
      "step": 1729
    },
    {
      "epoch": 0.12118346511159561,
      "grad_norm": 5.454882621765137,
      "learning_rate": 8.789302977232925e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 27848792,
      "step": 1730
    },
    {
      "epoch": 0.12125351335732486,
      "grad_norm": 5.065526485443115,
      "learning_rate": 8.788603152364274e-05,
      "loss": 1.0354,
      "num_input_tokens_seen": 27864688,
      "step": 1731
    },
    {
      "epoch": 0.1213235616030541,
      "grad_norm": 3.73103666305542,
      "learning_rate": 8.787903327495623e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 27881072,
      "step": 1732
    },
    {
      "epoch": 0.12139360984878335,
      "grad_norm": 3.971198081970215,
      "learning_rate": 8.787203502626971e-05,
      "loss": 1.1908,
      "num_input_tokens_seen": 27896912,
      "step": 1733
    },
    {
      "epoch": 0.1214636580945126,
      "grad_norm": 3.933809518814087,
      "learning_rate": 8.786503677758319e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 27913104,
      "step": 1734
    },
    {
      "epoch": 0.12153370634024184,
      "grad_norm": 3.92167329788208,
      "learning_rate": 8.785803852889668e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 27929488,
      "step": 1735
    },
    {
      "epoch": 0.12160375458597109,
      "grad_norm": 4.441089630126953,
      "learning_rate": 8.785104028021017e-05,
      "loss": 0.9748,
      "num_input_tokens_seen": 27945504,
      "step": 1736
    },
    {
      "epoch": 0.12167380283170033,
      "grad_norm": 4.023623466491699,
      "learning_rate": 8.784404203152364e-05,
      "loss": 0.8826,
      "num_input_tokens_seen": 27961888,
      "step": 1737
    },
    {
      "epoch": 0.12174385107742958,
      "grad_norm": 4.0328826904296875,
      "learning_rate": 8.783704378283713e-05,
      "loss": 1.2769,
      "num_input_tokens_seen": 27978024,
      "step": 1738
    },
    {
      "epoch": 0.12181389932315882,
      "grad_norm": 4.5445733070373535,
      "learning_rate": 8.783004553415062e-05,
      "loss": 1.3745,
      "num_input_tokens_seen": 27993840,
      "step": 1739
    },
    {
      "epoch": 0.12188394756888807,
      "grad_norm": 3.609834671020508,
      "learning_rate": 8.782304728546411e-05,
      "loss": 0.916,
      "num_input_tokens_seen": 28010224,
      "step": 1740
    },
    {
      "epoch": 0.12195399581461731,
      "grad_norm": 3.849306344985962,
      "learning_rate": 8.78160490367776e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 28026232,
      "step": 1741
    },
    {
      "epoch": 0.12202404406034656,
      "grad_norm": 4.11102294921875,
      "learning_rate": 8.780905078809107e-05,
      "loss": 1.2269,
      "num_input_tokens_seen": 28041880,
      "step": 1742
    },
    {
      "epoch": 0.1220940923060758,
      "grad_norm": 4.156986713409424,
      "learning_rate": 8.780205253940456e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 28058264,
      "step": 1743
    },
    {
      "epoch": 0.12216414055180505,
      "grad_norm": 3.9670159816741943,
      "learning_rate": 8.779505429071803e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 28073168,
      "step": 1744
    },
    {
      "epoch": 0.1222341887975343,
      "grad_norm": 5.342650890350342,
      "learning_rate": 8.778805604203152e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 28089552,
      "step": 1745
    },
    {
      "epoch": 0.12230423704326354,
      "grad_norm": 4.031285285949707,
      "learning_rate": 8.778105779334501e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 28105264,
      "step": 1746
    },
    {
      "epoch": 0.12237428528899279,
      "grad_norm": 3.5976450443267822,
      "learning_rate": 8.77740595446585e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 28121648,
      "step": 1747
    },
    {
      "epoch": 0.12244433353472203,
      "grad_norm": 4.947859764099121,
      "learning_rate": 8.776706129597199e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 28137640,
      "step": 1748
    },
    {
      "epoch": 0.12251438178045129,
      "grad_norm": 4.004949569702148,
      "learning_rate": 8.776006304728546e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 28154024,
      "step": 1749
    },
    {
      "epoch": 0.12258443002618054,
      "grad_norm": 3.9022445678710938,
      "learning_rate": 8.775306479859895e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 28170408,
      "step": 1750
    },
    {
      "epoch": 0.12265447827190978,
      "grad_norm": 4.171925067901611,
      "learning_rate": 8.774606654991244e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 28186792,
      "step": 1751
    },
    {
      "epoch": 0.12272452651763903,
      "grad_norm": 3.9387433528900146,
      "learning_rate": 8.773906830122592e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 28203176,
      "step": 1752
    },
    {
      "epoch": 0.12279457476336827,
      "grad_norm": 5.067278861999512,
      "learning_rate": 8.773207005253942e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 28219192,
      "step": 1753
    },
    {
      "epoch": 0.12286462300909752,
      "grad_norm": 3.673807144165039,
      "learning_rate": 8.77250718038529e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 28235576,
      "step": 1754
    },
    {
      "epoch": 0.12293467125482677,
      "grad_norm": 5.303588390350342,
      "learning_rate": 8.771807355516638e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 28251960,
      "step": 1755
    },
    {
      "epoch": 0.12300471950055601,
      "grad_norm": 5.343825340270996,
      "learning_rate": 8.771107530647986e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 28268344,
      "step": 1756
    },
    {
      "epoch": 0.12307476774628526,
      "grad_norm": 4.125874996185303,
      "learning_rate": 8.770407705779335e-05,
      "loss": 1.1497,
      "num_input_tokens_seen": 28284144,
      "step": 1757
    },
    {
      "epoch": 0.1231448159920145,
      "grad_norm": 4.628546714782715,
      "learning_rate": 8.769707880910683e-05,
      "loss": 1.1757,
      "num_input_tokens_seen": 28299896,
      "step": 1758
    },
    {
      "epoch": 0.12321486423774375,
      "grad_norm": 3.946603775024414,
      "learning_rate": 8.769008056042032e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 28316280,
      "step": 1759
    },
    {
      "epoch": 0.123284912483473,
      "grad_norm": 3.4837770462036133,
      "learning_rate": 8.768308231173381e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 28332128,
      "step": 1760
    },
    {
      "epoch": 0.12335496072920224,
      "grad_norm": 3.9601573944091797,
      "learning_rate": 8.767608406304729e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 28347488,
      "step": 1761
    },
    {
      "epoch": 0.12342500897493149,
      "grad_norm": 4.178001403808594,
      "learning_rate": 8.766908581436078e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 28363872,
      "step": 1762
    },
    {
      "epoch": 0.12349505722066073,
      "grad_norm": 3.9182498455047607,
      "learning_rate": 8.766208756567426e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 28380208,
      "step": 1763
    },
    {
      "epoch": 0.12356510546638998,
      "grad_norm": 4.071939468383789,
      "learning_rate": 8.765508931698774e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 28396592,
      "step": 1764
    },
    {
      "epoch": 0.12363515371211922,
      "grad_norm": 4.657908916473389,
      "learning_rate": 8.764809106830123e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 28412976,
      "step": 1765
    },
    {
      "epoch": 0.12370520195784847,
      "grad_norm": 3.9706201553344727,
      "learning_rate": 8.764109281961472e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 28429088,
      "step": 1766
    },
    {
      "epoch": 0.12377525020357771,
      "grad_norm": 4.571341514587402,
      "learning_rate": 8.76340945709282e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 28445472,
      "step": 1767
    },
    {
      "epoch": 0.12384529844930696,
      "grad_norm": 4.197002410888672,
      "learning_rate": 8.762709632224169e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 28461656,
      "step": 1768
    },
    {
      "epoch": 0.1239153466950362,
      "grad_norm": 5.376040935516357,
      "learning_rate": 8.762009807355517e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 28477088,
      "step": 1769
    },
    {
      "epoch": 0.12398539494076545,
      "grad_norm": 3.987495183944702,
      "learning_rate": 8.761309982486866e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 28493472,
      "step": 1770
    },
    {
      "epoch": 0.1240554431864947,
      "grad_norm": 4.379208564758301,
      "learning_rate": 8.760610157618213e-05,
      "loss": 1.2834,
      "num_input_tokens_seen": 28509856,
      "step": 1771
    },
    {
      "epoch": 0.12412549143222394,
      "grad_norm": 3.7258729934692383,
      "learning_rate": 8.759910332749562e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 28525664,
      "step": 1772
    },
    {
      "epoch": 0.12419553967795319,
      "grad_norm": 4.0574774742126465,
      "learning_rate": 8.759210507880911e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 28541920,
      "step": 1773
    },
    {
      "epoch": 0.12426558792368243,
      "grad_norm": 3.8423895835876465,
      "learning_rate": 8.75851068301226e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 28558216,
      "step": 1774
    },
    {
      "epoch": 0.12433563616941168,
      "grad_norm": 3.8898398876190186,
      "learning_rate": 8.757810858143609e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 28574536,
      "step": 1775
    },
    {
      "epoch": 0.12440568441514092,
      "grad_norm": 3.286412000656128,
      "learning_rate": 8.757111033274956e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 28590920,
      "step": 1776
    },
    {
      "epoch": 0.12447573266087017,
      "grad_norm": 3.7219464778900146,
      "learning_rate": 8.756411208406305e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 28607192,
      "step": 1777
    },
    {
      "epoch": 0.12454578090659942,
      "grad_norm": 3.8907012939453125,
      "learning_rate": 8.755711383537654e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 28623176,
      "step": 1778
    },
    {
      "epoch": 0.12461582915232866,
      "grad_norm": 3.8087925910949707,
      "learning_rate": 8.755011558669003e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 28639528,
      "step": 1779
    },
    {
      "epoch": 0.1246858773980579,
      "grad_norm": 4.8956217765808105,
      "learning_rate": 8.754311733800352e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 28654976,
      "step": 1780
    },
    {
      "epoch": 0.12475592564378715,
      "grad_norm": 3.7400572299957275,
      "learning_rate": 8.753611908931699e-05,
      "loss": 0.8787,
      "num_input_tokens_seen": 28671064,
      "step": 1781
    },
    {
      "epoch": 0.1248259738895164,
      "grad_norm": 4.689199924468994,
      "learning_rate": 8.752912084063048e-05,
      "loss": 1.2326,
      "num_input_tokens_seen": 28686664,
      "step": 1782
    },
    {
      "epoch": 0.12489602213524566,
      "grad_norm": 3.6594929695129395,
      "learning_rate": 8.752212259194395e-05,
      "loss": 1.1626,
      "num_input_tokens_seen": 28703048,
      "step": 1783
    },
    {
      "epoch": 0.1249660703809749,
      "grad_norm": 4.6070356369018555,
      "learning_rate": 8.751512434325744e-05,
      "loss": 1.358,
      "num_input_tokens_seen": 28719000,
      "step": 1784
    },
    {
      "epoch": 0.12503611862670413,
      "grad_norm": 4.658362865447998,
      "learning_rate": 8.750812609457093e-05,
      "loss": 1.2852,
      "num_input_tokens_seen": 28735384,
      "step": 1785
    },
    {
      "epoch": 0.1251061668724334,
      "grad_norm": 3.6963465213775635,
      "learning_rate": 8.750112784588442e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 28750856,
      "step": 1786
    },
    {
      "epoch": 0.12517621511816263,
      "grad_norm": 4.419562816619873,
      "learning_rate": 8.749412959719791e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 28766824,
      "step": 1787
    },
    {
      "epoch": 0.12524626336389189,
      "grad_norm": 4.601676940917969,
      "learning_rate": 8.74871313485114e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 28783208,
      "step": 1788
    },
    {
      "epoch": 0.12531631160962112,
      "grad_norm": 3.8597445487976074,
      "learning_rate": 8.748013309982487e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 28799160,
      "step": 1789
    },
    {
      "epoch": 0.12538635985535038,
      "grad_norm": 3.654649257659912,
      "learning_rate": 8.747313485113835e-05,
      "loss": 1.3127,
      "num_input_tokens_seen": 28815440,
      "step": 1790
    },
    {
      "epoch": 0.1254564081010796,
      "grad_norm": 4.043321132659912,
      "learning_rate": 8.746613660245184e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 28831824,
      "step": 1791
    },
    {
      "epoch": 0.12552645634680887,
      "grad_norm": 4.5223894119262695,
      "learning_rate": 8.745913835376532e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 28846984,
      "step": 1792
    },
    {
      "epoch": 0.1255965045925381,
      "grad_norm": 4.074361801147461,
      "learning_rate": 8.745214010507881e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 28863368,
      "step": 1793
    },
    {
      "epoch": 0.12566655283826736,
      "grad_norm": 4.661183834075928,
      "learning_rate": 8.74451418563923e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 28879752,
      "step": 1794
    },
    {
      "epoch": 0.1257366010839966,
      "grad_norm": 3.95831561088562,
      "learning_rate": 8.743814360770579e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 28895728,
      "step": 1795
    },
    {
      "epoch": 0.12580664932972585,
      "grad_norm": 4.271726131439209,
      "learning_rate": 8.743114535901927e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 28912112,
      "step": 1796
    },
    {
      "epoch": 0.12587669757545508,
      "grad_norm": 4.079075336456299,
      "learning_rate": 8.742414711033275e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 28928496,
      "step": 1797
    },
    {
      "epoch": 0.12594674582118434,
      "grad_norm": 4.030980587005615,
      "learning_rate": 8.741714886164623e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 28943968,
      "step": 1798
    },
    {
      "epoch": 0.12601679406691357,
      "grad_norm": 3.7285454273223877,
      "learning_rate": 8.741015061295973e-05,
      "loss": 0.8448,
      "num_input_tokens_seen": 28959800,
      "step": 1799
    },
    {
      "epoch": 0.12608684231264283,
      "grad_norm": 3.964663028717041,
      "learning_rate": 8.74031523642732e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 28976184,
      "step": 1800
    },
    {
      "epoch": 0.12608684231264283,
      "eval_loss": 1.1493111848831177,
      "eval_runtime": 0.196,
      "eval_samples_per_second": 5.102,
      "eval_steps_per_second": 5.102,
      "num_input_tokens_seen": 28976184,
      "step": 1800
    },
    {
      "epoch": 0.1261568905583721,
      "grad_norm": 4.2887396812438965,
      "learning_rate": 8.73961541155867e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 28992552,
      "step": 1801
    },
    {
      "epoch": 0.12622693880410132,
      "grad_norm": 5.139194011688232,
      "learning_rate": 8.738915586690018e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 29007480,
      "step": 1802
    },
    {
      "epoch": 0.12629698704983058,
      "grad_norm": 4.023421287536621,
      "learning_rate": 8.738215761821366e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 29023864,
      "step": 1803
    },
    {
      "epoch": 0.12636703529555982,
      "grad_norm": 3.6131162643432617,
      "learning_rate": 8.737515936952715e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 29039640,
      "step": 1804
    },
    {
      "epoch": 0.12643708354128907,
      "grad_norm": 4.477705478668213,
      "learning_rate": 8.736816112084064e-05,
      "loss": 0.8054,
      "num_input_tokens_seen": 29055816,
      "step": 1805
    },
    {
      "epoch": 0.1265071317870183,
      "grad_norm": 3.7637252807617188,
      "learning_rate": 8.736116287215412e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 29071456,
      "step": 1806
    },
    {
      "epoch": 0.12657718003274757,
      "grad_norm": 3.9611611366271973,
      "learning_rate": 8.735416462346761e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 29087840,
      "step": 1807
    },
    {
      "epoch": 0.1266472282784768,
      "grad_norm": 3.6022791862487793,
      "learning_rate": 8.734716637478109e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 29104224,
      "step": 1808
    },
    {
      "epoch": 0.12671727652420606,
      "grad_norm": 3.7403485774993896,
      "learning_rate": 8.734016812609458e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 29120608,
      "step": 1809
    },
    {
      "epoch": 0.1267873247699353,
      "grad_norm": 3.5624709129333496,
      "learning_rate": 8.733316987740805e-05,
      "loss": 1.0931,
      "num_input_tokens_seen": 29136840,
      "step": 1810
    },
    {
      "epoch": 0.12685737301566455,
      "grad_norm": 3.961516857147217,
      "learning_rate": 8.732617162872154e-05,
      "loss": 0.9529,
      "num_input_tokens_seen": 29153224,
      "step": 1811
    },
    {
      "epoch": 0.12692742126139378,
      "grad_norm": 4.895046234130859,
      "learning_rate": 8.731917338003503e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 29168336,
      "step": 1812
    },
    {
      "epoch": 0.12699746950712304,
      "grad_norm": 4.290217876434326,
      "learning_rate": 8.731217513134852e-05,
      "loss": 0.8945,
      "num_input_tokens_seen": 29184720,
      "step": 1813
    },
    {
      "epoch": 0.12706751775285227,
      "grad_norm": 3.6602399349212646,
      "learning_rate": 8.7305176882662e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 29200920,
      "step": 1814
    },
    {
      "epoch": 0.12713756599858153,
      "grad_norm": 3.7980921268463135,
      "learning_rate": 8.72981786339755e-05,
      "loss": 0.8915,
      "num_input_tokens_seen": 29217304,
      "step": 1815
    },
    {
      "epoch": 0.12720761424431076,
      "grad_norm": 3.646242141723633,
      "learning_rate": 8.729118038528897e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 29233688,
      "step": 1816
    },
    {
      "epoch": 0.12727766249004002,
      "grad_norm": 5.226564884185791,
      "learning_rate": 8.728418213660244e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 29247896,
      "step": 1817
    },
    {
      "epoch": 0.12734771073576925,
      "grad_norm": 3.8191912174224854,
      "learning_rate": 8.727718388791593e-05,
      "loss": 1.1548,
      "num_input_tokens_seen": 29263896,
      "step": 1818
    },
    {
      "epoch": 0.1274177589814985,
      "grad_norm": 4.349045276641846,
      "learning_rate": 8.727018563922944e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 29280224,
      "step": 1819
    },
    {
      "epoch": 0.12748780722722775,
      "grad_norm": 3.842888116836548,
      "learning_rate": 8.726318739054291e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 29296608,
      "step": 1820
    },
    {
      "epoch": 0.127557855472957,
      "grad_norm": 3.8854012489318848,
      "learning_rate": 8.72561891418564e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 29312992,
      "step": 1821
    },
    {
      "epoch": 0.12762790371868624,
      "grad_norm": 4.102949619293213,
      "learning_rate": 8.724919089316989e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 29328416,
      "step": 1822
    },
    {
      "epoch": 0.1276979519644155,
      "grad_norm": 5.0174336433410645,
      "learning_rate": 8.724219264448336e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 29344800,
      "step": 1823
    },
    {
      "epoch": 0.12776800021014473,
      "grad_norm": 3.6122186183929443,
      "learning_rate": 8.723519439579685e-05,
      "loss": 0.924,
      "num_input_tokens_seen": 29361184,
      "step": 1824
    },
    {
      "epoch": 0.127838048455874,
      "grad_norm": 4.086683750152588,
      "learning_rate": 8.722819614711034e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 29376840,
      "step": 1825
    },
    {
      "epoch": 0.12790809670160322,
      "grad_norm": 4.279770851135254,
      "learning_rate": 8.722119789842383e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 29393016,
      "step": 1826
    },
    {
      "epoch": 0.12797814494733248,
      "grad_norm": 5.032819747924805,
      "learning_rate": 8.72141996497373e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 29409400,
      "step": 1827
    },
    {
      "epoch": 0.1280481931930617,
      "grad_norm": 4.480144023895264,
      "learning_rate": 8.720720140105079e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 29425472,
      "step": 1828
    },
    {
      "epoch": 0.12811824143879097,
      "grad_norm": 3.6843478679656982,
      "learning_rate": 8.720020315236428e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 29441472,
      "step": 1829
    },
    {
      "epoch": 0.1281882896845202,
      "grad_norm": 3.7091941833496094,
      "learning_rate": 8.719320490367776e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 29457600,
      "step": 1830
    },
    {
      "epoch": 0.12825833793024946,
      "grad_norm": 4.122303009033203,
      "learning_rate": 8.718620665499124e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 29473984,
      "step": 1831
    },
    {
      "epoch": 0.1283283861759787,
      "grad_norm": 5.282047748565674,
      "learning_rate": 8.717920840630473e-05,
      "loss": 1.4479,
      "num_input_tokens_seen": 29490336,
      "step": 1832
    },
    {
      "epoch": 0.12839843442170795,
      "grad_norm": 4.0706586837768555,
      "learning_rate": 8.717221015761822e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 29506432,
      "step": 1833
    },
    {
      "epoch": 0.12846848266743718,
      "grad_norm": 3.856018543243408,
      "learning_rate": 8.716521190893171e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 29521744,
      "step": 1834
    },
    {
      "epoch": 0.12853853091316644,
      "grad_norm": 3.7059905529022217,
      "learning_rate": 8.715821366024518e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 29537104,
      "step": 1835
    },
    {
      "epoch": 0.1286085791588957,
      "grad_norm": 3.915038585662842,
      "learning_rate": 8.715121541155867e-05,
      "loss": 1.2072,
      "num_input_tokens_seen": 29552928,
      "step": 1836
    },
    {
      "epoch": 0.12867862740462493,
      "grad_norm": 3.6828839778900146,
      "learning_rate": 8.714421716287215e-05,
      "loss": 0.9849,
      "num_input_tokens_seen": 29569312,
      "step": 1837
    },
    {
      "epoch": 0.1287486756503542,
      "grad_norm": 4.3285441398620605,
      "learning_rate": 8.713721891418564e-05,
      "loss": 1.2812,
      "num_input_tokens_seen": 29584376,
      "step": 1838
    },
    {
      "epoch": 0.12881872389608343,
      "grad_norm": 4.646363258361816,
      "learning_rate": 8.713022066549914e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 29599856,
      "step": 1839
    },
    {
      "epoch": 0.12888877214181269,
      "grad_norm": 4.180859088897705,
      "learning_rate": 8.712322241681261e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 29616224,
      "step": 1840
    },
    {
      "epoch": 0.12895882038754192,
      "grad_norm": 3.666090250015259,
      "learning_rate": 8.71162241681261e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 29632608,
      "step": 1841
    },
    {
      "epoch": 0.12902886863327118,
      "grad_norm": 3.4623513221740723,
      "learning_rate": 8.710922591943959e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 29648992,
      "step": 1842
    },
    {
      "epoch": 0.1290989168790004,
      "grad_norm": 4.720603942871094,
      "learning_rate": 8.710222767075307e-05,
      "loss": 1.0566,
      "num_input_tokens_seen": 29665136,
      "step": 1843
    },
    {
      "epoch": 0.12916896512472967,
      "grad_norm": 4.208099365234375,
      "learning_rate": 8.709522942206654e-05,
      "loss": 1.1878,
      "num_input_tokens_seen": 29681520,
      "step": 1844
    },
    {
      "epoch": 0.1292390133704589,
      "grad_norm": 4.145462989807129,
      "learning_rate": 8.708823117338004e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 29697480,
      "step": 1845
    },
    {
      "epoch": 0.12930906161618816,
      "grad_norm": 3.9043843746185303,
      "learning_rate": 8.708123292469353e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 29713560,
      "step": 1846
    },
    {
      "epoch": 0.1293791098619174,
      "grad_norm": 4.092489242553711,
      "learning_rate": 8.707423467600701e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 29729944,
      "step": 1847
    },
    {
      "epoch": 0.12944915810764665,
      "grad_norm": 4.73677396774292,
      "learning_rate": 8.70672364273205e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 29746328,
      "step": 1848
    },
    {
      "epoch": 0.12951920635337588,
      "grad_norm": 6.134850025177002,
      "learning_rate": 8.706023817863398e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 29762584,
      "step": 1849
    },
    {
      "epoch": 0.12958925459910514,
      "grad_norm": 3.4841954708099365,
      "learning_rate": 8.705323992994746e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 29778968,
      "step": 1850
    },
    {
      "epoch": 0.12965930284483437,
      "grad_norm": 3.8816237449645996,
      "learning_rate": 8.704624168126095e-05,
      "loss": 0.7471,
      "num_input_tokens_seen": 29795352,
      "step": 1851
    },
    {
      "epoch": 0.12972935109056363,
      "grad_norm": 3.596538543701172,
      "learning_rate": 8.703924343257444e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 29811608,
      "step": 1852
    },
    {
      "epoch": 0.12979939933629286,
      "grad_norm": 3.9403269290924072,
      "learning_rate": 8.703224518388793e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 29827608,
      "step": 1853
    },
    {
      "epoch": 0.12986944758202212,
      "grad_norm": 4.586714744567871,
      "learning_rate": 8.70252469352014e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 29843992,
      "step": 1854
    },
    {
      "epoch": 0.12993949582775136,
      "grad_norm": 3.905280590057373,
      "learning_rate": 8.701824868651489e-05,
      "loss": 0.9115,
      "num_input_tokens_seen": 29860376,
      "step": 1855
    },
    {
      "epoch": 0.13000954407348062,
      "grad_norm": 4.974122524261475,
      "learning_rate": 8.701125043782838e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 29875880,
      "step": 1856
    },
    {
      "epoch": 0.13007959231920985,
      "grad_norm": 4.33966064453125,
      "learning_rate": 8.700425218914185e-05,
      "loss": 1.1955,
      "num_input_tokens_seen": 29891088,
      "step": 1857
    },
    {
      "epoch": 0.1301496405649391,
      "grad_norm": 4.593107223510742,
      "learning_rate": 8.699725394045534e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 29907472,
      "step": 1858
    },
    {
      "epoch": 0.13021968881066834,
      "grad_norm": 4.036941051483154,
      "learning_rate": 8.699025569176884e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 29923856,
      "step": 1859
    },
    {
      "epoch": 0.1302897370563976,
      "grad_norm": 3.887981653213501,
      "learning_rate": 8.698325744308232e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 29939872,
      "step": 1860
    },
    {
      "epoch": 0.13035978530212683,
      "grad_norm": 3.796053886413574,
      "learning_rate": 8.697625919439581e-05,
      "loss": 1.1678,
      "num_input_tokens_seen": 29955928,
      "step": 1861
    },
    {
      "epoch": 0.1304298335478561,
      "grad_norm": 4.5357184410095215,
      "learning_rate": 8.696926094570928e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 29970760,
      "step": 1862
    },
    {
      "epoch": 0.13049988179358532,
      "grad_norm": 5.54911994934082,
      "learning_rate": 8.696226269702277e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 29986408,
      "step": 1863
    },
    {
      "epoch": 0.13056993003931458,
      "grad_norm": 3.6517300605773926,
      "learning_rate": 8.695526444833625e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 30002792,
      "step": 1864
    },
    {
      "epoch": 0.1306399782850438,
      "grad_norm": 3.6885063648223877,
      "learning_rate": 8.694826619964975e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 30019176,
      "step": 1865
    },
    {
      "epoch": 0.13071002653077307,
      "grad_norm": 4.417117118835449,
      "learning_rate": 8.694126795096324e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 30034856,
      "step": 1866
    },
    {
      "epoch": 0.1307800747765023,
      "grad_norm": 4.070515155792236,
      "learning_rate": 8.693426970227671e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 30051240,
      "step": 1867
    },
    {
      "epoch": 0.13085012302223156,
      "grad_norm": 4.135226726531982,
      "learning_rate": 8.69272714535902e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 30067392,
      "step": 1868
    },
    {
      "epoch": 0.1309201712679608,
      "grad_norm": 4.304529666900635,
      "learning_rate": 8.692027320490369e-05,
      "loss": 1.0851,
      "num_input_tokens_seen": 30083640,
      "step": 1869
    },
    {
      "epoch": 0.13099021951369005,
      "grad_norm": 4.633643627166748,
      "learning_rate": 8.691327495621716e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 30099968,
      "step": 1870
    },
    {
      "epoch": 0.1310602677594193,
      "grad_norm": 3.6481478214263916,
      "learning_rate": 8.690627670753065e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 30116352,
      "step": 1871
    },
    {
      "epoch": 0.13113031600514855,
      "grad_norm": 4.15482234954834,
      "learning_rate": 8.689927845884414e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 30132256,
      "step": 1872
    },
    {
      "epoch": 0.1312003642508778,
      "grad_norm": 3.6562340259552,
      "learning_rate": 8.689228021015763e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 30147520,
      "step": 1873
    },
    {
      "epoch": 0.13127041249660704,
      "grad_norm": 5.323586463928223,
      "learning_rate": 8.68852819614711e-05,
      "loss": 1.4077,
      "num_input_tokens_seen": 30163880,
      "step": 1874
    },
    {
      "epoch": 0.1313404607423363,
      "grad_norm": 4.068235397338867,
      "learning_rate": 8.687828371278459e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 30180264,
      "step": 1875
    },
    {
      "epoch": 0.13141050898806553,
      "grad_norm": 3.743837594985962,
      "learning_rate": 8.687128546409808e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 30196520,
      "step": 1876
    },
    {
      "epoch": 0.1314805572337948,
      "grad_norm": 4.344557285308838,
      "learning_rate": 8.686428721541156e-05,
      "loss": 1.2745,
      "num_input_tokens_seen": 30212904,
      "step": 1877
    },
    {
      "epoch": 0.13155060547952402,
      "grad_norm": 4.048375129699707,
      "learning_rate": 8.685728896672505e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 30228464,
      "step": 1878
    },
    {
      "epoch": 0.13162065372525328,
      "grad_norm": 3.893768548965454,
      "learning_rate": 8.685029071803853e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 30244848,
      "step": 1879
    },
    {
      "epoch": 0.1316907019709825,
      "grad_norm": 4.469354629516602,
      "learning_rate": 8.684329246935202e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 30260744,
      "step": 1880
    },
    {
      "epoch": 0.13176075021671177,
      "grad_norm": 3.8471877574920654,
      "learning_rate": 8.68362942206655e-05,
      "loss": 0.8467,
      "num_input_tokens_seen": 30277128,
      "step": 1881
    },
    {
      "epoch": 0.131830798462441,
      "grad_norm": 4.37143611907959,
      "learning_rate": 8.682929597197899e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 30293184,
      "step": 1882
    },
    {
      "epoch": 0.13190084670817026,
      "grad_norm": 4.4709601402282715,
      "learning_rate": 8.682229772329247e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 30309568,
      "step": 1883
    },
    {
      "epoch": 0.1319708949538995,
      "grad_norm": 4.016445159912109,
      "learning_rate": 8.681529947460595e-05,
      "loss": 1.1499,
      "num_input_tokens_seen": 30325952,
      "step": 1884
    },
    {
      "epoch": 0.13204094319962875,
      "grad_norm": 3.6610453128814697,
      "learning_rate": 8.680830122591945e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 30341608,
      "step": 1885
    },
    {
      "epoch": 0.13211099144535798,
      "grad_norm": 4.226510524749756,
      "learning_rate": 8.680130297723294e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 30357992,
      "step": 1886
    },
    {
      "epoch": 0.13218103969108724,
      "grad_norm": 4.135020732879639,
      "learning_rate": 8.679430472854642e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 30373464,
      "step": 1887
    },
    {
      "epoch": 0.13225108793681647,
      "grad_norm": 3.858785629272461,
      "learning_rate": 8.67873064798599e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 30389336,
      "step": 1888
    },
    {
      "epoch": 0.13232113618254573,
      "grad_norm": 3.5424365997314453,
      "learning_rate": 8.678030823117338e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 30405720,
      "step": 1889
    },
    {
      "epoch": 0.13239118442827497,
      "grad_norm": 4.177000522613525,
      "learning_rate": 8.677330998248687e-05,
      "loss": 1.2172,
      "num_input_tokens_seen": 30422104,
      "step": 1890
    },
    {
      "epoch": 0.13246123267400423,
      "grad_norm": 4.08710241317749,
      "learning_rate": 8.676631173380036e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 30437560,
      "step": 1891
    },
    {
      "epoch": 0.13253128091973346,
      "grad_norm": 3.889277219772339,
      "learning_rate": 8.675931348511384e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 30453944,
      "step": 1892
    },
    {
      "epoch": 0.13260132916546272,
      "grad_norm": 3.7967042922973633,
      "learning_rate": 8.675231523642733e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 30469480,
      "step": 1893
    },
    {
      "epoch": 0.13267137741119195,
      "grad_norm": 4.2189202308654785,
      "learning_rate": 8.674531698774081e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 30485536,
      "step": 1894
    },
    {
      "epoch": 0.1327414256569212,
      "grad_norm": 4.682656764984131,
      "learning_rate": 8.67383187390543e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 30501720,
      "step": 1895
    },
    {
      "epoch": 0.13281147390265044,
      "grad_norm": 4.151151657104492,
      "learning_rate": 8.673132049036779e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 30518104,
      "step": 1896
    },
    {
      "epoch": 0.1328815221483797,
      "grad_norm": 3.700916290283203,
      "learning_rate": 8.672432224168126e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 30534488,
      "step": 1897
    },
    {
      "epoch": 0.13295157039410893,
      "grad_norm": 3.512343406677246,
      "learning_rate": 8.671732399299475e-05,
      "loss": 1.0569,
      "num_input_tokens_seen": 30550872,
      "step": 1898
    },
    {
      "epoch": 0.1330216186398382,
      "grad_norm": 3.5579488277435303,
      "learning_rate": 8.671032574430824e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 30567256,
      "step": 1899
    },
    {
      "epoch": 0.13309166688556742,
      "grad_norm": 3.7006070613861084,
      "learning_rate": 8.670332749562173e-05,
      "loss": 0.9628,
      "num_input_tokens_seen": 30582784,
      "step": 1900
    },
    {
      "epoch": 0.13316171513129668,
      "grad_norm": 4.373071670532227,
      "learning_rate": 8.66963292469352e-05,
      "loss": 1.2223,
      "num_input_tokens_seen": 30599168,
      "step": 1901
    },
    {
      "epoch": 0.1332317633770259,
      "grad_norm": 4.459958076477051,
      "learning_rate": 8.668933099824869e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 30615552,
      "step": 1902
    },
    {
      "epoch": 0.13330181162275517,
      "grad_norm": 4.919619560241699,
      "learning_rate": 8.668233274956218e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 30631936,
      "step": 1903
    },
    {
      "epoch": 0.1333718598684844,
      "grad_norm": 3.709568977355957,
      "learning_rate": 8.667533450087565e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 30648320,
      "step": 1904
    },
    {
      "epoch": 0.13344190811421366,
      "grad_norm": 4.097365379333496,
      "learning_rate": 8.666833625218916e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 30664704,
      "step": 1905
    },
    {
      "epoch": 0.13351195635994292,
      "grad_norm": 4.702358722686768,
      "learning_rate": 8.666133800350263e-05,
      "loss": 1.2809,
      "num_input_tokens_seen": 30681088,
      "step": 1906
    },
    {
      "epoch": 0.13358200460567216,
      "grad_norm": 3.7732086181640625,
      "learning_rate": 8.665433975481612e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 30697472,
      "step": 1907
    },
    {
      "epoch": 0.13365205285140142,
      "grad_norm": 5.318485260009766,
      "learning_rate": 8.66473415061296e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 30712336,
      "step": 1908
    },
    {
      "epoch": 0.13372210109713065,
      "grad_norm": 4.364311695098877,
      "learning_rate": 8.664034325744308e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 30728600,
      "step": 1909
    },
    {
      "epoch": 0.1337921493428599,
      "grad_norm": 4.860876083374023,
      "learning_rate": 8.663334500875657e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 30744832,
      "step": 1910
    },
    {
      "epoch": 0.13386219758858914,
      "grad_norm": 4.455454349517822,
      "learning_rate": 8.662634676007006e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 30761216,
      "step": 1911
    },
    {
      "epoch": 0.1339322458343184,
      "grad_norm": 4.70845365524292,
      "learning_rate": 8.661934851138355e-05,
      "loss": 1.2774,
      "num_input_tokens_seen": 30776600,
      "step": 1912
    },
    {
      "epoch": 0.13400229408004763,
      "grad_norm": 3.9769747257232666,
      "learning_rate": 8.661235026269704e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 30792632,
      "step": 1913
    },
    {
      "epoch": 0.1340723423257769,
      "grad_norm": 4.387015342712402,
      "learning_rate": 8.660535201401051e-05,
      "loss": 1.1839,
      "num_input_tokens_seen": 30809016,
      "step": 1914
    },
    {
      "epoch": 0.13414239057150612,
      "grad_norm": 4.786890506744385,
      "learning_rate": 8.6598353765324e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 30825136,
      "step": 1915
    },
    {
      "epoch": 0.13421243881723538,
      "grad_norm": 3.502570629119873,
      "learning_rate": 8.659135551663748e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 30841472,
      "step": 1916
    },
    {
      "epoch": 0.1342824870629646,
      "grad_norm": 4.2404913902282715,
      "learning_rate": 8.658435726795096e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 30857856,
      "step": 1917
    },
    {
      "epoch": 0.13435253530869387,
      "grad_norm": 4.230425834655762,
      "learning_rate": 8.657735901926445e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 30874240,
      "step": 1918
    },
    {
      "epoch": 0.1344225835544231,
      "grad_norm": 3.9034597873687744,
      "learning_rate": 8.657036077057794e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 30890560,
      "step": 1919
    },
    {
      "epoch": 0.13449263180015236,
      "grad_norm": 3.829190492630005,
      "learning_rate": 8.656336252189143e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 30906480,
      "step": 1920
    },
    {
      "epoch": 0.1345626800458816,
      "grad_norm": 3.9801993370056152,
      "learning_rate": 8.65563642732049e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 30922160,
      "step": 1921
    },
    {
      "epoch": 0.13463272829161085,
      "grad_norm": 5.018815994262695,
      "learning_rate": 8.65493660245184e-05,
      "loss": 1.1155,
      "num_input_tokens_seen": 30938544,
      "step": 1922
    },
    {
      "epoch": 0.13470277653734009,
      "grad_norm": 3.6515283584594727,
      "learning_rate": 8.654236777583188e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 30954088,
      "step": 1923
    },
    {
      "epoch": 0.13477282478306934,
      "grad_norm": 4.440131664276123,
      "learning_rate": 8.653536952714536e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 30970472,
      "step": 1924
    },
    {
      "epoch": 0.13484287302879858,
      "grad_norm": 5.27577543258667,
      "learning_rate": 8.652837127845885e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 30985544,
      "step": 1925
    },
    {
      "epoch": 0.13491292127452784,
      "grad_norm": 4.632978916168213,
      "learning_rate": 8.652137302977233e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 31001928,
      "step": 1926
    },
    {
      "epoch": 0.13498296952025707,
      "grad_norm": 3.9239861965179443,
      "learning_rate": 8.651437478108582e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 31018312,
      "step": 1927
    },
    {
      "epoch": 0.13505301776598633,
      "grad_norm": 4.819107532501221,
      "learning_rate": 8.65073765323993e-05,
      "loss": 1.1631,
      "num_input_tokens_seen": 31033568,
      "step": 1928
    },
    {
      "epoch": 0.13512306601171556,
      "grad_norm": 3.5287766456604004,
      "learning_rate": 8.650037828371279e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 31049952,
      "step": 1929
    },
    {
      "epoch": 0.13519311425744482,
      "grad_norm": 3.536736488342285,
      "learning_rate": 8.649338003502628e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 31066336,
      "step": 1930
    },
    {
      "epoch": 0.13526316250317405,
      "grad_norm": 5.148278713226318,
      "learning_rate": 8.648638178633976e-05,
      "loss": 1.2137,
      "num_input_tokens_seen": 31082136,
      "step": 1931
    },
    {
      "epoch": 0.1353332107489033,
      "grad_norm": 4.076564788818359,
      "learning_rate": 8.647938353765325e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 31098520,
      "step": 1932
    },
    {
      "epoch": 0.13540325899463254,
      "grad_norm": 4.747740745544434,
      "learning_rate": 8.647238528896673e-05,
      "loss": 1.1989,
      "num_input_tokens_seen": 31114560,
      "step": 1933
    },
    {
      "epoch": 0.1354733072403618,
      "grad_norm": 3.662280797958374,
      "learning_rate": 8.646538704028022e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 31130944,
      "step": 1934
    },
    {
      "epoch": 0.13554335548609103,
      "grad_norm": 3.8747782707214355,
      "learning_rate": 8.645838879159369e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 31146544,
      "step": 1935
    },
    {
      "epoch": 0.1356134037318203,
      "grad_norm": 3.465095281600952,
      "learning_rate": 8.645139054290718e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 31162928,
      "step": 1936
    },
    {
      "epoch": 0.13568345197754952,
      "grad_norm": 4.640190124511719,
      "learning_rate": 8.644439229422067e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 31177712,
      "step": 1937
    },
    {
      "epoch": 0.13575350022327878,
      "grad_norm": 3.88620924949646,
      "learning_rate": 8.643739404553416e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 31193640,
      "step": 1938
    },
    {
      "epoch": 0.13582354846900804,
      "grad_norm": 3.657331705093384,
      "learning_rate": 8.643039579684765e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 31209112,
      "step": 1939
    },
    {
      "epoch": 0.13589359671473727,
      "grad_norm": 6.8866448402404785,
      "learning_rate": 8.642339754816113e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 31223968,
      "step": 1940
    },
    {
      "epoch": 0.13596364496046653,
      "grad_norm": 5.0794172286987305,
      "learning_rate": 8.641639929947461e-05,
      "loss": 1.1988,
      "num_input_tokens_seen": 31240352,
      "step": 1941
    },
    {
      "epoch": 0.13603369320619577,
      "grad_norm": 4.631995677947998,
      "learning_rate": 8.64094010507881e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 31256736,
      "step": 1942
    },
    {
      "epoch": 0.13610374145192503,
      "grad_norm": 5.566014766693115,
      "learning_rate": 8.640240280210157e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 31273120,
      "step": 1943
    },
    {
      "epoch": 0.13617378969765426,
      "grad_norm": 3.940988302230835,
      "learning_rate": 8.639540455341506e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 31289504,
      "step": 1944
    },
    {
      "epoch": 0.13624383794338352,
      "grad_norm": 3.9979453086853027,
      "learning_rate": 8.638840630472855e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 31305888,
      "step": 1945
    },
    {
      "epoch": 0.13631388618911275,
      "grad_norm": 5.303500175476074,
      "learning_rate": 8.638140805604204e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 31321512,
      "step": 1946
    },
    {
      "epoch": 0.136383934434842,
      "grad_norm": 4.6745429039001465,
      "learning_rate": 8.637440980735553e-05,
      "loss": 1.3665,
      "num_input_tokens_seen": 31337896,
      "step": 1947
    },
    {
      "epoch": 0.13645398268057124,
      "grad_norm": 4.203839302062988,
      "learning_rate": 8.6367411558669e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 31354176,
      "step": 1948
    },
    {
      "epoch": 0.1365240309263005,
      "grad_norm": 4.802511215209961,
      "learning_rate": 8.636041330998249e-05,
      "loss": 1.2427,
      "num_input_tokens_seen": 31369976,
      "step": 1949
    },
    {
      "epoch": 0.13659407917202973,
      "grad_norm": 4.077885627746582,
      "learning_rate": 8.635341506129598e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 31386360,
      "step": 1950
    },
    {
      "epoch": 0.136664127417759,
      "grad_norm": 5.009285926818848,
      "learning_rate": 8.634641681260947e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 31402744,
      "step": 1951
    },
    {
      "epoch": 0.13673417566348822,
      "grad_norm": 3.539872646331787,
      "learning_rate": 8.633941856392294e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 31419128,
      "step": 1952
    },
    {
      "epoch": 0.13680422390921748,
      "grad_norm": 4.664520740509033,
      "learning_rate": 8.633242031523643e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 31435400,
      "step": 1953
    },
    {
      "epoch": 0.1368742721549467,
      "grad_norm": 3.8469269275665283,
      "learning_rate": 8.632542206654992e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 31451408,
      "step": 1954
    },
    {
      "epoch": 0.13694432040067597,
      "grad_norm": 4.064670085906982,
      "learning_rate": 8.63184238178634e-05,
      "loss": 0.8825,
      "num_input_tokens_seen": 31467504,
      "step": 1955
    },
    {
      "epoch": 0.1370143686464052,
      "grad_norm": 3.9931817054748535,
      "learning_rate": 8.631142556917688e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 31483528,
      "step": 1956
    },
    {
      "epoch": 0.13708441689213446,
      "grad_norm": 4.136581897735596,
      "learning_rate": 8.630442732049037e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 31499912,
      "step": 1957
    },
    {
      "epoch": 0.1371544651378637,
      "grad_norm": 3.7189536094665527,
      "learning_rate": 8.629742907180386e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 31515560,
      "step": 1958
    },
    {
      "epoch": 0.13722451338359296,
      "grad_norm": 3.7821719646453857,
      "learning_rate": 8.629043082311735e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 31531944,
      "step": 1959
    },
    {
      "epoch": 0.1372945616293222,
      "grad_norm": 6.815886497497559,
      "learning_rate": 8.628343257443082e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 31548248,
      "step": 1960
    },
    {
      "epoch": 0.13736460987505145,
      "grad_norm": 7.490451812744141,
      "learning_rate": 8.627643432574431e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 31562560,
      "step": 1961
    },
    {
      "epoch": 0.13743465812078068,
      "grad_norm": 4.918768405914307,
      "learning_rate": 8.626943607705779e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 31578944,
      "step": 1962
    },
    {
      "epoch": 0.13750470636650994,
      "grad_norm": 5.567696571350098,
      "learning_rate": 8.626243782837128e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 31594312,
      "step": 1963
    },
    {
      "epoch": 0.13757475461223917,
      "grad_norm": 4.24015474319458,
      "learning_rate": 8.625543957968477e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 31609656,
      "step": 1964
    },
    {
      "epoch": 0.13764480285796843,
      "grad_norm": 5.664759635925293,
      "learning_rate": 8.624844133099825e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 31626040,
      "step": 1965
    },
    {
      "epoch": 0.13771485110369766,
      "grad_norm": 3.7281267642974854,
      "learning_rate": 8.624144308231174e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 31642424,
      "step": 1966
    },
    {
      "epoch": 0.13778489934942692,
      "grad_norm": 4.112753391265869,
      "learning_rate": 8.623444483362523e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 31658808,
      "step": 1967
    },
    {
      "epoch": 0.13785494759515615,
      "grad_norm": 3.8851754665374756,
      "learning_rate": 8.62274465849387e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 31675192,
      "step": 1968
    },
    {
      "epoch": 0.1379249958408854,
      "grad_norm": 4.161825656890869,
      "learning_rate": 8.62204483362522e-05,
      "loss": 1.03,
      "num_input_tokens_seen": 31691576,
      "step": 1969
    },
    {
      "epoch": 0.13799504408661464,
      "grad_norm": 4.802804470062256,
      "learning_rate": 8.621345008756567e-05,
      "loss": 1.4374,
      "num_input_tokens_seen": 31707960,
      "step": 1970
    },
    {
      "epoch": 0.1380650923323439,
      "grad_norm": 3.752012252807617,
      "learning_rate": 8.620645183887917e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 31724344,
      "step": 1971
    },
    {
      "epoch": 0.13813514057807313,
      "grad_norm": 3.8039815425872803,
      "learning_rate": 8.619945359019265e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 31740456,
      "step": 1972
    },
    {
      "epoch": 0.1382051888238024,
      "grad_norm": 4.029634952545166,
      "learning_rate": 8.619245534150614e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 31756776,
      "step": 1973
    },
    {
      "epoch": 0.13827523706953165,
      "grad_norm": 5.531665802001953,
      "learning_rate": 8.618545709281962e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 31772480,
      "step": 1974
    },
    {
      "epoch": 0.13834528531526089,
      "grad_norm": 4.6494646072387695,
      "learning_rate": 8.61784588441331e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 31788504,
      "step": 1975
    },
    {
      "epoch": 0.13841533356099014,
      "grad_norm": 4.201340675354004,
      "learning_rate": 8.617146059544659e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 31804888,
      "step": 1976
    },
    {
      "epoch": 0.13848538180671938,
      "grad_norm": 4.272038459777832,
      "learning_rate": 8.616446234676008e-05,
      "loss": 1.2557,
      "num_input_tokens_seen": 31821272,
      "step": 1977
    },
    {
      "epoch": 0.13855543005244864,
      "grad_norm": 3.729841947555542,
      "learning_rate": 8.615746409807357e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 31837656,
      "step": 1978
    },
    {
      "epoch": 0.13862547829817787,
      "grad_norm": 3.5615944862365723,
      "learning_rate": 8.615046584938704e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 31854040,
      "step": 1979
    },
    {
      "epoch": 0.13869552654390713,
      "grad_norm": 3.7658376693725586,
      "learning_rate": 8.614346760070053e-05,
      "loss": 1.3268,
      "num_input_tokens_seen": 31870424,
      "step": 1980
    },
    {
      "epoch": 0.13876557478963636,
      "grad_norm": 4.124275207519531,
      "learning_rate": 8.613646935201402e-05,
      "loss": 1.2736,
      "num_input_tokens_seen": 31886808,
      "step": 1981
    },
    {
      "epoch": 0.13883562303536562,
      "grad_norm": 5.348685264587402,
      "learning_rate": 8.612947110332749e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 31902880,
      "step": 1982
    },
    {
      "epoch": 0.13890567128109485,
      "grad_norm": 5.311651706695557,
      "learning_rate": 8.612247285464098e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 31918704,
      "step": 1983
    },
    {
      "epoch": 0.1389757195268241,
      "grad_norm": 4.194555759429932,
      "learning_rate": 8.611547460595447e-05,
      "loss": 1.2802,
      "num_input_tokens_seen": 31935088,
      "step": 1984
    },
    {
      "epoch": 0.13904576777255334,
      "grad_norm": 3.6576390266418457,
      "learning_rate": 8.610847635726796e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 31951472,
      "step": 1985
    },
    {
      "epoch": 0.1391158160182826,
      "grad_norm": 4.169801235198975,
      "learning_rate": 8.610147810858145e-05,
      "loss": 1.1668,
      "num_input_tokens_seen": 31967856,
      "step": 1986
    },
    {
      "epoch": 0.13918586426401183,
      "grad_norm": 3.79791259765625,
      "learning_rate": 8.609447985989492e-05,
      "loss": 1.2546,
      "num_input_tokens_seen": 31984232,
      "step": 1987
    },
    {
      "epoch": 0.1392559125097411,
      "grad_norm": 3.726701021194458,
      "learning_rate": 8.608748161120841e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 31999720,
      "step": 1988
    },
    {
      "epoch": 0.13932596075547032,
      "grad_norm": 3.7376129627227783,
      "learning_rate": 8.608048336252189e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 32016104,
      "step": 1989
    },
    {
      "epoch": 0.13939600900119958,
      "grad_norm": 4.290423393249512,
      "learning_rate": 8.607348511383537e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 32031992,
      "step": 1990
    },
    {
      "epoch": 0.13946605724692882,
      "grad_norm": 3.592384099960327,
      "learning_rate": 8.606648686514888e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 32047904,
      "step": 1991
    },
    {
      "epoch": 0.13953610549265807,
      "grad_norm": 3.753692626953125,
      "learning_rate": 8.605948861646235e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 32063720,
      "step": 1992
    },
    {
      "epoch": 0.1396061537383873,
      "grad_norm": 4.698465347290039,
      "learning_rate": 8.605249036777584e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 32079768,
      "step": 1993
    },
    {
      "epoch": 0.13967620198411657,
      "grad_norm": 4.187407970428467,
      "learning_rate": 8.604549211908933e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 32095120,
      "step": 1994
    },
    {
      "epoch": 0.1397462502298458,
      "grad_norm": 4.256275653839111,
      "learning_rate": 8.60384938704028e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 32111408,
      "step": 1995
    },
    {
      "epoch": 0.13981629847557506,
      "grad_norm": 6.693331241607666,
      "learning_rate": 8.603149562171629e-05,
      "loss": 1.2317,
      "num_input_tokens_seen": 32127792,
      "step": 1996
    },
    {
      "epoch": 0.1398863467213043,
      "grad_norm": 3.69393253326416,
      "learning_rate": 8.602449737302978e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 32143792,
      "step": 1997
    },
    {
      "epoch": 0.13995639496703355,
      "grad_norm": 4.117836952209473,
      "learning_rate": 8.601749912434327e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 32158624,
      "step": 1998
    },
    {
      "epoch": 0.14002644321276278,
      "grad_norm": 5.14541482925415,
      "learning_rate": 8.601050087565674e-05,
      "loss": 1.1787,
      "num_input_tokens_seen": 32175008,
      "step": 1999
    },
    {
      "epoch": 0.14009649145849204,
      "grad_norm": 4.0103535652160645,
      "learning_rate": 8.600350262697023e-05,
      "loss": 1.1308,
      "num_input_tokens_seen": 32191392,
      "step": 2000
    },
    {
      "epoch": 0.14009649145849204,
      "eval_loss": 1.1461617946624756,
      "eval_runtime": 0.1945,
      "eval_samples_per_second": 5.141,
      "eval_steps_per_second": 5.141,
      "num_input_tokens_seen": 32191392,
      "step": 2000
    },
    {
      "epoch": 0.14016653970422127,
      "grad_norm": 3.8072049617767334,
      "learning_rate": 8.599650437828372e-05,
      "loss": 0.8617,
      "num_input_tokens_seen": 32207712,
      "step": 2001
    },
    {
      "epoch": 0.14023658794995053,
      "grad_norm": 4.034494400024414,
      "learning_rate": 8.59895061295972e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 32223440,
      "step": 2002
    },
    {
      "epoch": 0.14030663619567976,
      "grad_norm": 3.9485251903533936,
      "learning_rate": 8.598250788091069e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 32239824,
      "step": 2003
    },
    {
      "epoch": 0.14037668444140902,
      "grad_norm": 5.427109241485596,
      "learning_rate": 8.597550963222417e-05,
      "loss": 1.1922,
      "num_input_tokens_seen": 32255976,
      "step": 2004
    },
    {
      "epoch": 0.14044673268713825,
      "grad_norm": 4.4832000732421875,
      "learning_rate": 8.596851138353766e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 32272304,
      "step": 2005
    },
    {
      "epoch": 0.1405167809328675,
      "grad_norm": 4.4699859619140625,
      "learning_rate": 8.596151313485114e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 32288688,
      "step": 2006
    },
    {
      "epoch": 0.14058682917859674,
      "grad_norm": 6.007316589355469,
      "learning_rate": 8.595451488616463e-05,
      "loss": 1.2402,
      "num_input_tokens_seen": 32304992,
      "step": 2007
    },
    {
      "epoch": 0.140656877424326,
      "grad_norm": 5.460748195648193,
      "learning_rate": 8.594751663747811e-05,
      "loss": 1.2683,
      "num_input_tokens_seen": 32320104,
      "step": 2008
    },
    {
      "epoch": 0.14072692567005526,
      "grad_norm": 4.430675029754639,
      "learning_rate": 8.594051838879159e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 32336040,
      "step": 2009
    },
    {
      "epoch": 0.1407969739157845,
      "grad_norm": 4.469089508056641,
      "learning_rate": 8.593352014010508e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 32352424,
      "step": 2010
    },
    {
      "epoch": 0.14086702216151376,
      "grad_norm": 4.94099760055542,
      "learning_rate": 8.592652189141858e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 32367944,
      "step": 2011
    },
    {
      "epoch": 0.140937070407243,
      "grad_norm": 5.430322170257568,
      "learning_rate": 8.591952364273206e-05,
      "loss": 1.2256,
      "num_input_tokens_seen": 32384328,
      "step": 2012
    },
    {
      "epoch": 0.14100711865297225,
      "grad_norm": 3.847569704055786,
      "learning_rate": 8.591252539404554e-05,
      "loss": 1.019,
      "num_input_tokens_seen": 32400712,
      "step": 2013
    },
    {
      "epoch": 0.14107716689870148,
      "grad_norm": 3.7531189918518066,
      "learning_rate": 8.590552714535902e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 32417096,
      "step": 2014
    },
    {
      "epoch": 0.14114721514443074,
      "grad_norm": 4.070606708526611,
      "learning_rate": 8.589852889667251e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 32432504,
      "step": 2015
    },
    {
      "epoch": 0.14121726339015997,
      "grad_norm": 4.791952610015869,
      "learning_rate": 8.589153064798598e-05,
      "loss": 0.8467,
      "num_input_tokens_seen": 32448008,
      "step": 2016
    },
    {
      "epoch": 0.14128731163588923,
      "grad_norm": 4.672977924346924,
      "learning_rate": 8.588453239929948e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 32463792,
      "step": 2017
    },
    {
      "epoch": 0.14135735988161846,
      "grad_norm": 6.187239170074463,
      "learning_rate": 8.587753415061297e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 32480104,
      "step": 2018
    },
    {
      "epoch": 0.14142740812734772,
      "grad_norm": 4.058189392089844,
      "learning_rate": 8.587053590192645e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 32495824,
      "step": 2019
    },
    {
      "epoch": 0.14149745637307695,
      "grad_norm": 3.862661838531494,
      "learning_rate": 8.586353765323994e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 32512208,
      "step": 2020
    },
    {
      "epoch": 0.1415675046188062,
      "grad_norm": 3.7348716259002686,
      "learning_rate": 8.585653940455343e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 32528592,
      "step": 2021
    },
    {
      "epoch": 0.14163755286453544,
      "grad_norm": 4.405923843383789,
      "learning_rate": 8.58495411558669e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 32544120,
      "step": 2022
    },
    {
      "epoch": 0.1417076011102647,
      "grad_norm": 3.73984694480896,
      "learning_rate": 8.584254290718039e-05,
      "loss": 1.0797,
      "num_input_tokens_seen": 32560504,
      "step": 2023
    },
    {
      "epoch": 0.14177764935599393,
      "grad_norm": 5.73613166809082,
      "learning_rate": 8.583554465849388e-05,
      "loss": 1.2119,
      "num_input_tokens_seen": 32576888,
      "step": 2024
    },
    {
      "epoch": 0.1418476976017232,
      "grad_norm": 6.435116291046143,
      "learning_rate": 8.582854640980737e-05,
      "loss": 1.3408,
      "num_input_tokens_seen": 32591592,
      "step": 2025
    },
    {
      "epoch": 0.14191774584745243,
      "grad_norm": 4.520002365112305,
      "learning_rate": 8.582154816112084e-05,
      "loss": 1.1654,
      "num_input_tokens_seen": 32607448,
      "step": 2026
    },
    {
      "epoch": 0.14198779409318169,
      "grad_norm": 4.01891565322876,
      "learning_rate": 8.581454991243433e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 32623104,
      "step": 2027
    },
    {
      "epoch": 0.14205784233891092,
      "grad_norm": 3.8237030506134033,
      "learning_rate": 8.580755166374782e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 32639376,
      "step": 2028
    },
    {
      "epoch": 0.14212789058464018,
      "grad_norm": 4.893499851226807,
      "learning_rate": 8.58005534150613e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 32654800,
      "step": 2029
    },
    {
      "epoch": 0.1421979388303694,
      "grad_norm": 3.6075315475463867,
      "learning_rate": 8.579355516637478e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 32671184,
      "step": 2030
    },
    {
      "epoch": 0.14226798707609867,
      "grad_norm": 4.696410179138184,
      "learning_rate": 8.578655691768827e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 32687360,
      "step": 2031
    },
    {
      "epoch": 0.1423380353218279,
      "grad_norm": 3.9465558528900146,
      "learning_rate": 8.577955866900176e-05,
      "loss": 1.1378,
      "num_input_tokens_seen": 32703744,
      "step": 2032
    },
    {
      "epoch": 0.14240808356755716,
      "grad_norm": 3.933898448944092,
      "learning_rate": 8.577256042031523e-05,
      "loss": 0.8353,
      "num_input_tokens_seen": 32720128,
      "step": 2033
    },
    {
      "epoch": 0.1424781318132864,
      "grad_norm": 3.865894317626953,
      "learning_rate": 8.576556217162872e-05,
      "loss": 0.9827,
      "num_input_tokens_seen": 32735976,
      "step": 2034
    },
    {
      "epoch": 0.14254818005901565,
      "grad_norm": 3.9533474445343018,
      "learning_rate": 8.575856392294221e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 32752240,
      "step": 2035
    },
    {
      "epoch": 0.14261822830474488,
      "grad_norm": 3.5534164905548096,
      "learning_rate": 8.575156567425569e-05,
      "loss": 1.0887,
      "num_input_tokens_seen": 32768624,
      "step": 2036
    },
    {
      "epoch": 0.14268827655047414,
      "grad_norm": 5.689724922180176,
      "learning_rate": 8.574456742556918e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 32784600,
      "step": 2037
    },
    {
      "epoch": 0.14275832479620337,
      "grad_norm": 4.010136604309082,
      "learning_rate": 8.573756917688268e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 32799824,
      "step": 2038
    },
    {
      "epoch": 0.14282837304193263,
      "grad_norm": 4.153547763824463,
      "learning_rate": 8.573057092819615e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 32815744,
      "step": 2039
    },
    {
      "epoch": 0.14289842128766186,
      "grad_norm": 3.976120948791504,
      "learning_rate": 8.572357267950964e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 32831664,
      "step": 2040
    },
    {
      "epoch": 0.14296846953339112,
      "grad_norm": 3.9593231678009033,
      "learning_rate": 8.571657443082312e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 32848048,
      "step": 2041
    },
    {
      "epoch": 0.14303851777912036,
      "grad_norm": 3.920823097229004,
      "learning_rate": 8.57095761821366e-05,
      "loss": 1.1891,
      "num_input_tokens_seen": 32863168,
      "step": 2042
    },
    {
      "epoch": 0.14310856602484961,
      "grad_norm": 4.754055976867676,
      "learning_rate": 8.57025779334501e-05,
      "loss": 1.1123,
      "num_input_tokens_seen": 32879552,
      "step": 2043
    },
    {
      "epoch": 0.14317861427057887,
      "grad_norm": 3.6835105419158936,
      "learning_rate": 8.569557968476358e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 32895864,
      "step": 2044
    },
    {
      "epoch": 0.1432486625163081,
      "grad_norm": 4.115698337554932,
      "learning_rate": 8.568858143607707e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 32912232,
      "step": 2045
    },
    {
      "epoch": 0.14331871076203737,
      "grad_norm": 6.536626815795898,
      "learning_rate": 8.568158318739055e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 32928616,
      "step": 2046
    },
    {
      "epoch": 0.1433887590077666,
      "grad_norm": 5.43113899230957,
      "learning_rate": 8.567458493870403e-05,
      "loss": 0.9645,
      "num_input_tokens_seen": 32945000,
      "step": 2047
    },
    {
      "epoch": 0.14345880725349586,
      "grad_norm": 3.8677239418029785,
      "learning_rate": 8.566758669001752e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 32961384,
      "step": 2048
    },
    {
      "epoch": 0.1435288554992251,
      "grad_norm": 6.913444995880127,
      "learning_rate": 8.5660588441331e-05,
      "loss": 1.2204,
      "num_input_tokens_seen": 32977768,
      "step": 2049
    },
    {
      "epoch": 0.14359890374495435,
      "grad_norm": 4.870579719543457,
      "learning_rate": 8.565359019264449e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 32994152,
      "step": 2050
    },
    {
      "epoch": 0.14366895199068358,
      "grad_norm": 4.057044982910156,
      "learning_rate": 8.564659194395797e-05,
      "loss": 1.0599,
      "num_input_tokens_seen": 33010536,
      "step": 2051
    },
    {
      "epoch": 0.14373900023641284,
      "grad_norm": 8.405828475952148,
      "learning_rate": 8.563959369527146e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 33025192,
      "step": 2052
    },
    {
      "epoch": 0.14380904848214207,
      "grad_norm": 4.188510894775391,
      "learning_rate": 8.563259544658494e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 33041576,
      "step": 2053
    },
    {
      "epoch": 0.14387909672787133,
      "grad_norm": 6.505815505981445,
      "learning_rate": 8.562559719789843e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 33057800,
      "step": 2054
    },
    {
      "epoch": 0.14394914497360056,
      "grad_norm": 4.021209716796875,
      "learning_rate": 8.561859894921192e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 33073872,
      "step": 2055
    },
    {
      "epoch": 0.14401919321932982,
      "grad_norm": 3.924671173095703,
      "learning_rate": 8.561160070052539e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 33090256,
      "step": 2056
    },
    {
      "epoch": 0.14408924146505905,
      "grad_norm": 4.1323418617248535,
      "learning_rate": 8.560460245183888e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 33106256,
      "step": 2057
    },
    {
      "epoch": 0.1441592897107883,
      "grad_norm": 4.236043930053711,
      "learning_rate": 8.559760420315237e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 33122352,
      "step": 2058
    },
    {
      "epoch": 0.14422933795651754,
      "grad_norm": 3.4836020469665527,
      "learning_rate": 8.559060595446586e-05,
      "loss": 1.0136,
      "num_input_tokens_seen": 33138736,
      "step": 2059
    },
    {
      "epoch": 0.1442993862022468,
      "grad_norm": 4.363439083099365,
      "learning_rate": 8.558360770577933e-05,
      "loss": 1.1382,
      "num_input_tokens_seen": 33153936,
      "step": 2060
    },
    {
      "epoch": 0.14436943444797604,
      "grad_norm": 5.099925994873047,
      "learning_rate": 8.557660945709282e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 33170320,
      "step": 2061
    },
    {
      "epoch": 0.1444394826937053,
      "grad_norm": 4.438295364379883,
      "learning_rate": 8.556961120840631e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 33186704,
      "step": 2062
    },
    {
      "epoch": 0.14450953093943453,
      "grad_norm": 3.7912747859954834,
      "learning_rate": 8.556261295971978e-05,
      "loss": 1.0708,
      "num_input_tokens_seen": 33203088,
      "step": 2063
    },
    {
      "epoch": 0.1445795791851638,
      "grad_norm": 4.679794788360596,
      "learning_rate": 8.555561471103329e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 33218936,
      "step": 2064
    },
    {
      "epoch": 0.14464962743089302,
      "grad_norm": 4.092919826507568,
      "learning_rate": 8.554861646234677e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 33235320,
      "step": 2065
    },
    {
      "epoch": 0.14471967567662228,
      "grad_norm": 4.13189172744751,
      "learning_rate": 8.554161821366025e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 33251704,
      "step": 2066
    },
    {
      "epoch": 0.1447897239223515,
      "grad_norm": 3.618739366531372,
      "learning_rate": 8.553461996497374e-05,
      "loss": 1.0026,
      "num_input_tokens_seen": 33268088,
      "step": 2067
    },
    {
      "epoch": 0.14485977216808077,
      "grad_norm": 4.197813034057617,
      "learning_rate": 8.552762171628721e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 33284472,
      "step": 2068
    },
    {
      "epoch": 0.14492982041381,
      "grad_norm": 4.159245491027832,
      "learning_rate": 8.55206234676007e-05,
      "loss": 0.9579,
      "num_input_tokens_seen": 33300560,
      "step": 2069
    },
    {
      "epoch": 0.14499986865953926,
      "grad_norm": 3.970898389816284,
      "learning_rate": 8.551362521891419e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 33316744,
      "step": 2070
    },
    {
      "epoch": 0.1450699169052685,
      "grad_norm": 5.635775089263916,
      "learning_rate": 8.550662697022768e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 33333128,
      "step": 2071
    },
    {
      "epoch": 0.14513996515099775,
      "grad_norm": 5.239542007446289,
      "learning_rate": 8.549962872154117e-05,
      "loss": 1.166,
      "num_input_tokens_seen": 33348392,
      "step": 2072
    },
    {
      "epoch": 0.14521001339672698,
      "grad_norm": 3.8646957874298096,
      "learning_rate": 8.549263047285464e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 33364504,
      "step": 2073
    },
    {
      "epoch": 0.14528006164245624,
      "grad_norm": 4.49400520324707,
      "learning_rate": 8.548563222416813e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 33379680,
      "step": 2074
    },
    {
      "epoch": 0.14535010988818547,
      "grad_norm": 3.782107353210449,
      "learning_rate": 8.547863397548162e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 33396064,
      "step": 2075
    },
    {
      "epoch": 0.14542015813391473,
      "grad_norm": 3.8171703815460205,
      "learning_rate": 8.54716357267951e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 33411640,
      "step": 2076
    },
    {
      "epoch": 0.14549020637964397,
      "grad_norm": 3.56487774848938,
      "learning_rate": 8.546463747810858e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 33428024,
      "step": 2077
    },
    {
      "epoch": 0.14556025462537323,
      "grad_norm": 5.169209003448486,
      "learning_rate": 8.545763922942207e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 33444408,
      "step": 2078
    },
    {
      "epoch": 0.14563030287110248,
      "grad_norm": 3.718086004257202,
      "learning_rate": 8.545064098073556e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 33460416,
      "step": 2079
    },
    {
      "epoch": 0.14570035111683172,
      "grad_norm": 4.2451372146606445,
      "learning_rate": 8.544364273204904e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 33476800,
      "step": 2080
    },
    {
      "epoch": 0.14577039936256098,
      "grad_norm": 3.7441632747650146,
      "learning_rate": 8.543664448336252e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 33492536,
      "step": 2081
    },
    {
      "epoch": 0.1458404476082902,
      "grad_norm": 3.8453383445739746,
      "learning_rate": 8.542964623467601e-05,
      "loss": 1.2527,
      "num_input_tokens_seen": 33508920,
      "step": 2082
    },
    {
      "epoch": 0.14591049585401947,
      "grad_norm": 3.6744494438171387,
      "learning_rate": 8.542264798598949e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 33525304,
      "step": 2083
    },
    {
      "epoch": 0.1459805440997487,
      "grad_norm": 4.209956645965576,
      "learning_rate": 8.541564973730299e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 33541544,
      "step": 2084
    },
    {
      "epoch": 0.14605059234547796,
      "grad_norm": 4.347019672393799,
      "learning_rate": 8.540865148861647e-05,
      "loss": 1.1078,
      "num_input_tokens_seen": 33557928,
      "step": 2085
    },
    {
      "epoch": 0.1461206405912072,
      "grad_norm": 5.323390483856201,
      "learning_rate": 8.540165323992995e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 33573152,
      "step": 2086
    },
    {
      "epoch": 0.14619068883693645,
      "grad_norm": 3.632425308227539,
      "learning_rate": 8.539465499124343e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 33588848,
      "step": 2087
    },
    {
      "epoch": 0.14626073708266568,
      "grad_norm": 4.460893154144287,
      "learning_rate": 8.538765674255692e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 33604984,
      "step": 2088
    },
    {
      "epoch": 0.14633078532839494,
      "grad_norm": 4.059104919433594,
      "learning_rate": 8.53806584938704e-05,
      "loss": 1.1818,
      "num_input_tokens_seen": 33620384,
      "step": 2089
    },
    {
      "epoch": 0.14640083357412417,
      "grad_norm": 6.023964881896973,
      "learning_rate": 8.53736602451839e-05,
      "loss": 1.1589,
      "num_input_tokens_seen": 33636416,
      "step": 2090
    },
    {
      "epoch": 0.14647088181985343,
      "grad_norm": 4.462921619415283,
      "learning_rate": 8.536666199649738e-05,
      "loss": 0.9362,
      "num_input_tokens_seen": 33652504,
      "step": 2091
    },
    {
      "epoch": 0.14654093006558266,
      "grad_norm": 4.003902435302734,
      "learning_rate": 8.535966374781087e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 33668888,
      "step": 2092
    },
    {
      "epoch": 0.14661097831131192,
      "grad_norm": 4.161351680755615,
      "learning_rate": 8.535266549912435e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 33685272,
      "step": 2093
    },
    {
      "epoch": 0.14668102655704116,
      "grad_norm": 4.424163341522217,
      "learning_rate": 8.534566725043784e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 33700872,
      "step": 2094
    },
    {
      "epoch": 0.14675107480277041,
      "grad_norm": 4.2255072593688965,
      "learning_rate": 8.533866900175131e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 33717256,
      "step": 2095
    },
    {
      "epoch": 0.14682112304849965,
      "grad_norm": 4.204975605010986,
      "learning_rate": 8.53316707530648e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 33732544,
      "step": 2096
    },
    {
      "epoch": 0.1468911712942289,
      "grad_norm": 3.7058298587799072,
      "learning_rate": 8.532467250437829e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 33748928,
      "step": 2097
    },
    {
      "epoch": 0.14696121953995814,
      "grad_norm": 6.157133102416992,
      "learning_rate": 8.531767425569178e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 33765312,
      "step": 2098
    },
    {
      "epoch": 0.1470312677856874,
      "grad_norm": 4.3684210777282715,
      "learning_rate": 8.531067600700526e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 33781552,
      "step": 2099
    },
    {
      "epoch": 0.14710131603141663,
      "grad_norm": 3.957848310470581,
      "learning_rate": 8.530367775831874e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 33797464,
      "step": 2100
    },
    {
      "epoch": 0.1471713642771459,
      "grad_norm": 4.9368486404418945,
      "learning_rate": 8.529667950963223e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 33812672,
      "step": 2101
    },
    {
      "epoch": 0.14724141252287512,
      "grad_norm": 3.8515660762786865,
      "learning_rate": 8.528968126094572e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 33829024,
      "step": 2102
    },
    {
      "epoch": 0.14731146076860438,
      "grad_norm": 3.961448907852173,
      "learning_rate": 8.528268301225919e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 33845408,
      "step": 2103
    },
    {
      "epoch": 0.1473815090143336,
      "grad_norm": 4.101677894592285,
      "learning_rate": 8.52756847635727e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 33861240,
      "step": 2104
    },
    {
      "epoch": 0.14745155726006287,
      "grad_norm": 3.886634349822998,
      "learning_rate": 8.526868651488617e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 33876832,
      "step": 2105
    },
    {
      "epoch": 0.1475216055057921,
      "grad_norm": 3.7241156101226807,
      "learning_rate": 8.526168826619966e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 33893216,
      "step": 2106
    },
    {
      "epoch": 0.14759165375152136,
      "grad_norm": 3.829458236694336,
      "learning_rate": 8.525469001751313e-05,
      "loss": 0.8324,
      "num_input_tokens_seen": 33909224,
      "step": 2107
    },
    {
      "epoch": 0.1476617019972506,
      "grad_norm": 5.275660514831543,
      "learning_rate": 8.524769176882662e-05,
      "loss": 1.0253,
      "num_input_tokens_seen": 33924768,
      "step": 2108
    },
    {
      "epoch": 0.14773175024297985,
      "grad_norm": 4.207718372344971,
      "learning_rate": 8.524069352014011e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 33940288,
      "step": 2109
    },
    {
      "epoch": 0.14780179848870909,
      "grad_norm": 4.205242156982422,
      "learning_rate": 8.52336952714536e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 33956512,
      "step": 2110
    },
    {
      "epoch": 0.14787184673443834,
      "grad_norm": 4.365423202514648,
      "learning_rate": 8.522669702276709e-05,
      "loss": 1.2043,
      "num_input_tokens_seen": 33972896,
      "step": 2111
    },
    {
      "epoch": 0.1479418949801676,
      "grad_norm": 4.416136741638184,
      "learning_rate": 8.521969877408056e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 33989280,
      "step": 2112
    },
    {
      "epoch": 0.14801194322589684,
      "grad_norm": 4.375226974487305,
      "learning_rate": 8.521270052539405e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 34005664,
      "step": 2113
    },
    {
      "epoch": 0.1480819914716261,
      "grad_norm": 5.2603840827941895,
      "learning_rate": 8.520570227670753e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 34021576,
      "step": 2114
    },
    {
      "epoch": 0.14815203971735533,
      "grad_norm": 4.02445125579834,
      "learning_rate": 8.519870402802101e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 34037960,
      "step": 2115
    },
    {
      "epoch": 0.1482220879630846,
      "grad_norm": 3.6527910232543945,
      "learning_rate": 8.51917057793345e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 34053240,
      "step": 2116
    },
    {
      "epoch": 0.14829213620881382,
      "grad_norm": 4.170680999755859,
      "learning_rate": 8.518470753064799e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 34068896,
      "step": 2117
    },
    {
      "epoch": 0.14836218445454308,
      "grad_norm": 4.366664886474609,
      "learning_rate": 8.517770928196148e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 34085280,
      "step": 2118
    },
    {
      "epoch": 0.1484322327002723,
      "grad_norm": 3.50757098197937,
      "learning_rate": 8.517071103327497e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 34101664,
      "step": 2119
    },
    {
      "epoch": 0.14850228094600157,
      "grad_norm": 4.607417106628418,
      "learning_rate": 8.516371278458844e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 34117752,
      "step": 2120
    },
    {
      "epoch": 0.1485723291917308,
      "grad_norm": 3.959874391555786,
      "learning_rate": 8.515671453590193e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 34133576,
      "step": 2121
    },
    {
      "epoch": 0.14864237743746006,
      "grad_norm": 4.708366870880127,
      "learning_rate": 8.514971628721541e-05,
      "loss": 1.1201,
      "num_input_tokens_seen": 34149952,
      "step": 2122
    },
    {
      "epoch": 0.1487124256831893,
      "grad_norm": 3.6237339973449707,
      "learning_rate": 8.51427180385289e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 34166336,
      "step": 2123
    },
    {
      "epoch": 0.14878247392891855,
      "grad_norm": 4.606329917907715,
      "learning_rate": 8.513571978984238e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 34181128,
      "step": 2124
    },
    {
      "epoch": 0.14885252217464778,
      "grad_norm": 4.559760093688965,
      "learning_rate": 8.512872154115587e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 34197512,
      "step": 2125
    },
    {
      "epoch": 0.14892257042037704,
      "grad_norm": 3.870089292526245,
      "learning_rate": 8.512172329246936e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 34213896,
      "step": 2126
    },
    {
      "epoch": 0.14899261866610627,
      "grad_norm": 4.441296100616455,
      "learning_rate": 8.511472504378284e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 34229472,
      "step": 2127
    },
    {
      "epoch": 0.14906266691183553,
      "grad_norm": 3.8565545082092285,
      "learning_rate": 8.510772679509633e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 34245856,
      "step": 2128
    },
    {
      "epoch": 0.14913271515756477,
      "grad_norm": 3.563889741897583,
      "learning_rate": 8.510072854640981e-05,
      "loss": 0.8612,
      "num_input_tokens_seen": 34262240,
      "step": 2129
    },
    {
      "epoch": 0.14920276340329403,
      "grad_norm": 4.2634429931640625,
      "learning_rate": 8.50937302977233e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 34278624,
      "step": 2130
    },
    {
      "epoch": 0.14927281164902326,
      "grad_norm": 5.418450355529785,
      "learning_rate": 8.508673204903679e-05,
      "loss": 0.9869,
      "num_input_tokens_seen": 34294216,
      "step": 2131
    },
    {
      "epoch": 0.14934285989475252,
      "grad_norm": 3.511815309524536,
      "learning_rate": 8.507973380035027e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 34310592,
      "step": 2132
    },
    {
      "epoch": 0.14941290814048175,
      "grad_norm": 4.088070392608643,
      "learning_rate": 8.507273555166375e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 34326352,
      "step": 2133
    },
    {
      "epoch": 0.149482956386211,
      "grad_norm": 3.8594932556152344,
      "learning_rate": 8.506573730297723e-05,
      "loss": 1.0281,
      "num_input_tokens_seen": 34342512,
      "step": 2134
    },
    {
      "epoch": 0.14955300463194024,
      "grad_norm": 5.410063743591309,
      "learning_rate": 8.505873905429072e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 34358896,
      "step": 2135
    },
    {
      "epoch": 0.1496230528776695,
      "grad_norm": 4.02821159362793,
      "learning_rate": 8.505174080560421e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 34375280,
      "step": 2136
    },
    {
      "epoch": 0.14969310112339873,
      "grad_norm": 3.866480827331543,
      "learning_rate": 8.50447425569177e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 34391584,
      "step": 2137
    },
    {
      "epoch": 0.149763149369128,
      "grad_norm": 3.667064905166626,
      "learning_rate": 8.503774430823118e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 34407264,
      "step": 2138
    },
    {
      "epoch": 0.14983319761485722,
      "grad_norm": 5.41308069229126,
      "learning_rate": 8.503074605954466e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 34423648,
      "step": 2139
    },
    {
      "epoch": 0.14990324586058648,
      "grad_norm": 4.1716485023498535,
      "learning_rate": 8.502374781085815e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 34439512,
      "step": 2140
    },
    {
      "epoch": 0.1499732941063157,
      "grad_norm": 4.1403913497924805,
      "learning_rate": 8.501674956217162e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 34455896,
      "step": 2141
    },
    {
      "epoch": 0.15004334235204497,
      "grad_norm": 3.75219988822937,
      "learning_rate": 8.500975131348511e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 34472280,
      "step": 2142
    },
    {
      "epoch": 0.1501133905977742,
      "grad_norm": 4.339532852172852,
      "learning_rate": 8.50027530647986e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 34488664,
      "step": 2143
    },
    {
      "epoch": 0.15018343884350346,
      "grad_norm": 4.259124755859375,
      "learning_rate": 8.499575481611209e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 34505048,
      "step": 2144
    },
    {
      "epoch": 0.1502534870892327,
      "grad_norm": 5.031396865844727,
      "learning_rate": 8.498875656742558e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 34521432,
      "step": 2145
    },
    {
      "epoch": 0.15032353533496196,
      "grad_norm": 5.313172340393066,
      "learning_rate": 8.498175831873907e-05,
      "loss": 1.1737,
      "num_input_tokens_seen": 34536344,
      "step": 2146
    },
    {
      "epoch": 0.15039358358069121,
      "grad_norm": 4.844740390777588,
      "learning_rate": 8.497476007005254e-05,
      "loss": 1.4095,
      "num_input_tokens_seen": 34552728,
      "step": 2147
    },
    {
      "epoch": 0.15046363182642045,
      "grad_norm": 4.231154441833496,
      "learning_rate": 8.496776182136603e-05,
      "loss": 1.1196,
      "num_input_tokens_seen": 34569016,
      "step": 2148
    },
    {
      "epoch": 0.1505336800721497,
      "grad_norm": 4.176802635192871,
      "learning_rate": 8.49607635726795e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 34585376,
      "step": 2149
    },
    {
      "epoch": 0.15060372831787894,
      "grad_norm": 4.710334777832031,
      "learning_rate": 8.4953765323993e-05,
      "loss": 1.0085,
      "num_input_tokens_seen": 34600400,
      "step": 2150
    },
    {
      "epoch": 0.1506737765636082,
      "grad_norm": 3.9053258895874023,
      "learning_rate": 8.494676707530648e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 34616688,
      "step": 2151
    },
    {
      "epoch": 0.15074382480933743,
      "grad_norm": 4.043003559112549,
      "learning_rate": 8.493976882661997e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 34631920,
      "step": 2152
    },
    {
      "epoch": 0.1508138730550667,
      "grad_norm": 5.230721473693848,
      "learning_rate": 8.493277057793346e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 34648128,
      "step": 2153
    },
    {
      "epoch": 0.15088392130079592,
      "grad_norm": 4.098349094390869,
      "learning_rate": 8.492577232924693e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 34664512,
      "step": 2154
    },
    {
      "epoch": 0.15095396954652518,
      "grad_norm": 4.803813457489014,
      "learning_rate": 8.491877408056042e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 34680560,
      "step": 2155
    },
    {
      "epoch": 0.1510240177922544,
      "grad_norm": 4.25751256942749,
      "learning_rate": 8.491177583187391e-05,
      "loss": 1.2481,
      "num_input_tokens_seen": 34696944,
      "step": 2156
    },
    {
      "epoch": 0.15109406603798367,
      "grad_norm": 6.600613117218018,
      "learning_rate": 8.49047775831874e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 34712416,
      "step": 2157
    },
    {
      "epoch": 0.1511641142837129,
      "grad_norm": 5.649744987487793,
      "learning_rate": 8.489777933450087e-05,
      "loss": 1.3045,
      "num_input_tokens_seen": 34728520,
      "step": 2158
    },
    {
      "epoch": 0.15123416252944216,
      "grad_norm": 5.778639316558838,
      "learning_rate": 8.489078108581436e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 34744776,
      "step": 2159
    },
    {
      "epoch": 0.1513042107751714,
      "grad_norm": 5.944733619689941,
      "learning_rate": 8.488378283712785e-05,
      "loss": 1.3293,
      "num_input_tokens_seen": 34761160,
      "step": 2160
    },
    {
      "epoch": 0.15137425902090065,
      "grad_norm": 3.7783594131469727,
      "learning_rate": 8.487678458844133e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 34777544,
      "step": 2161
    },
    {
      "epoch": 0.15144430726662989,
      "grad_norm": 5.126344680786133,
      "learning_rate": 8.486978633975482e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 34793072,
      "step": 2162
    },
    {
      "epoch": 0.15151435551235914,
      "grad_norm": 4.689150333404541,
      "learning_rate": 8.48627880910683e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 34809456,
      "step": 2163
    },
    {
      "epoch": 0.15158440375808838,
      "grad_norm": 3.7559547424316406,
      "learning_rate": 8.485578984238179e-05,
      "loss": 1.1414,
      "num_input_tokens_seen": 34825208,
      "step": 2164
    },
    {
      "epoch": 0.15165445200381764,
      "grad_norm": 3.9225172996520996,
      "learning_rate": 8.484879159369528e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 34841592,
      "step": 2165
    },
    {
      "epoch": 0.15172450024954687,
      "grad_norm": 4.264125347137451,
      "learning_rate": 8.484179334500876e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 34857928,
      "step": 2166
    },
    {
      "epoch": 0.15179454849527613,
      "grad_norm": 4.0784382820129395,
      "learning_rate": 8.483479509632224e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 34873224,
      "step": 2167
    },
    {
      "epoch": 0.15186459674100536,
      "grad_norm": 4.371130466461182,
      "learning_rate": 8.482779684763572e-05,
      "loss": 1.3854,
      "num_input_tokens_seen": 34889608,
      "step": 2168
    },
    {
      "epoch": 0.15193464498673462,
      "grad_norm": 3.7022883892059326,
      "learning_rate": 8.482079859894921e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 34905984,
      "step": 2169
    },
    {
      "epoch": 0.15200469323246385,
      "grad_norm": 4.196985721588135,
      "learning_rate": 8.481380035026271e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 34922368,
      "step": 2170
    },
    {
      "epoch": 0.1520747414781931,
      "grad_norm": 4.0252580642700195,
      "learning_rate": 8.480680210157619e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 34938752,
      "step": 2171
    },
    {
      "epoch": 0.15214478972392234,
      "grad_norm": 4.03692626953125,
      "learning_rate": 8.479980385288967e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 34954176,
      "step": 2172
    },
    {
      "epoch": 0.1522148379696516,
      "grad_norm": 4.183175563812256,
      "learning_rate": 8.479280560420316e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 34969880,
      "step": 2173
    },
    {
      "epoch": 0.15228488621538083,
      "grad_norm": 3.757636070251465,
      "learning_rate": 8.478580735551664e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 34985576,
      "step": 2174
    },
    {
      "epoch": 0.1523549344611101,
      "grad_norm": 3.9442903995513916,
      "learning_rate": 8.477880910683013e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 35001896,
      "step": 2175
    },
    {
      "epoch": 0.15242498270683932,
      "grad_norm": 4.092566013336182,
      "learning_rate": 8.477181085814362e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 35018280,
      "step": 2176
    },
    {
      "epoch": 0.15249503095256858,
      "grad_norm": 5.494921684265137,
      "learning_rate": 8.47648126094571e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 35034544,
      "step": 2177
    },
    {
      "epoch": 0.15256507919829781,
      "grad_norm": 7.327289581298828,
      "learning_rate": 8.475781436077058e-05,
      "loss": 1.1879,
      "num_input_tokens_seen": 35050928,
      "step": 2178
    },
    {
      "epoch": 0.15263512744402707,
      "grad_norm": 4.048150539398193,
      "learning_rate": 8.475081611208407e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 35067000,
      "step": 2179
    },
    {
      "epoch": 0.1527051756897563,
      "grad_norm": 6.388006210327148,
      "learning_rate": 8.474381786339756e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 35082064,
      "step": 2180
    },
    {
      "epoch": 0.15277522393548557,
      "grad_norm": 4.289052963256836,
      "learning_rate": 8.473681961471103e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 35098448,
      "step": 2181
    },
    {
      "epoch": 0.15284527218121483,
      "grad_norm": 4.288560390472412,
      "learning_rate": 8.472982136602452e-05,
      "loss": 1.2723,
      "num_input_tokens_seen": 35114832,
      "step": 2182
    },
    {
      "epoch": 0.15291532042694406,
      "grad_norm": 4.17701530456543,
      "learning_rate": 8.472282311733801e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 35131216,
      "step": 2183
    },
    {
      "epoch": 0.15298536867267332,
      "grad_norm": 4.975949764251709,
      "learning_rate": 8.47158248686515e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 35147600,
      "step": 2184
    },
    {
      "epoch": 0.15305541691840255,
      "grad_norm": 5.465437889099121,
      "learning_rate": 8.470882661996497e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 35162464,
      "step": 2185
    },
    {
      "epoch": 0.1531254651641318,
      "grad_norm": 3.329401731491089,
      "learning_rate": 8.470182837127846e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 35178744,
      "step": 2186
    },
    {
      "epoch": 0.15319551340986104,
      "grad_norm": 5.962124824523926,
      "learning_rate": 8.469483012259195e-05,
      "loss": 1.2799,
      "num_input_tokens_seen": 35194736,
      "step": 2187
    },
    {
      "epoch": 0.1532655616555903,
      "grad_norm": 3.897841691970825,
      "learning_rate": 8.468783187390542e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 35211120,
      "step": 2188
    },
    {
      "epoch": 0.15333560990131953,
      "grad_norm": 3.9668943881988525,
      "learning_rate": 8.468083362521891e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 35227504,
      "step": 2189
    },
    {
      "epoch": 0.1534056581470488,
      "grad_norm": 3.8960444927215576,
      "learning_rate": 8.467383537653241e-05,
      "loss": 0.891,
      "num_input_tokens_seen": 35243584,
      "step": 2190
    },
    {
      "epoch": 0.15347570639277802,
      "grad_norm": 3.7700982093811035,
      "learning_rate": 8.466683712784589e-05,
      "loss": 1.1744,
      "num_input_tokens_seen": 35259968,
      "step": 2191
    },
    {
      "epoch": 0.15354575463850728,
      "grad_norm": 4.65008020401001,
      "learning_rate": 8.465983887915938e-05,
      "loss": 1.2807,
      "num_input_tokens_seen": 35276352,
      "step": 2192
    },
    {
      "epoch": 0.1536158028842365,
      "grad_norm": 3.5371146202087402,
      "learning_rate": 8.465284063047285e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 35292736,
      "step": 2193
    },
    {
      "epoch": 0.15368585112996577,
      "grad_norm": 4.395732879638672,
      "learning_rate": 8.464584238178634e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 35309120,
      "step": 2194
    },
    {
      "epoch": 0.153755899375695,
      "grad_norm": 5.01919412612915,
      "learning_rate": 8.463884413309982e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 35325504,
      "step": 2195
    },
    {
      "epoch": 0.15382594762142426,
      "grad_norm": 3.7417054176330566,
      "learning_rate": 8.463184588441332e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 35341376,
      "step": 2196
    },
    {
      "epoch": 0.1538959958671535,
      "grad_norm": 4.119459629058838,
      "learning_rate": 8.462484763572681e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 35357520,
      "step": 2197
    },
    {
      "epoch": 0.15396604411288276,
      "grad_norm": 6.938751220703125,
      "learning_rate": 8.461784938704028e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 35372920,
      "step": 2198
    },
    {
      "epoch": 0.154036092358612,
      "grad_norm": 5.000339984893799,
      "learning_rate": 8.461085113835377e-05,
      "loss": 1.1508,
      "num_input_tokens_seen": 35389304,
      "step": 2199
    },
    {
      "epoch": 0.15410614060434125,
      "grad_norm": 3.6554362773895264,
      "learning_rate": 8.460385288966726e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 35405688,
      "step": 2200
    },
    {
      "epoch": 0.15410614060434125,
      "eval_loss": 1.145054578781128,
      "eval_runtime": 0.1886,
      "eval_samples_per_second": 5.303,
      "eval_steps_per_second": 5.303,
      "num_input_tokens_seen": 35405688,
      "step": 2200
    },
    {
      "epoch": 0.15417618885007048,
      "grad_norm": 3.718207836151123,
      "learning_rate": 8.459685464098073e-05,
      "loss": 0.8814,
      "num_input_tokens_seen": 35422072,
      "step": 2201
    },
    {
      "epoch": 0.15424623709579974,
      "grad_norm": 4.98813533782959,
      "learning_rate": 8.458985639229422e-05,
      "loss": 1.1814,
      "num_input_tokens_seen": 35438456,
      "step": 2202
    },
    {
      "epoch": 0.15431628534152897,
      "grad_norm": 3.550008535385132,
      "learning_rate": 8.458285814360771e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 35454840,
      "step": 2203
    },
    {
      "epoch": 0.15438633358725823,
      "grad_norm": 3.8408641815185547,
      "learning_rate": 8.45758598949212e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 35471080,
      "step": 2204
    },
    {
      "epoch": 0.15445638183298746,
      "grad_norm": 4.515852451324463,
      "learning_rate": 8.456886164623468e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 35486904,
      "step": 2205
    },
    {
      "epoch": 0.15452643007871672,
      "grad_norm": 3.6536715030670166,
      "learning_rate": 8.456186339754816e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 35503064,
      "step": 2206
    },
    {
      "epoch": 0.15459647832444595,
      "grad_norm": 4.071808338165283,
      "learning_rate": 8.455486514886165e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 35518880,
      "step": 2207
    },
    {
      "epoch": 0.1546665265701752,
      "grad_norm": 4.329566955566406,
      "learning_rate": 8.454786690017513e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 35535256,
      "step": 2208
    },
    {
      "epoch": 0.15473657481590444,
      "grad_norm": 4.243298053741455,
      "learning_rate": 8.454086865148862e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 35551376,
      "step": 2209
    },
    {
      "epoch": 0.1548066230616337,
      "grad_norm": 4.154253959655762,
      "learning_rate": 8.453387040280212e-05,
      "loss": 1.0458,
      "num_input_tokens_seen": 35567696,
      "step": 2210
    },
    {
      "epoch": 0.15487667130736293,
      "grad_norm": 4.0564494132995605,
      "learning_rate": 8.45268721541156e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 35583576,
      "step": 2211
    },
    {
      "epoch": 0.1549467195530922,
      "grad_norm": 3.735724687576294,
      "learning_rate": 8.451987390542907e-05,
      "loss": 0.92,
      "num_input_tokens_seen": 35599536,
      "step": 2212
    },
    {
      "epoch": 0.15501676779882143,
      "grad_norm": 4.651454925537109,
      "learning_rate": 8.451287565674256e-05,
      "loss": 1.2097,
      "num_input_tokens_seen": 35615920,
      "step": 2213
    },
    {
      "epoch": 0.15508681604455068,
      "grad_norm": 5.01883602142334,
      "learning_rate": 8.450587740805605e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 35631208,
      "step": 2214
    },
    {
      "epoch": 0.15515686429027992,
      "grad_norm": 4.435250282287598,
      "learning_rate": 8.449887915936952e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 35647328,
      "step": 2215
    },
    {
      "epoch": 0.15522691253600918,
      "grad_norm": 3.495476245880127,
      "learning_rate": 8.449188091068302e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 35663472,
      "step": 2216
    },
    {
      "epoch": 0.15529696078173844,
      "grad_norm": 4.461013317108154,
      "learning_rate": 8.448488266199651e-05,
      "loss": 1.1098,
      "num_input_tokens_seen": 35679856,
      "step": 2217
    },
    {
      "epoch": 0.15536700902746767,
      "grad_norm": 5.4857683181762695,
      "learning_rate": 8.447788441330999e-05,
      "loss": 1.143,
      "num_input_tokens_seen": 35695616,
      "step": 2218
    },
    {
      "epoch": 0.15543705727319693,
      "grad_norm": 4.20158052444458,
      "learning_rate": 8.447088616462348e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 35711432,
      "step": 2219
    },
    {
      "epoch": 0.15550710551892616,
      "grad_norm": 4.289988040924072,
      "learning_rate": 8.446388791593695e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 35727552,
      "step": 2220
    },
    {
      "epoch": 0.15557715376465542,
      "grad_norm": 3.7897555828094482,
      "learning_rate": 8.445688966725044e-05,
      "loss": 1.255,
      "num_input_tokens_seen": 35743800,
      "step": 2221
    },
    {
      "epoch": 0.15564720201038465,
      "grad_norm": 4.405816078186035,
      "learning_rate": 8.444989141856393e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 35760184,
      "step": 2222
    },
    {
      "epoch": 0.1557172502561139,
      "grad_norm": 4.2683610916137695,
      "learning_rate": 8.444289316987742e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 35776568,
      "step": 2223
    },
    {
      "epoch": 0.15578729850184314,
      "grad_norm": 3.9999659061431885,
      "learning_rate": 8.44358949211909e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 35792952,
      "step": 2224
    },
    {
      "epoch": 0.1558573467475724,
      "grad_norm": 3.6252965927124023,
      "learning_rate": 8.442889667250438e-05,
      "loss": 0.9755,
      "num_input_tokens_seen": 35809176,
      "step": 2225
    },
    {
      "epoch": 0.15592739499330163,
      "grad_norm": 3.9726274013519287,
      "learning_rate": 8.442189842381787e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 35825560,
      "step": 2226
    },
    {
      "epoch": 0.1559974432390309,
      "grad_norm": 5.004739761352539,
      "learning_rate": 8.441490017513136e-05,
      "loss": 1.2484,
      "num_input_tokens_seen": 35841936,
      "step": 2227
    },
    {
      "epoch": 0.15606749148476012,
      "grad_norm": 5.432271480560303,
      "learning_rate": 8.440790192644483e-05,
      "loss": 0.9799,
      "num_input_tokens_seen": 35857944,
      "step": 2228
    },
    {
      "epoch": 0.15613753973048938,
      "grad_norm": 4.553518295288086,
      "learning_rate": 8.440090367775832e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 35873920,
      "step": 2229
    },
    {
      "epoch": 0.15620758797621861,
      "grad_norm": 5.924668312072754,
      "learning_rate": 8.439390542907181e-05,
      "loss": 1.2937,
      "num_input_tokens_seen": 35888872,
      "step": 2230
    },
    {
      "epoch": 0.15627763622194787,
      "grad_norm": 4.276167392730713,
      "learning_rate": 8.43869071803853e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 35905256,
      "step": 2231
    },
    {
      "epoch": 0.1563476844676771,
      "grad_norm": 3.719632863998413,
      "learning_rate": 8.437990893169877e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 35921640,
      "step": 2232
    },
    {
      "epoch": 0.15641773271340637,
      "grad_norm": 4.769368648529053,
      "learning_rate": 8.437291068301226e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 35936256,
      "step": 2233
    },
    {
      "epoch": 0.1564877809591356,
      "grad_norm": 4.957282543182373,
      "learning_rate": 8.436591243432575e-05,
      "loss": 1.0535,
      "num_input_tokens_seen": 35952640,
      "step": 2234
    },
    {
      "epoch": 0.15655782920486486,
      "grad_norm": 4.782018661499023,
      "learning_rate": 8.435891418563923e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 35967880,
      "step": 2235
    },
    {
      "epoch": 0.1566278774505941,
      "grad_norm": 4.716582775115967,
      "learning_rate": 8.435191593695273e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 35984016,
      "step": 2236
    },
    {
      "epoch": 0.15669792569632335,
      "grad_norm": 4.36606502532959,
      "learning_rate": 8.434491768826622e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 35999904,
      "step": 2237
    },
    {
      "epoch": 0.15676797394205258,
      "grad_norm": 3.8300321102142334,
      "learning_rate": 8.433791943957969e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 36016216,
      "step": 2238
    },
    {
      "epoch": 0.15683802218778184,
      "grad_norm": 3.7595677375793457,
      "learning_rate": 8.433092119089317e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 36032600,
      "step": 2239
    },
    {
      "epoch": 0.15690807043351107,
      "grad_norm": 4.783555030822754,
      "learning_rate": 8.432392294220665e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 36048984,
      "step": 2240
    },
    {
      "epoch": 0.15697811867924033,
      "grad_norm": 4.393221855163574,
      "learning_rate": 8.431692469352014e-05,
      "loss": 1.2196,
      "num_input_tokens_seen": 36065368,
      "step": 2241
    },
    {
      "epoch": 0.15704816692496956,
      "grad_norm": 3.8634722232818604,
      "learning_rate": 8.430992644483363e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 36081752,
      "step": 2242
    },
    {
      "epoch": 0.15711821517069882,
      "grad_norm": 4.5091233253479,
      "learning_rate": 8.430292819614712e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 36097672,
      "step": 2243
    },
    {
      "epoch": 0.15718826341642805,
      "grad_norm": 3.89699387550354,
      "learning_rate": 8.429592994746061e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 36114048,
      "step": 2244
    },
    {
      "epoch": 0.1572583116621573,
      "grad_norm": 3.8859546184539795,
      "learning_rate": 8.428893169877408e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 36130024,
      "step": 2245
    },
    {
      "epoch": 0.15732835990788654,
      "grad_norm": 4.236848831176758,
      "learning_rate": 8.428193345008757e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 36146408,
      "step": 2246
    },
    {
      "epoch": 0.1573984081536158,
      "grad_norm": 6.742307662963867,
      "learning_rate": 8.427493520140105e-05,
      "loss": 0.9674,
      "num_input_tokens_seen": 36161440,
      "step": 2247
    },
    {
      "epoch": 0.15746845639934504,
      "grad_norm": 3.332416534423828,
      "learning_rate": 8.426793695271454e-05,
      "loss": 0.7694,
      "num_input_tokens_seen": 36177824,
      "step": 2248
    },
    {
      "epoch": 0.1575385046450743,
      "grad_norm": 4.672734260559082,
      "learning_rate": 8.426093870402802e-05,
      "loss": 0.9228,
      "num_input_tokens_seen": 36193320,
      "step": 2249
    },
    {
      "epoch": 0.15760855289080353,
      "grad_norm": 4.437155246734619,
      "learning_rate": 8.425394045534151e-05,
      "loss": 1.2712,
      "num_input_tokens_seen": 36209704,
      "step": 2250
    },
    {
      "epoch": 0.1576786011365328,
      "grad_norm": 4.112512111663818,
      "learning_rate": 8.4246942206655e-05,
      "loss": 1.3494,
      "num_input_tokens_seen": 36226088,
      "step": 2251
    },
    {
      "epoch": 0.15774864938226205,
      "grad_norm": 4.432194709777832,
      "learning_rate": 8.423994395796848e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 36242472,
      "step": 2252
    },
    {
      "epoch": 0.15781869762799128,
      "grad_norm": 4.322375297546387,
      "learning_rate": 8.423294570928197e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 36258680,
      "step": 2253
    },
    {
      "epoch": 0.15788874587372054,
      "grad_norm": 3.848836660385132,
      "learning_rate": 8.422594746059545e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 36274512,
      "step": 2254
    },
    {
      "epoch": 0.15795879411944977,
      "grad_norm": 4.022729396820068,
      "learning_rate": 8.421894921190893e-05,
      "loss": 1.0584,
      "num_input_tokens_seen": 36289568,
      "step": 2255
    },
    {
      "epoch": 0.15802884236517903,
      "grad_norm": 3.8060622215270996,
      "learning_rate": 8.421195096322243e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 36305256,
      "step": 2256
    },
    {
      "epoch": 0.15809889061090826,
      "grad_norm": 4.685004234313965,
      "learning_rate": 8.42049527145359e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 36321008,
      "step": 2257
    },
    {
      "epoch": 0.15816893885663752,
      "grad_norm": 3.4483463764190674,
      "learning_rate": 8.41979544658494e-05,
      "loss": 0.9563,
      "num_input_tokens_seen": 36337000,
      "step": 2258
    },
    {
      "epoch": 0.15823898710236675,
      "grad_norm": 3.7172203063964844,
      "learning_rate": 8.419095621716287e-05,
      "loss": 1.1463,
      "num_input_tokens_seen": 36353160,
      "step": 2259
    },
    {
      "epoch": 0.158309035348096,
      "grad_norm": 5.734589099884033,
      "learning_rate": 8.418395796847636e-05,
      "loss": 0.9321,
      "num_input_tokens_seen": 36369248,
      "step": 2260
    },
    {
      "epoch": 0.15837908359382524,
      "grad_norm": 4.060257911682129,
      "learning_rate": 8.417695971978985e-05,
      "loss": 1.2162,
      "num_input_tokens_seen": 36384736,
      "step": 2261
    },
    {
      "epoch": 0.1584491318395545,
      "grad_norm": 5.240515232086182,
      "learning_rate": 8.416996147110334e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 36401120,
      "step": 2262
    },
    {
      "epoch": 0.15851918008528373,
      "grad_norm": 5.482649803161621,
      "learning_rate": 8.416296322241682e-05,
      "loss": 1.207,
      "num_input_tokens_seen": 36417504,
      "step": 2263
    },
    {
      "epoch": 0.158589228331013,
      "grad_norm": 3.9862253665924072,
      "learning_rate": 8.415596497373031e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 36433888,
      "step": 2264
    },
    {
      "epoch": 0.15865927657674223,
      "grad_norm": 6.322808742523193,
      "learning_rate": 8.414896672504379e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 36449552,
      "step": 2265
    },
    {
      "epoch": 0.15872932482247148,
      "grad_norm": 4.312921524047852,
      "learning_rate": 8.414196847635726e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 36465936,
      "step": 2266
    },
    {
      "epoch": 0.15879937306820072,
      "grad_norm": 4.178677082061768,
      "learning_rate": 8.413497022767075e-05,
      "loss": 1.2539,
      "num_input_tokens_seen": 36482184,
      "step": 2267
    },
    {
      "epoch": 0.15886942131392998,
      "grad_norm": 4.304810523986816,
      "learning_rate": 8.412797197898424e-05,
      "loss": 1.199,
      "num_input_tokens_seen": 36498320,
      "step": 2268
    },
    {
      "epoch": 0.1589394695596592,
      "grad_norm": 3.723483085632324,
      "learning_rate": 8.412097373029773e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 36514704,
      "step": 2269
    },
    {
      "epoch": 0.15900951780538847,
      "grad_norm": 4.285789489746094,
      "learning_rate": 8.411397548161122e-05,
      "loss": 1.2463,
      "num_input_tokens_seen": 36531032,
      "step": 2270
    },
    {
      "epoch": 0.1590795660511177,
      "grad_norm": 3.5788466930389404,
      "learning_rate": 8.41069772329247e-05,
      "loss": 0.7809,
      "num_input_tokens_seen": 36547416,
      "step": 2271
    },
    {
      "epoch": 0.15914961429684696,
      "grad_norm": 5.785874366760254,
      "learning_rate": 8.409997898423818e-05,
      "loss": 1.2832,
      "num_input_tokens_seen": 36563800,
      "step": 2272
    },
    {
      "epoch": 0.1592196625425762,
      "grad_norm": 3.914402723312378,
      "learning_rate": 8.409298073555167e-05,
      "loss": 1.2065,
      "num_input_tokens_seen": 36580184,
      "step": 2273
    },
    {
      "epoch": 0.15928971078830545,
      "grad_norm": 3.878512144088745,
      "learning_rate": 8.408598248686514e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 36596568,
      "step": 2274
    },
    {
      "epoch": 0.15935975903403468,
      "grad_norm": 4.195454120635986,
      "learning_rate": 8.407898423817863e-05,
      "loss": 1.2628,
      "num_input_tokens_seen": 36612952,
      "step": 2275
    },
    {
      "epoch": 0.15942980727976394,
      "grad_norm": 3.847649097442627,
      "learning_rate": 8.407198598949212e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 36628752,
      "step": 2276
    },
    {
      "epoch": 0.15949985552549317,
      "grad_norm": 5.284397125244141,
      "learning_rate": 8.406498774080561e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 36645136,
      "step": 2277
    },
    {
      "epoch": 0.15956990377122243,
      "grad_norm": 4.10982084274292,
      "learning_rate": 8.40579894921191e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 36661392,
      "step": 2278
    },
    {
      "epoch": 0.15963995201695166,
      "grad_norm": 3.8282828330993652,
      "learning_rate": 8.405099124343257e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 36676856,
      "step": 2279
    },
    {
      "epoch": 0.15971000026268092,
      "grad_norm": 4.115365028381348,
      "learning_rate": 8.404399299474606e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 36693080,
      "step": 2280
    },
    {
      "epoch": 0.15978004850841016,
      "grad_norm": 3.6131088733673096,
      "learning_rate": 8.403699474605955e-05,
      "loss": 0.8565,
      "num_input_tokens_seen": 36709440,
      "step": 2281
    },
    {
      "epoch": 0.15985009675413941,
      "grad_norm": 3.83146071434021,
      "learning_rate": 8.402999649737304e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 36725496,
      "step": 2282
    },
    {
      "epoch": 0.15992014499986865,
      "grad_norm": 3.8456339836120605,
      "learning_rate": 8.402299824868653e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 36741544,
      "step": 2283
    },
    {
      "epoch": 0.1599901932455979,
      "grad_norm": 3.717014789581299,
      "learning_rate": 8.4016e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 36757928,
      "step": 2284
    },
    {
      "epoch": 0.16006024149132717,
      "grad_norm": 4.3730854988098145,
      "learning_rate": 8.400900175131349e-05,
      "loss": 1.1639,
      "num_input_tokens_seen": 36774144,
      "step": 2285
    },
    {
      "epoch": 0.1601302897370564,
      "grad_norm": 3.6635241508483887,
      "learning_rate": 8.400200350262697e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 36790248,
      "step": 2286
    },
    {
      "epoch": 0.16020033798278566,
      "grad_norm": 3.9058330059051514,
      "learning_rate": 8.399500525394046e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 36806632,
      "step": 2287
    },
    {
      "epoch": 0.1602703862285149,
      "grad_norm": 3.60127854347229,
      "learning_rate": 8.398800700525394e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 36823016,
      "step": 2288
    },
    {
      "epoch": 0.16034043447424415,
      "grad_norm": 5.762889385223389,
      "learning_rate": 8.398100875656743e-05,
      "loss": 0.9572,
      "num_input_tokens_seen": 36838576,
      "step": 2289
    },
    {
      "epoch": 0.16041048271997338,
      "grad_norm": 3.495436191558838,
      "learning_rate": 8.397401050788092e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 36854960,
      "step": 2290
    },
    {
      "epoch": 0.16048053096570264,
      "grad_norm": 4.083384037017822,
      "learning_rate": 8.396701225919441e-05,
      "loss": 1.1724,
      "num_input_tokens_seen": 36870672,
      "step": 2291
    },
    {
      "epoch": 0.16055057921143187,
      "grad_norm": 3.7010245323181152,
      "learning_rate": 8.396001401050788e-05,
      "loss": 0.8871,
      "num_input_tokens_seen": 36887056,
      "step": 2292
    },
    {
      "epoch": 0.16062062745716113,
      "grad_norm": 3.419485330581665,
      "learning_rate": 8.395301576182136e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 36903144,
      "step": 2293
    },
    {
      "epoch": 0.16069067570289036,
      "grad_norm": 3.593970537185669,
      "learning_rate": 8.394601751313485e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 36919192,
      "step": 2294
    },
    {
      "epoch": 0.16076072394861962,
      "grad_norm": 3.729038953781128,
      "learning_rate": 8.393901926444834e-05,
      "loss": 1.288,
      "num_input_tokens_seen": 36935576,
      "step": 2295
    },
    {
      "epoch": 0.16083077219434885,
      "grad_norm": 3.60687255859375,
      "learning_rate": 8.393202101576183e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 36951960,
      "step": 2296
    },
    {
      "epoch": 0.1609008204400781,
      "grad_norm": 3.4520435333251953,
      "learning_rate": 8.392502276707531e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 36968344,
      "step": 2297
    },
    {
      "epoch": 0.16097086868580734,
      "grad_norm": 3.71907639503479,
      "learning_rate": 8.39180245183888e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 36984440,
      "step": 2298
    },
    {
      "epoch": 0.1610409169315366,
      "grad_norm": 3.8897864818573,
      "learning_rate": 8.391102626970228e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 37000824,
      "step": 2299
    },
    {
      "epoch": 0.16111096517726584,
      "grad_norm": 3.579921245574951,
      "learning_rate": 8.390402802101577e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 37017088,
      "step": 2300
    },
    {
      "epoch": 0.1611810134229951,
      "grad_norm": 3.9658427238464355,
      "learning_rate": 8.389702977232924e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 37033232,
      "step": 2301
    },
    {
      "epoch": 0.16125106166872433,
      "grad_norm": 4.2862725257873535,
      "learning_rate": 8.389003152364274e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 37049616,
      "step": 2302
    },
    {
      "epoch": 0.1613211099144536,
      "grad_norm": 3.7523694038391113,
      "learning_rate": 8.388303327495622e-05,
      "loss": 1.0806,
      "num_input_tokens_seen": 37065784,
      "step": 2303
    },
    {
      "epoch": 0.16139115816018282,
      "grad_norm": 3.9068679809570312,
      "learning_rate": 8.387603502626971e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 37082168,
      "step": 2304
    },
    {
      "epoch": 0.16146120640591208,
      "grad_norm": 5.876891613006592,
      "learning_rate": 8.38690367775832e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 37097072,
      "step": 2305
    },
    {
      "epoch": 0.1615312546516413,
      "grad_norm": 4.040335655212402,
      "learning_rate": 8.386203852889667e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 37112936,
      "step": 2306
    },
    {
      "epoch": 0.16160130289737057,
      "grad_norm": 3.566763401031494,
      "learning_rate": 8.385504028021016e-05,
      "loss": 0.9164,
      "num_input_tokens_seen": 37129320,
      "step": 2307
    },
    {
      "epoch": 0.1616713511430998,
      "grad_norm": 3.7780325412750244,
      "learning_rate": 8.384804203152365e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 37144832,
      "step": 2308
    },
    {
      "epoch": 0.16174139938882906,
      "grad_norm": 4.291510105133057,
      "learning_rate": 8.384104378283714e-05,
      "loss": 1.2579,
      "num_input_tokens_seen": 37160312,
      "step": 2309
    },
    {
      "epoch": 0.1618114476345583,
      "grad_norm": 3.721531629562378,
      "learning_rate": 8.383404553415063e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 37176696,
      "step": 2310
    },
    {
      "epoch": 0.16188149588028755,
      "grad_norm": 3.883301258087158,
      "learning_rate": 8.38270472854641e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 37192632,
      "step": 2311
    },
    {
      "epoch": 0.16195154412601678,
      "grad_norm": 4.240591049194336,
      "learning_rate": 8.382004903677759e-05,
      "loss": 1.2263,
      "num_input_tokens_seen": 37208448,
      "step": 2312
    },
    {
      "epoch": 0.16202159237174604,
      "grad_norm": 4.711728572845459,
      "learning_rate": 8.381305078809106e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 37223176,
      "step": 2313
    },
    {
      "epoch": 0.16209164061747527,
      "grad_norm": 4.733399391174316,
      "learning_rate": 8.380605253940455e-05,
      "loss": 1.2512,
      "num_input_tokens_seen": 37239560,
      "step": 2314
    },
    {
      "epoch": 0.16216168886320453,
      "grad_norm": 5.842257976531982,
      "learning_rate": 8.379905429071804e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 37255536,
      "step": 2315
    },
    {
      "epoch": 0.16223173710893377,
      "grad_norm": 3.9891135692596436,
      "learning_rate": 8.379205604203153e-05,
      "loss": 1.0767,
      "num_input_tokens_seen": 37271920,
      "step": 2316
    },
    {
      "epoch": 0.16230178535466303,
      "grad_norm": 3.3596630096435547,
      "learning_rate": 8.378505779334502e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 37288024,
      "step": 2317
    },
    {
      "epoch": 0.16237183360039226,
      "grad_norm": 4.346104621887207,
      "learning_rate": 8.377805954465851e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 37304264,
      "step": 2318
    },
    {
      "epoch": 0.16244188184612152,
      "grad_norm": 3.5524039268493652,
      "learning_rate": 8.377106129597198e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 37320648,
      "step": 2319
    },
    {
      "epoch": 0.16251193009185078,
      "grad_norm": 4.335781574249268,
      "learning_rate": 8.376406304728546e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 37336104,
      "step": 2320
    },
    {
      "epoch": 0.16258197833758,
      "grad_norm": 3.7356534004211426,
      "learning_rate": 8.375706479859895e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 37352488,
      "step": 2321
    },
    {
      "epoch": 0.16265202658330927,
      "grad_norm": 3.842710494995117,
      "learning_rate": 8.375006654991243e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 37368872,
      "step": 2322
    },
    {
      "epoch": 0.1627220748290385,
      "grad_norm": 3.9485390186309814,
      "learning_rate": 8.374306830122592e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 37385256,
      "step": 2323
    },
    {
      "epoch": 0.16279212307476776,
      "grad_norm": 3.9196622371673584,
      "learning_rate": 8.373607005253941e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 37401224,
      "step": 2324
    },
    {
      "epoch": 0.162862171320497,
      "grad_norm": 4.2444844245910645,
      "learning_rate": 8.37290718038529e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 37417016,
      "step": 2325
    },
    {
      "epoch": 0.16293221956622625,
      "grad_norm": 3.974438190460205,
      "learning_rate": 8.372207355516638e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 37433400,
      "step": 2326
    },
    {
      "epoch": 0.16300226781195548,
      "grad_norm": 3.833350658416748,
      "learning_rate": 8.371507530647986e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 37449784,
      "step": 2327
    },
    {
      "epoch": 0.16307231605768474,
      "grad_norm": 4.566055774688721,
      "learning_rate": 8.370807705779335e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 37465720,
      "step": 2328
    },
    {
      "epoch": 0.16314236430341397,
      "grad_norm": 3.455068588256836,
      "learning_rate": 8.370107880910684e-05,
      "loss": 0.9396,
      "num_input_tokens_seen": 37482104,
      "step": 2329
    },
    {
      "epoch": 0.16321241254914323,
      "grad_norm": 4.584096908569336,
      "learning_rate": 8.369408056042032e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 37498488,
      "step": 2330
    },
    {
      "epoch": 0.16328246079487246,
      "grad_norm": 4.0225958824157715,
      "learning_rate": 8.36870823117338e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 37514264,
      "step": 2331
    },
    {
      "epoch": 0.16335250904060172,
      "grad_norm": 5.311272144317627,
      "learning_rate": 8.368008406304729e-05,
      "loss": 1.2248,
      "num_input_tokens_seen": 37529280,
      "step": 2332
    },
    {
      "epoch": 0.16342255728633096,
      "grad_norm": 3.752720594406128,
      "learning_rate": 8.367308581436077e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 37545664,
      "step": 2333
    },
    {
      "epoch": 0.16349260553206021,
      "grad_norm": 3.8337442874908447,
      "learning_rate": 8.366608756567426e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 37562048,
      "step": 2334
    },
    {
      "epoch": 0.16356265377778945,
      "grad_norm": 3.818251132965088,
      "learning_rate": 8.365908931698775e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 37577848,
      "step": 2335
    },
    {
      "epoch": 0.1636327020235187,
      "grad_norm": 3.7170960903167725,
      "learning_rate": 8.365209106830123e-05,
      "loss": 1.2297,
      "num_input_tokens_seen": 37594232,
      "step": 2336
    },
    {
      "epoch": 0.16370275026924794,
      "grad_norm": 3.984950304031372,
      "learning_rate": 8.364509281961472e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 37610248,
      "step": 2337
    },
    {
      "epoch": 0.1637727985149772,
      "grad_norm": 3.4384636878967285,
      "learning_rate": 8.36380945709282e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 37626632,
      "step": 2338
    },
    {
      "epoch": 0.16384284676070643,
      "grad_norm": 3.952625274658203,
      "learning_rate": 8.363109632224169e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 37643016,
      "step": 2339
    },
    {
      "epoch": 0.1639128950064357,
      "grad_norm": 3.7193119525909424,
      "learning_rate": 8.362409807355516e-05,
      "loss": 0.9054,
      "num_input_tokens_seen": 37658216,
      "step": 2340
    },
    {
      "epoch": 0.16398294325216492,
      "grad_norm": 3.977997303009033,
      "learning_rate": 8.361709982486865e-05,
      "loss": 1.2904,
      "num_input_tokens_seen": 37674600,
      "step": 2341
    },
    {
      "epoch": 0.16405299149789418,
      "grad_norm": 5.108094215393066,
      "learning_rate": 8.361010157618214e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 37690184,
      "step": 2342
    },
    {
      "epoch": 0.1641230397436234,
      "grad_norm": 4.881065845489502,
      "learning_rate": 8.360310332749563e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 37705352,
      "step": 2343
    },
    {
      "epoch": 0.16419308798935267,
      "grad_norm": 4.128891468048096,
      "learning_rate": 8.359610507880912e-05,
      "loss": 0.8745,
      "num_input_tokens_seen": 37721736,
      "step": 2344
    },
    {
      "epoch": 0.1642631362350819,
      "grad_norm": 4.006495475769043,
      "learning_rate": 8.35891068301226e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 37738120,
      "step": 2345
    },
    {
      "epoch": 0.16433318448081116,
      "grad_norm": 3.877427101135254,
      "learning_rate": 8.358210858143608e-05,
      "loss": 0.9334,
      "num_input_tokens_seen": 37754504,
      "step": 2346
    },
    {
      "epoch": 0.1644032327265404,
      "grad_norm": 3.7013916969299316,
      "learning_rate": 8.357511033274955e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 37770792,
      "step": 2347
    },
    {
      "epoch": 0.16447328097226965,
      "grad_norm": 3.675049066543579,
      "learning_rate": 8.356811208406304e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 37786800,
      "step": 2348
    },
    {
      "epoch": 0.16454332921799888,
      "grad_norm": 4.831826210021973,
      "learning_rate": 8.356111383537654e-05,
      "loss": 0.9077,
      "num_input_tokens_seen": 37801760,
      "step": 2349
    },
    {
      "epoch": 0.16461337746372814,
      "grad_norm": 4.207952499389648,
      "learning_rate": 8.355411558669002e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 37818144,
      "step": 2350
    },
    {
      "epoch": 0.16468342570945738,
      "grad_norm": 3.9083497524261475,
      "learning_rate": 8.354711733800351e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 37833896,
      "step": 2351
    },
    {
      "epoch": 0.16475347395518664,
      "grad_norm": 4.307275295257568,
      "learning_rate": 8.3540119089317e-05,
      "loss": 1.0692,
      "num_input_tokens_seen": 37850280,
      "step": 2352
    },
    {
      "epoch": 0.16482352220091587,
      "grad_norm": 3.9434409141540527,
      "learning_rate": 8.353312084063047e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 37866664,
      "step": 2353
    },
    {
      "epoch": 0.16489357044664513,
      "grad_norm": 4.162476539611816,
      "learning_rate": 8.352612259194396e-05,
      "loss": 1.2,
      "num_input_tokens_seen": 37883048,
      "step": 2354
    },
    {
      "epoch": 0.1649636186923744,
      "grad_norm": 4.3073506355285645,
      "learning_rate": 8.351912434325745e-05,
      "loss": 1.2625,
      "num_input_tokens_seen": 37899264,
      "step": 2355
    },
    {
      "epoch": 0.16503366693810362,
      "grad_norm": 3.9900870323181152,
      "learning_rate": 8.351212609457094e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 37915648,
      "step": 2356
    },
    {
      "epoch": 0.16510371518383288,
      "grad_norm": 3.599282741546631,
      "learning_rate": 8.350512784588441e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 37932032,
      "step": 2357
    },
    {
      "epoch": 0.1651737634295621,
      "grad_norm": 3.796546697616577,
      "learning_rate": 8.34981295971979e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 37948416,
      "step": 2358
    },
    {
      "epoch": 0.16524381167529137,
      "grad_norm": 4.0810017585754395,
      "learning_rate": 8.349113134851139e-05,
      "loss": 0.9083,
      "num_input_tokens_seen": 37964072,
      "step": 2359
    },
    {
      "epoch": 0.1653138599210206,
      "grad_norm": 4.155765533447266,
      "learning_rate": 8.348413309982487e-05,
      "loss": 1.1827,
      "num_input_tokens_seen": 37980320,
      "step": 2360
    },
    {
      "epoch": 0.16538390816674986,
      "grad_norm": 4.131893634796143,
      "learning_rate": 8.347713485113835e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 37995872,
      "step": 2361
    },
    {
      "epoch": 0.1654539564124791,
      "grad_norm": 4.266848564147949,
      "learning_rate": 8.347013660245184e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 38011856,
      "step": 2362
    },
    {
      "epoch": 0.16552400465820835,
      "grad_norm": 3.8229875564575195,
      "learning_rate": 8.346313835376533e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 38028080,
      "step": 2363
    },
    {
      "epoch": 0.16559405290393758,
      "grad_norm": 4.0808234214782715,
      "learning_rate": 8.345614010507882e-05,
      "loss": 0.9667,
      "num_input_tokens_seen": 38043992,
      "step": 2364
    },
    {
      "epoch": 0.16566410114966684,
      "grad_norm": 4.470417022705078,
      "learning_rate": 8.34491418563923e-05,
      "loss": 1.2859,
      "num_input_tokens_seen": 38059848,
      "step": 2365
    },
    {
      "epoch": 0.16573414939539607,
      "grad_norm": 3.459963798522949,
      "learning_rate": 8.344214360770578e-05,
      "loss": 1.0801,
      "num_input_tokens_seen": 38076232,
      "step": 2366
    },
    {
      "epoch": 0.16580419764112533,
      "grad_norm": 3.6845312118530273,
      "learning_rate": 8.343514535901926e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 38092616,
      "step": 2367
    },
    {
      "epoch": 0.16587424588685457,
      "grad_norm": 3.683866500854492,
      "learning_rate": 8.342814711033275e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 38108880,
      "step": 2368
    },
    {
      "epoch": 0.16594429413258382,
      "grad_norm": 4.3266191482543945,
      "learning_rate": 8.342114886164625e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 38125264,
      "step": 2369
    },
    {
      "epoch": 0.16601434237831306,
      "grad_norm": 3.9031660556793213,
      "learning_rate": 8.341415061295972e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 38141648,
      "step": 2370
    },
    {
      "epoch": 0.16608439062404232,
      "grad_norm": 5.415440082550049,
      "learning_rate": 8.340715236427321e-05,
      "loss": 1.2011,
      "num_input_tokens_seen": 38157328,
      "step": 2371
    },
    {
      "epoch": 0.16615443886977155,
      "grad_norm": 4.017500877380371,
      "learning_rate": 8.34001541155867e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 38173096,
      "step": 2372
    },
    {
      "epoch": 0.1662244871155008,
      "grad_norm": 3.855212926864624,
      "learning_rate": 8.339315586690018e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 38189480,
      "step": 2373
    },
    {
      "epoch": 0.16629453536123004,
      "grad_norm": 3.8502743244171143,
      "learning_rate": 8.338615761821365e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 38205416,
      "step": 2374
    },
    {
      "epoch": 0.1663645836069593,
      "grad_norm": 6.8746867179870605,
      "learning_rate": 8.337915936952715e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 38221800,
      "step": 2375
    },
    {
      "epoch": 0.16643463185268853,
      "grad_norm": 3.9708571434020996,
      "learning_rate": 8.337216112084064e-05,
      "loss": 0.9832,
      "num_input_tokens_seen": 38237208,
      "step": 2376
    },
    {
      "epoch": 0.1665046800984178,
      "grad_norm": 4.927229404449463,
      "learning_rate": 8.336516287215412e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 38253592,
      "step": 2377
    },
    {
      "epoch": 0.16657472834414702,
      "grad_norm": 3.9976963996887207,
      "learning_rate": 8.33581646234676e-05,
      "loss": 1.1451,
      "num_input_tokens_seen": 38269184,
      "step": 2378
    },
    {
      "epoch": 0.16664477658987628,
      "grad_norm": 3.680177927017212,
      "learning_rate": 8.33511663747811e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 38285568,
      "step": 2379
    },
    {
      "epoch": 0.1667148248356055,
      "grad_norm": 3.768069267272949,
      "learning_rate": 8.334416812609457e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 38301952,
      "step": 2380
    },
    {
      "epoch": 0.16678487308133477,
      "grad_norm": 4.554010391235352,
      "learning_rate": 8.333716987740806e-05,
      "loss": 1.3037,
      "num_input_tokens_seen": 38318336,
      "step": 2381
    },
    {
      "epoch": 0.166854921327064,
      "grad_norm": 3.6799368858337402,
      "learning_rate": 8.333017162872155e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 38333544,
      "step": 2382
    },
    {
      "epoch": 0.16692496957279326,
      "grad_norm": 3.5584356784820557,
      "learning_rate": 8.332317338003503e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 38349632,
      "step": 2383
    },
    {
      "epoch": 0.1669950178185225,
      "grad_norm": 5.978849411010742,
      "learning_rate": 8.331617513134851e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 38364872,
      "step": 2384
    },
    {
      "epoch": 0.16706506606425175,
      "grad_norm": 4.641121864318848,
      "learning_rate": 8.3309176882662e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 38379800,
      "step": 2385
    },
    {
      "epoch": 0.167135114309981,
      "grad_norm": 3.895772695541382,
      "learning_rate": 8.330217863397549e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 38395744,
      "step": 2386
    },
    {
      "epoch": 0.16720516255571025,
      "grad_norm": 3.48437762260437,
      "learning_rate": 8.329518038528896e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 38412056,
      "step": 2387
    },
    {
      "epoch": 0.16727521080143948,
      "grad_norm": 4.2831549644470215,
      "learning_rate": 8.328818213660245e-05,
      "loss": 0.8967,
      "num_input_tokens_seen": 38426768,
      "step": 2388
    },
    {
      "epoch": 0.16734525904716874,
      "grad_norm": 3.7090001106262207,
      "learning_rate": 8.328118388791595e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 38442296,
      "step": 2389
    },
    {
      "epoch": 0.167415307292898,
      "grad_norm": 4.253223896026611,
      "learning_rate": 8.327418563922943e-05,
      "loss": 1.0169,
      "num_input_tokens_seen": 38458664,
      "step": 2390
    },
    {
      "epoch": 0.16748535553862723,
      "grad_norm": 4.919910907745361,
      "learning_rate": 8.326718739054292e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 38475048,
      "step": 2391
    },
    {
      "epoch": 0.1675554037843565,
      "grad_norm": 7.881314277648926,
      "learning_rate": 8.326018914185639e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 38491432,
      "step": 2392
    },
    {
      "epoch": 0.16762545203008572,
      "grad_norm": 6.979029655456543,
      "learning_rate": 8.325319089316988e-05,
      "loss": 0.9189,
      "num_input_tokens_seen": 38506312,
      "step": 2393
    },
    {
      "epoch": 0.16769550027581498,
      "grad_norm": 4.141571044921875,
      "learning_rate": 8.324619264448336e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 38522696,
      "step": 2394
    },
    {
      "epoch": 0.1677655485215442,
      "grad_norm": 4.306760311126709,
      "learning_rate": 8.323919439579686e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 38539080,
      "step": 2395
    },
    {
      "epoch": 0.16783559676727347,
      "grad_norm": 4.089770793914795,
      "learning_rate": 8.323219614711035e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 38555464,
      "step": 2396
    },
    {
      "epoch": 0.1679056450130027,
      "grad_norm": 3.648800849914551,
      "learning_rate": 8.322519789842382e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 38571848,
      "step": 2397
    },
    {
      "epoch": 0.16797569325873196,
      "grad_norm": 4.310317516326904,
      "learning_rate": 8.321819964973731e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 38587616,
      "step": 2398
    },
    {
      "epoch": 0.1680457415044612,
      "grad_norm": 5.373032093048096,
      "learning_rate": 8.32112014010508e-05,
      "loss": 0.9952,
      "num_input_tokens_seen": 38604000,
      "step": 2399
    },
    {
      "epoch": 0.16811578975019045,
      "grad_norm": 3.7830634117126465,
      "learning_rate": 8.320420315236427e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 38620384,
      "step": 2400
    },
    {
      "epoch": 0.16811578975019045,
      "eval_loss": 1.1429402828216553,
      "eval_runtime": 0.205,
      "eval_samples_per_second": 4.878,
      "eval_steps_per_second": 4.878,
      "num_input_tokens_seen": 38620384,
      "step": 2400
    },
    {
      "epoch": 0.16818583799591968,
      "grad_norm": 6.3896684646606445,
      "learning_rate": 8.319720490367776e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 38636288,
      "step": 2401
    },
    {
      "epoch": 0.16825588624164894,
      "grad_norm": 4.178726673126221,
      "learning_rate": 8.319020665499125e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 38652352,
      "step": 2402
    },
    {
      "epoch": 0.16832593448737818,
      "grad_norm": 3.7572708129882812,
      "learning_rate": 8.318320840630474e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 38668712,
      "step": 2403
    },
    {
      "epoch": 0.16839598273310744,
      "grad_norm": 3.688552141189575,
      "learning_rate": 8.317621015761821e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 38685096,
      "step": 2404
    },
    {
      "epoch": 0.16846603097883667,
      "grad_norm": 4.2040510177612305,
      "learning_rate": 8.31692119089317e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 38701480,
      "step": 2405
    },
    {
      "epoch": 0.16853607922456593,
      "grad_norm": 3.9412119388580322,
      "learning_rate": 8.316221366024519e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 38716904,
      "step": 2406
    },
    {
      "epoch": 0.16860612747029516,
      "grad_norm": 3.9538826942443848,
      "learning_rate": 8.315521541155867e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 38733288,
      "step": 2407
    },
    {
      "epoch": 0.16867617571602442,
      "grad_norm": 3.803135871887207,
      "learning_rate": 8.314821716287215e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 38749672,
      "step": 2408
    },
    {
      "epoch": 0.16874622396175365,
      "grad_norm": 4.323539733886719,
      "learning_rate": 8.314121891418564e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 38766056,
      "step": 2409
    },
    {
      "epoch": 0.1688162722074829,
      "grad_norm": 3.84000825881958,
      "learning_rate": 8.313422066549913e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 38782440,
      "step": 2410
    },
    {
      "epoch": 0.16888632045321214,
      "grad_norm": 6.76428747177124,
      "learning_rate": 8.312722241681261e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 38798824,
      "step": 2411
    },
    {
      "epoch": 0.1689563686989414,
      "grad_norm": 5.8638224601745605,
      "learning_rate": 8.31202241681261e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 38815112,
      "step": 2412
    },
    {
      "epoch": 0.16902641694467063,
      "grad_norm": 4.254051685333252,
      "learning_rate": 8.311322591943958e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 38831192,
      "step": 2413
    },
    {
      "epoch": 0.1690964651903999,
      "grad_norm": 3.641663074493408,
      "learning_rate": 8.310622767075306e-05,
      "loss": 0.9888,
      "num_input_tokens_seen": 38847360,
      "step": 2414
    },
    {
      "epoch": 0.16916651343612912,
      "grad_norm": 3.594768762588501,
      "learning_rate": 8.309922942206656e-05,
      "loss": 1.1156,
      "num_input_tokens_seen": 38863744,
      "step": 2415
    },
    {
      "epoch": 0.16923656168185838,
      "grad_norm": 3.6955742835998535,
      "learning_rate": 8.309223117338005e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 38879880,
      "step": 2416
    },
    {
      "epoch": 0.16930660992758761,
      "grad_norm": 3.64803409576416,
      "learning_rate": 8.308523292469353e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 38896264,
      "step": 2417
    },
    {
      "epoch": 0.16937665817331687,
      "grad_norm": 3.7921512126922607,
      "learning_rate": 8.307823467600701e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 38912648,
      "step": 2418
    },
    {
      "epoch": 0.1694467064190461,
      "grad_norm": 4.777346611022949,
      "learning_rate": 8.307123642732049e-05,
      "loss": 1.239,
      "num_input_tokens_seen": 38929032,
      "step": 2419
    },
    {
      "epoch": 0.16951675466477537,
      "grad_norm": 4.417767524719238,
      "learning_rate": 8.306423817863398e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 38945416,
      "step": 2420
    },
    {
      "epoch": 0.1695868029105046,
      "grad_norm": 4.257672309875488,
      "learning_rate": 8.305723992994747e-05,
      "loss": 0.8461,
      "num_input_tokens_seen": 38961800,
      "step": 2421
    },
    {
      "epoch": 0.16965685115623386,
      "grad_norm": 4.098975658416748,
      "learning_rate": 8.305024168126095e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 38978184,
      "step": 2422
    },
    {
      "epoch": 0.1697268994019631,
      "grad_norm": 5.206361293792725,
      "learning_rate": 8.304324343257444e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 38994568,
      "step": 2423
    },
    {
      "epoch": 0.16979694764769235,
      "grad_norm": 3.638395309448242,
      "learning_rate": 8.303624518388792e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 39010136,
      "step": 2424
    },
    {
      "epoch": 0.1698669958934216,
      "grad_norm": 3.4154045581817627,
      "learning_rate": 8.30292469352014e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 39026520,
      "step": 2425
    },
    {
      "epoch": 0.16993704413915084,
      "grad_norm": 3.923617362976074,
      "learning_rate": 8.30222486865149e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 39042816,
      "step": 2426
    },
    {
      "epoch": 0.1700070923848801,
      "grad_norm": 4.469310760498047,
      "learning_rate": 8.301525043782837e-05,
      "loss": 1.3424,
      "num_input_tokens_seen": 39059040,
      "step": 2427
    },
    {
      "epoch": 0.17007714063060933,
      "grad_norm": 4.111564636230469,
      "learning_rate": 8.300825218914186e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 39074992,
      "step": 2428
    },
    {
      "epoch": 0.1701471888763386,
      "grad_norm": 3.7809438705444336,
      "learning_rate": 8.300125394045535e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 39090840,
      "step": 2429
    },
    {
      "epoch": 0.17021723712206782,
      "grad_norm": 3.704542875289917,
      "learning_rate": 8.299425569176884e-05,
      "loss": 1.1784,
      "num_input_tokens_seen": 39107136,
      "step": 2430
    },
    {
      "epoch": 0.17028728536779708,
      "grad_norm": 4.356417179107666,
      "learning_rate": 8.298725744308231e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 39123520,
      "step": 2431
    },
    {
      "epoch": 0.1703573336135263,
      "grad_norm": 3.400228500366211,
      "learning_rate": 8.29802591943958e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 39139904,
      "step": 2432
    },
    {
      "epoch": 0.17042738185925557,
      "grad_norm": 4.777987480163574,
      "learning_rate": 8.297326094570929e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 39156288,
      "step": 2433
    },
    {
      "epoch": 0.1704974301049848,
      "grad_norm": 5.600007057189941,
      "learning_rate": 8.296626269702276e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 39171928,
      "step": 2434
    },
    {
      "epoch": 0.17056747835071406,
      "grad_norm": 3.72717022895813,
      "learning_rate": 8.295926444833627e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 39188032,
      "step": 2435
    },
    {
      "epoch": 0.1706375265964433,
      "grad_norm": 7.264038562774658,
      "learning_rate": 8.295226619964974e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 39203816,
      "step": 2436
    },
    {
      "epoch": 0.17070757484217255,
      "grad_norm": 4.103509426116943,
      "learning_rate": 8.294526795096323e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 39220200,
      "step": 2437
    },
    {
      "epoch": 0.1707776230879018,
      "grad_norm": 3.6456661224365234,
      "learning_rate": 8.29382697022767e-05,
      "loss": 0.9107,
      "num_input_tokens_seen": 39236584,
      "step": 2438
    },
    {
      "epoch": 0.17084767133363105,
      "grad_norm": 3.750075340270996,
      "learning_rate": 8.293127145359019e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 39252968,
      "step": 2439
    },
    {
      "epoch": 0.17091771957936028,
      "grad_norm": 4.5003581047058105,
      "learning_rate": 8.292427320490368e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 39269192,
      "step": 2440
    },
    {
      "epoch": 0.17098776782508954,
      "grad_norm": 4.513885498046875,
      "learning_rate": 8.291727495621717e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 39285576,
      "step": 2441
    },
    {
      "epoch": 0.17105781607081877,
      "grad_norm": 4.220343589782715,
      "learning_rate": 8.291027670753066e-05,
      "loss": 1.3044,
      "num_input_tokens_seen": 39300864,
      "step": 2442
    },
    {
      "epoch": 0.17112786431654803,
      "grad_norm": 4.986631393432617,
      "learning_rate": 8.290327845884415e-05,
      "loss": 1.0377,
      "num_input_tokens_seen": 39317208,
      "step": 2443
    },
    {
      "epoch": 0.17119791256227726,
      "grad_norm": 7.632670879364014,
      "learning_rate": 8.289628021015762e-05,
      "loss": 1.1749,
      "num_input_tokens_seen": 39332392,
      "step": 2444
    },
    {
      "epoch": 0.17126796080800652,
      "grad_norm": 3.588841199874878,
      "learning_rate": 8.288928196147111e-05,
      "loss": 0.8124,
      "num_input_tokens_seen": 39348600,
      "step": 2445
    },
    {
      "epoch": 0.17133800905373575,
      "grad_norm": 4.311728477478027,
      "learning_rate": 8.288228371278459e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 39364456,
      "step": 2446
    },
    {
      "epoch": 0.171408057299465,
      "grad_norm": 6.236140251159668,
      "learning_rate": 8.287528546409807e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 39379496,
      "step": 2447
    },
    {
      "epoch": 0.17147810554519424,
      "grad_norm": 4.228808403015137,
      "learning_rate": 8.286828721541156e-05,
      "loss": 1.0185,
      "num_input_tokens_seen": 39395880,
      "step": 2448
    },
    {
      "epoch": 0.1715481537909235,
      "grad_norm": 3.873366117477417,
      "learning_rate": 8.286128896672505e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 39412264,
      "step": 2449
    },
    {
      "epoch": 0.17161820203665273,
      "grad_norm": 3.797846794128418,
      "learning_rate": 8.285429071803854e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 39428648,
      "step": 2450
    },
    {
      "epoch": 0.171688250282382,
      "grad_norm": 3.798875093460083,
      "learning_rate": 8.284729246935202e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 39445032,
      "step": 2451
    },
    {
      "epoch": 0.17175829852811123,
      "grad_norm": 5.118900299072266,
      "learning_rate": 8.28402942206655e-05,
      "loss": 1.14,
      "num_input_tokens_seen": 39460168,
      "step": 2452
    },
    {
      "epoch": 0.17182834677384048,
      "grad_norm": 4.157371520996094,
      "learning_rate": 8.283329597197899e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 39476544,
      "step": 2453
    },
    {
      "epoch": 0.17189839501956972,
      "grad_norm": 3.760786771774292,
      "learning_rate": 8.282629772329247e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 39492928,
      "step": 2454
    },
    {
      "epoch": 0.17196844326529898,
      "grad_norm": 4.252779960632324,
      "learning_rate": 8.281929947460596e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 39509312,
      "step": 2455
    },
    {
      "epoch": 0.1720384915110282,
      "grad_norm": 4.374740123748779,
      "learning_rate": 8.281230122591944e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 39525696,
      "step": 2456
    },
    {
      "epoch": 0.17210853975675747,
      "grad_norm": 4.460380554199219,
      "learning_rate": 8.280530297723293e-05,
      "loss": 1.1876,
      "num_input_tokens_seen": 39541864,
      "step": 2457
    },
    {
      "epoch": 0.17217858800248673,
      "grad_norm": 4.22148323059082,
      "learning_rate": 8.279830472854641e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 39557944,
      "step": 2458
    },
    {
      "epoch": 0.17224863624821596,
      "grad_norm": 4.310081481933594,
      "learning_rate": 8.27913064798599e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 39574328,
      "step": 2459
    },
    {
      "epoch": 0.17231868449394522,
      "grad_norm": 4.15192174911499,
      "learning_rate": 8.278430823117339e-05,
      "loss": 0.7793,
      "num_input_tokens_seen": 39589312,
      "step": 2460
    },
    {
      "epoch": 0.17238873273967445,
      "grad_norm": 4.6561455726623535,
      "learning_rate": 8.277730998248687e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 39605456,
      "step": 2461
    },
    {
      "epoch": 0.1724587809854037,
      "grad_norm": 4.273087978363037,
      "learning_rate": 8.277031173380036e-05,
      "loss": 1.1436,
      "num_input_tokens_seen": 39621840,
      "step": 2462
    },
    {
      "epoch": 0.17252882923113294,
      "grad_norm": 4.575830459594727,
      "learning_rate": 8.276331348511384e-05,
      "loss": 1.2589,
      "num_input_tokens_seen": 39638224,
      "step": 2463
    },
    {
      "epoch": 0.1725988774768622,
      "grad_norm": 3.9122824668884277,
      "learning_rate": 8.275631523642733e-05,
      "loss": 1.0634,
      "num_input_tokens_seen": 39654608,
      "step": 2464
    },
    {
      "epoch": 0.17266892572259143,
      "grad_norm": 4.991362571716309,
      "learning_rate": 8.27493169877408e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 39669824,
      "step": 2465
    },
    {
      "epoch": 0.1727389739683207,
      "grad_norm": 4.688175678253174,
      "learning_rate": 8.274231873905429e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 39686208,
      "step": 2466
    },
    {
      "epoch": 0.17280902221404992,
      "grad_norm": 3.779524087905884,
      "learning_rate": 8.273532049036778e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 39702336,
      "step": 2467
    },
    {
      "epoch": 0.17287907045977918,
      "grad_norm": 4.117679595947266,
      "learning_rate": 8.272832224168127e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 39718232,
      "step": 2468
    },
    {
      "epoch": 0.17294911870550841,
      "grad_norm": 3.810084819793701,
      "learning_rate": 8.272132399299476e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 39733584,
      "step": 2469
    },
    {
      "epoch": 0.17301916695123767,
      "grad_norm": 3.7730038166046143,
      "learning_rate": 8.271432574430824e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 39749968,
      "step": 2470
    },
    {
      "epoch": 0.1730892151969669,
      "grad_norm": 7.299304008483887,
      "learning_rate": 8.270732749562172e-05,
      "loss": 1.3425,
      "num_input_tokens_seen": 39765552,
      "step": 2471
    },
    {
      "epoch": 0.17315926344269617,
      "grad_norm": 4.079380512237549,
      "learning_rate": 8.270032924693521e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 39781936,
      "step": 2472
    },
    {
      "epoch": 0.1732293116884254,
      "grad_norm": 3.736607789993286,
      "learning_rate": 8.269333099824868e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 39797688,
      "step": 2473
    },
    {
      "epoch": 0.17329935993415466,
      "grad_norm": 5.587291240692139,
      "learning_rate": 8.268633274956217e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 39814072,
      "step": 2474
    },
    {
      "epoch": 0.1733694081798839,
      "grad_norm": 3.5963592529296875,
      "learning_rate": 8.267933450087566e-05,
      "loss": 0.9947,
      "num_input_tokens_seen": 39830456,
      "step": 2475
    },
    {
      "epoch": 0.17343945642561315,
      "grad_norm": 5.241317272186279,
      "learning_rate": 8.267233625218915e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 39846728,
      "step": 2476
    },
    {
      "epoch": 0.17350950467134238,
      "grad_norm": 4.194108009338379,
      "learning_rate": 8.266533800350264e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 39863112,
      "step": 2477
    },
    {
      "epoch": 0.17357955291707164,
      "grad_norm": 4.698538780212402,
      "learning_rate": 8.265833975481611e-05,
      "loss": 1.3673,
      "num_input_tokens_seen": 39878624,
      "step": 2478
    },
    {
      "epoch": 0.17364960116280087,
      "grad_norm": 5.960018634796143,
      "learning_rate": 8.26513415061296e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 39894944,
      "step": 2479
    },
    {
      "epoch": 0.17371964940853013,
      "grad_norm": 4.386090278625488,
      "learning_rate": 8.264434325744309e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 39911040,
      "step": 2480
    },
    {
      "epoch": 0.17378969765425936,
      "grad_norm": 3.7272467613220215,
      "learning_rate": 8.263734500875658e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 39927408,
      "step": 2481
    },
    {
      "epoch": 0.17385974589998862,
      "grad_norm": 4.296888828277588,
      "learning_rate": 8.263034676007005e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 39943792,
      "step": 2482
    },
    {
      "epoch": 0.17392979414571785,
      "grad_norm": 5.469056606292725,
      "learning_rate": 8.262334851138354e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 39958176,
      "step": 2483
    },
    {
      "epoch": 0.1739998423914471,
      "grad_norm": 5.6080498695373535,
      "learning_rate": 8.261635026269703e-05,
      "loss": 1.2713,
      "num_input_tokens_seen": 39973592,
      "step": 2484
    },
    {
      "epoch": 0.17406989063717634,
      "grad_norm": 3.6164181232452393,
      "learning_rate": 8.26093520140105e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 39989792,
      "step": 2485
    },
    {
      "epoch": 0.1741399388829056,
      "grad_norm": 3.757291078567505,
      "learning_rate": 8.2602353765324e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 40005672,
      "step": 2486
    },
    {
      "epoch": 0.17420998712863484,
      "grad_norm": 5.1490559577941895,
      "learning_rate": 8.259535551663748e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 40021816,
      "step": 2487
    },
    {
      "epoch": 0.1742800353743641,
      "grad_norm": 3.9055886268615723,
      "learning_rate": 8.258835726795097e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 40038200,
      "step": 2488
    },
    {
      "epoch": 0.17435008362009333,
      "grad_norm": 5.496553897857666,
      "learning_rate": 8.258135901926446e-05,
      "loss": 1.3214,
      "num_input_tokens_seen": 40054584,
      "step": 2489
    },
    {
      "epoch": 0.1744201318658226,
      "grad_norm": 4.069197177886963,
      "learning_rate": 8.257436077057793e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 40070968,
      "step": 2490
    },
    {
      "epoch": 0.17449018011155182,
      "grad_norm": 5.098565101623535,
      "learning_rate": 8.256736252189142e-05,
      "loss": 0.9918,
      "num_input_tokens_seen": 40087352,
      "step": 2491
    },
    {
      "epoch": 0.17456022835728108,
      "grad_norm": 4.083621025085449,
      "learning_rate": 8.25603642732049e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 40103736,
      "step": 2492
    },
    {
      "epoch": 0.17463027660301034,
      "grad_norm": 3.8676462173461914,
      "learning_rate": 8.255336602451839e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 40120120,
      "step": 2493
    },
    {
      "epoch": 0.17470032484873957,
      "grad_norm": 3.8799197673797607,
      "learning_rate": 8.254636777583188e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 40136504,
      "step": 2494
    },
    {
      "epoch": 0.17477037309446883,
      "grad_norm": 5.469006538391113,
      "learning_rate": 8.253936952714536e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 40152888,
      "step": 2495
    },
    {
      "epoch": 0.17484042134019806,
      "grad_norm": 4.163306713104248,
      "learning_rate": 8.253237127845885e-05,
      "loss": 1.2059,
      "num_input_tokens_seen": 40169272,
      "step": 2496
    },
    {
      "epoch": 0.17491046958592732,
      "grad_norm": 3.792062282562256,
      "learning_rate": 8.252537302977234e-05,
      "loss": 1.0806,
      "num_input_tokens_seen": 40185656,
      "step": 2497
    },
    {
      "epoch": 0.17498051783165655,
      "grad_norm": 3.6881046295166016,
      "learning_rate": 8.251837478108582e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 40202040,
      "step": 2498
    },
    {
      "epoch": 0.1750505660773858,
      "grad_norm": 3.818491220474243,
      "learning_rate": 8.25113765323993e-05,
      "loss": 1.2193,
      "num_input_tokens_seen": 40218424,
      "step": 2499
    },
    {
      "epoch": 0.17512061432311504,
      "grad_norm": 3.77933931350708,
      "learning_rate": 8.250437828371278e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 40234216,
      "step": 2500
    },
    {
      "epoch": 0.1751906625688443,
      "grad_norm": 4.106552600860596,
      "learning_rate": 8.249738003502628e-05,
      "loss": 0.8558,
      "num_input_tokens_seen": 40250368,
      "step": 2501
    },
    {
      "epoch": 0.17526071081457353,
      "grad_norm": 4.9382710456848145,
      "learning_rate": 8.249038178633976e-05,
      "loss": 1.3082,
      "num_input_tokens_seen": 40266600,
      "step": 2502
    },
    {
      "epoch": 0.1753307590603028,
      "grad_norm": 3.8894200325012207,
      "learning_rate": 8.248338353765325e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 40282984,
      "step": 2503
    },
    {
      "epoch": 0.17540080730603202,
      "grad_norm": 3.793044328689575,
      "learning_rate": 8.247638528896673e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 40299368,
      "step": 2504
    },
    {
      "epoch": 0.17547085555176128,
      "grad_norm": 4.874731540679932,
      "learning_rate": 8.246938704028021e-05,
      "loss": 1.2598,
      "num_input_tokens_seen": 40315752,
      "step": 2505
    },
    {
      "epoch": 0.17554090379749052,
      "grad_norm": 3.908191680908203,
      "learning_rate": 8.24623887915937e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 40332136,
      "step": 2506
    },
    {
      "epoch": 0.17561095204321978,
      "grad_norm": 3.585002899169922,
      "learning_rate": 8.245539054290719e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 40348520,
      "step": 2507
    },
    {
      "epoch": 0.175681000288949,
      "grad_norm": 3.9742348194122314,
      "learning_rate": 8.244839229422068e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 40364760,
      "step": 2508
    },
    {
      "epoch": 0.17575104853467827,
      "grad_norm": 5.1725921630859375,
      "learning_rate": 8.244139404553415e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 40380072,
      "step": 2509
    },
    {
      "epoch": 0.1758210967804075,
      "grad_norm": 4.038326263427734,
      "learning_rate": 8.243439579684764e-05,
      "loss": 1.2252,
      "num_input_tokens_seen": 40395472,
      "step": 2510
    },
    {
      "epoch": 0.17589114502613676,
      "grad_norm": 3.7381017208099365,
      "learning_rate": 8.242739754816113e-05,
      "loss": 1.041,
      "num_input_tokens_seen": 40411280,
      "step": 2511
    },
    {
      "epoch": 0.175961193271866,
      "grad_norm": 4.327959060668945,
      "learning_rate": 8.24203992994746e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 40427664,
      "step": 2512
    },
    {
      "epoch": 0.17603124151759525,
      "grad_norm": 3.720078706741333,
      "learning_rate": 8.241340105078809e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 40443760,
      "step": 2513
    },
    {
      "epoch": 0.17610128976332448,
      "grad_norm": 4.0901618003845215,
      "learning_rate": 8.240640280210158e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 40460144,
      "step": 2514
    },
    {
      "epoch": 0.17617133800905374,
      "grad_norm": 4.013705730438232,
      "learning_rate": 8.239940455341507e-05,
      "loss": 1.0817,
      "num_input_tokens_seen": 40476528,
      "step": 2515
    },
    {
      "epoch": 0.17624138625478297,
      "grad_norm": 3.8833489418029785,
      "learning_rate": 8.239240630472856e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 40492768,
      "step": 2516
    },
    {
      "epoch": 0.17631143450051223,
      "grad_norm": 7.381611347198486,
      "learning_rate": 8.238540805604203e-05,
      "loss": 1.3033,
      "num_input_tokens_seen": 40507344,
      "step": 2517
    },
    {
      "epoch": 0.17638148274624146,
      "grad_norm": 3.8792364597320557,
      "learning_rate": 8.237840980735552e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 40523552,
      "step": 2518
    },
    {
      "epoch": 0.17645153099197072,
      "grad_norm": 5.19634485244751,
      "learning_rate": 8.2371411558669e-05,
      "loss": 1.2186,
      "num_input_tokens_seen": 40538640,
      "step": 2519
    },
    {
      "epoch": 0.17652157923769995,
      "grad_norm": 4.081907749176025,
      "learning_rate": 8.236441330998248e-05,
      "loss": 1.1075,
      "num_input_tokens_seen": 40555024,
      "step": 2520
    },
    {
      "epoch": 0.17659162748342921,
      "grad_norm": 4.296377182006836,
      "learning_rate": 8.235741506129599e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 40570480,
      "step": 2521
    },
    {
      "epoch": 0.17666167572915845,
      "grad_norm": 4.143492221832275,
      "learning_rate": 8.235041681260946e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 40586664,
      "step": 2522
    },
    {
      "epoch": 0.1767317239748877,
      "grad_norm": 3.9894590377807617,
      "learning_rate": 8.234341856392295e-05,
      "loss": 0.913,
      "num_input_tokens_seen": 40603048,
      "step": 2523
    },
    {
      "epoch": 0.17680177222061694,
      "grad_norm": 4.283662796020508,
      "learning_rate": 8.233642031523644e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 40618440,
      "step": 2524
    },
    {
      "epoch": 0.1768718204663462,
      "grad_norm": 4.126082420349121,
      "learning_rate": 8.232942206654991e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 40634824,
      "step": 2525
    },
    {
      "epoch": 0.17694186871207543,
      "grad_norm": 4.252981662750244,
      "learning_rate": 8.23224238178634e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 40650640,
      "step": 2526
    },
    {
      "epoch": 0.1770119169578047,
      "grad_norm": 3.768542528152466,
      "learning_rate": 8.231542556917689e-05,
      "loss": 0.8221,
      "num_input_tokens_seen": 40667000,
      "step": 2527
    },
    {
      "epoch": 0.17708196520353395,
      "grad_norm": 4.067849636077881,
      "learning_rate": 8.230842732049038e-05,
      "loss": 1.2117,
      "num_input_tokens_seen": 40683288,
      "step": 2528
    },
    {
      "epoch": 0.17715201344926318,
      "grad_norm": 4.7552995681762695,
      "learning_rate": 8.230142907180385e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 40699304,
      "step": 2529
    },
    {
      "epoch": 0.17722206169499244,
      "grad_norm": 4.099888324737549,
      "learning_rate": 8.229443082311734e-05,
      "loss": 1.3335,
      "num_input_tokens_seen": 40715688,
      "step": 2530
    },
    {
      "epoch": 0.17729210994072167,
      "grad_norm": 4.219737529754639,
      "learning_rate": 8.228743257443083e-05,
      "loss": 1.3004,
      "num_input_tokens_seen": 40731640,
      "step": 2531
    },
    {
      "epoch": 0.17736215818645093,
      "grad_norm": 4.125600337982178,
      "learning_rate": 8.22804343257443e-05,
      "loss": 1.1828,
      "num_input_tokens_seen": 40747664,
      "step": 2532
    },
    {
      "epoch": 0.17743220643218016,
      "grad_norm": 3.7761423587799072,
      "learning_rate": 8.22734360770578e-05,
      "loss": 1.1082,
      "num_input_tokens_seen": 40764048,
      "step": 2533
    },
    {
      "epoch": 0.17750225467790942,
      "grad_norm": 5.0669026374816895,
      "learning_rate": 8.226643782837128e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 40779160,
      "step": 2534
    },
    {
      "epoch": 0.17757230292363865,
      "grad_norm": 4.688200950622559,
      "learning_rate": 8.225943957968477e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 40795416,
      "step": 2535
    },
    {
      "epoch": 0.1776423511693679,
      "grad_norm": 3.62204647064209,
      "learning_rate": 8.225244133099825e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 40811800,
      "step": 2536
    },
    {
      "epoch": 0.17771239941509714,
      "grad_norm": 4.086390495300293,
      "learning_rate": 8.224544308231174e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 40826960,
      "step": 2537
    },
    {
      "epoch": 0.1777824476608264,
      "grad_norm": 5.574249744415283,
      "learning_rate": 8.223844483362522e-05,
      "loss": 1.2678,
      "num_input_tokens_seen": 40843344,
      "step": 2538
    },
    {
      "epoch": 0.17785249590655564,
      "grad_norm": 3.9721264839172363,
      "learning_rate": 8.22314465849387e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 40859448,
      "step": 2539
    },
    {
      "epoch": 0.1779225441522849,
      "grad_norm": 4.220152854919434,
      "learning_rate": 8.222444833625219e-05,
      "loss": 1.1014,
      "num_input_tokens_seen": 40875128,
      "step": 2540
    },
    {
      "epoch": 0.17799259239801413,
      "grad_norm": 3.905205011367798,
      "learning_rate": 8.221745008756569e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 40890624,
      "step": 2541
    },
    {
      "epoch": 0.1780626406437434,
      "grad_norm": 4.114316463470459,
      "learning_rate": 8.221045183887917e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 40905624,
      "step": 2542
    },
    {
      "epoch": 0.17813268888947262,
      "grad_norm": 4.810879230499268,
      "learning_rate": 8.220345359019265e-05,
      "loss": 0.9447,
      "num_input_tokens_seen": 40922008,
      "step": 2543
    },
    {
      "epoch": 0.17820273713520188,
      "grad_norm": 4.224065780639648,
      "learning_rate": 8.219645534150613e-05,
      "loss": 1.1176,
      "num_input_tokens_seen": 40938392,
      "step": 2544
    },
    {
      "epoch": 0.1782727853809311,
      "grad_norm": 3.7086703777313232,
      "learning_rate": 8.218945709281962e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 40954776,
      "step": 2545
    },
    {
      "epoch": 0.17834283362666037,
      "grad_norm": 4.346426963806152,
      "learning_rate": 8.218245884413309e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 40971160,
      "step": 2546
    },
    {
      "epoch": 0.1784128818723896,
      "grad_norm": 3.9295589923858643,
      "learning_rate": 8.21754605954466e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 40987544,
      "step": 2547
    },
    {
      "epoch": 0.17848293011811886,
      "grad_norm": 4.224534034729004,
      "learning_rate": 8.216846234676008e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 41002816,
      "step": 2548
    },
    {
      "epoch": 0.1785529783638481,
      "grad_norm": 3.940401315689087,
      "learning_rate": 8.216146409807356e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 41018560,
      "step": 2549
    },
    {
      "epoch": 0.17862302660957735,
      "grad_norm": 4.010072231292725,
      "learning_rate": 8.215446584938705e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 41033976,
      "step": 2550
    },
    {
      "epoch": 0.17869307485530658,
      "grad_norm": 4.192416191101074,
      "learning_rate": 8.214746760070054e-05,
      "loss": 1.0954,
      "num_input_tokens_seen": 41049384,
      "step": 2551
    },
    {
      "epoch": 0.17876312310103584,
      "grad_norm": 3.765962600708008,
      "learning_rate": 8.214046935201401e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 41065528,
      "step": 2552
    },
    {
      "epoch": 0.17883317134676507,
      "grad_norm": 3.7856082916259766,
      "learning_rate": 8.21334711033275e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 41081912,
      "step": 2553
    },
    {
      "epoch": 0.17890321959249433,
      "grad_norm": 4.845935821533203,
      "learning_rate": 8.212647285464099e-05,
      "loss": 1.2907,
      "num_input_tokens_seen": 41098056,
      "step": 2554
    },
    {
      "epoch": 0.17897326783822357,
      "grad_norm": 4.835206985473633,
      "learning_rate": 8.211947460595448e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 41114376,
      "step": 2555
    },
    {
      "epoch": 0.17904331608395282,
      "grad_norm": 3.9637155532836914,
      "learning_rate": 8.211247635726795e-05,
      "loss": 1.1689,
      "num_input_tokens_seen": 41130760,
      "step": 2556
    },
    {
      "epoch": 0.17911336432968206,
      "grad_norm": 3.5001652240753174,
      "learning_rate": 8.210547810858144e-05,
      "loss": 0.9798,
      "num_input_tokens_seen": 41147040,
      "step": 2557
    },
    {
      "epoch": 0.17918341257541132,
      "grad_norm": 5.54505729675293,
      "learning_rate": 8.209847985989493e-05,
      "loss": 1.3004,
      "num_input_tokens_seen": 41163312,
      "step": 2558
    },
    {
      "epoch": 0.17925346082114055,
      "grad_norm": 4.122933387756348,
      "learning_rate": 8.20914816112084e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 41179632,
      "step": 2559
    },
    {
      "epoch": 0.1793235090668698,
      "grad_norm": 4.166035175323486,
      "learning_rate": 8.208448336252189e-05,
      "loss": 1.0022,
      "num_input_tokens_seen": 41196000,
      "step": 2560
    },
    {
      "epoch": 0.17939355731259904,
      "grad_norm": 4.129281520843506,
      "learning_rate": 8.20774851138354e-05,
      "loss": 1.2342,
      "num_input_tokens_seen": 41211944,
      "step": 2561
    },
    {
      "epoch": 0.1794636055583283,
      "grad_norm": 3.9011406898498535,
      "learning_rate": 8.207048686514887e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 41227680,
      "step": 2562
    },
    {
      "epoch": 0.17953365380405756,
      "grad_norm": 3.717945098876953,
      "learning_rate": 8.206348861646234e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 41244064,
      "step": 2563
    },
    {
      "epoch": 0.1796037020497868,
      "grad_norm": 5.05475378036499,
      "learning_rate": 8.205649036777583e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 41260448,
      "step": 2564
    },
    {
      "epoch": 0.17967375029551605,
      "grad_norm": 4.52910041809082,
      "learning_rate": 8.204949211908932e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 41276832,
      "step": 2565
    },
    {
      "epoch": 0.17974379854124528,
      "grad_norm": 4.6492695808410645,
      "learning_rate": 8.20424938704028e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 41293216,
      "step": 2566
    },
    {
      "epoch": 0.17981384678697454,
      "grad_norm": 4.7587456703186035,
      "learning_rate": 8.20354956217163e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 41309600,
      "step": 2567
    },
    {
      "epoch": 0.17988389503270377,
      "grad_norm": 6.8467817306518555,
      "learning_rate": 8.202849737302979e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 41325984,
      "step": 2568
    },
    {
      "epoch": 0.17995394327843303,
      "grad_norm": 3.7771074771881104,
      "learning_rate": 8.202149912434326e-05,
      "loss": 1.1603,
      "num_input_tokens_seen": 41342368,
      "step": 2569
    },
    {
      "epoch": 0.18002399152416226,
      "grad_norm": 3.8494906425476074,
      "learning_rate": 8.201450087565675e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 41357992,
      "step": 2570
    },
    {
      "epoch": 0.18009403976989152,
      "grad_norm": 4.079790115356445,
      "learning_rate": 8.200750262697023e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 41374256,
      "step": 2571
    },
    {
      "epoch": 0.18016408801562075,
      "grad_norm": 7.093918800354004,
      "learning_rate": 8.200050437828371e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 41388728,
      "step": 2572
    },
    {
      "epoch": 0.18023413626135001,
      "grad_norm": 4.636250972747803,
      "learning_rate": 8.19935061295972e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 41404488,
      "step": 2573
    },
    {
      "epoch": 0.18030418450707925,
      "grad_norm": 3.789625644683838,
      "learning_rate": 8.198650788091069e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 41420200,
      "step": 2574
    },
    {
      "epoch": 0.1803742327528085,
      "grad_norm": 3.849637508392334,
      "learning_rate": 8.197950963222418e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 41436496,
      "step": 2575
    },
    {
      "epoch": 0.18044428099853774,
      "grad_norm": 3.6819775104522705,
      "learning_rate": 8.197251138353766e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 41452736,
      "step": 2576
    },
    {
      "epoch": 0.180514329244267,
      "grad_norm": 4.505229473114014,
      "learning_rate": 8.196551313485114e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 41468976,
      "step": 2577
    },
    {
      "epoch": 0.18058437748999623,
      "grad_norm": 5.465007781982422,
      "learning_rate": 8.195851488616463e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 41485064,
      "step": 2578
    },
    {
      "epoch": 0.1806544257357255,
      "grad_norm": 3.993953227996826,
      "learning_rate": 8.195151663747811e-05,
      "loss": 1.3406,
      "num_input_tokens_seen": 41501448,
      "step": 2579
    },
    {
      "epoch": 0.18072447398145472,
      "grad_norm": 5.29327392578125,
      "learning_rate": 8.19445183887916e-05,
      "loss": 1.2397,
      "num_input_tokens_seen": 41517832,
      "step": 2580
    },
    {
      "epoch": 0.18079452222718398,
      "grad_norm": 4.132434844970703,
      "learning_rate": 8.193752014010508e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 41532976,
      "step": 2581
    },
    {
      "epoch": 0.1808645704729132,
      "grad_norm": 5.620279788970947,
      "learning_rate": 8.193052189141857e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 41548784,
      "step": 2582
    },
    {
      "epoch": 0.18093461871864247,
      "grad_norm": 3.9721081256866455,
      "learning_rate": 8.192352364273205e-05,
      "loss": 1.0458,
      "num_input_tokens_seen": 41565168,
      "step": 2583
    },
    {
      "epoch": 0.1810046669643717,
      "grad_norm": 5.015312194824219,
      "learning_rate": 8.191652539404554e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 41580584,
      "step": 2584
    },
    {
      "epoch": 0.18107471521010096,
      "grad_norm": 5.385783672332764,
      "learning_rate": 8.190952714535903e-05,
      "loss": 1.0853,
      "num_input_tokens_seen": 41596656,
      "step": 2585
    },
    {
      "epoch": 0.1811447634558302,
      "grad_norm": 4.1005120277404785,
      "learning_rate": 8.19025288966725e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 41611752,
      "step": 2586
    },
    {
      "epoch": 0.18121481170155945,
      "grad_norm": 3.6853153705596924,
      "learning_rate": 8.1895530647986e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 41627408,
      "step": 2587
    },
    {
      "epoch": 0.18128485994728868,
      "grad_norm": 3.7818400859832764,
      "learning_rate": 8.188853239929949e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 41643792,
      "step": 2588
    },
    {
      "epoch": 0.18135490819301794,
      "grad_norm": 4.564868450164795,
      "learning_rate": 8.188153415061297e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 41658768,
      "step": 2589
    },
    {
      "epoch": 0.18142495643874718,
      "grad_norm": 4.092021465301514,
      "learning_rate": 8.187453590192644e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 41675088,
      "step": 2590
    },
    {
      "epoch": 0.18149500468447644,
      "grad_norm": 5.051564693450928,
      "learning_rate": 8.186753765323993e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 41690376,
      "step": 2591
    },
    {
      "epoch": 0.18156505293020567,
      "grad_norm": 3.5786261558532715,
      "learning_rate": 8.186053940455342e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 41706760,
      "step": 2592
    },
    {
      "epoch": 0.18163510117593493,
      "grad_norm": 4.11420202255249,
      "learning_rate": 8.185354115586691e-05,
      "loss": 1.1234,
      "num_input_tokens_seen": 41721760,
      "step": 2593
    },
    {
      "epoch": 0.18170514942166416,
      "grad_norm": 4.445348262786865,
      "learning_rate": 8.18465429071804e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 41737640,
      "step": 2594
    },
    {
      "epoch": 0.18177519766739342,
      "grad_norm": 5.705301284790039,
      "learning_rate": 8.183954465849388e-05,
      "loss": 1.2254,
      "num_input_tokens_seen": 41753784,
      "step": 2595
    },
    {
      "epoch": 0.18184524591312265,
      "grad_norm": 3.7948646545410156,
      "learning_rate": 8.183254640980736e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 41770120,
      "step": 2596
    },
    {
      "epoch": 0.1819152941588519,
      "grad_norm": 4.296072959899902,
      "learning_rate": 8.182554816112085e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 41786504,
      "step": 2597
    },
    {
      "epoch": 0.18198534240458117,
      "grad_norm": 4.750889778137207,
      "learning_rate": 8.181854991243432e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 41802888,
      "step": 2598
    },
    {
      "epoch": 0.1820553906503104,
      "grad_norm": 4.031731128692627,
      "learning_rate": 8.181155166374781e-05,
      "loss": 1.1096,
      "num_input_tokens_seen": 41819264,
      "step": 2599
    },
    {
      "epoch": 0.18212543889603966,
      "grad_norm": 4.620563507080078,
      "learning_rate": 8.18045534150613e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 41835016,
      "step": 2600
    },
    {
      "epoch": 0.18212543889603966,
      "eval_loss": 1.1404880285263062,
      "eval_runtime": 0.189,
      "eval_samples_per_second": 5.291,
      "eval_steps_per_second": 5.291,
      "num_input_tokens_seen": 41835016,
      "step": 2600
    },
    {
      "epoch": 0.1821954871417689,
      "grad_norm": 3.8487257957458496,
      "learning_rate": 8.179755516637479e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 41851400,
      "step": 2601
    },
    {
      "epoch": 0.18226553538749815,
      "grad_norm": 4.221493244171143,
      "learning_rate": 8.179055691768828e-05,
      "loss": 1.1346,
      "num_input_tokens_seen": 41867784,
      "step": 2602
    },
    {
      "epoch": 0.18233558363322738,
      "grad_norm": 3.88747239112854,
      "learning_rate": 8.178355866900175e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 41884024,
      "step": 2603
    },
    {
      "epoch": 0.18240563187895664,
      "grad_norm": 3.845551013946533,
      "learning_rate": 8.177656042031524e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 41899936,
      "step": 2604
    },
    {
      "epoch": 0.18247568012468587,
      "grad_norm": 3.6964564323425293,
      "learning_rate": 8.176956217162873e-05,
      "loss": 0.815,
      "num_input_tokens_seen": 41915512,
      "step": 2605
    },
    {
      "epoch": 0.18254572837041513,
      "grad_norm": 4.105105400085449,
      "learning_rate": 8.17625639229422e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 41931728,
      "step": 2606
    },
    {
      "epoch": 0.18261577661614437,
      "grad_norm": 5.0245842933654785,
      "learning_rate": 8.175556567425569e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 41947528,
      "step": 2607
    },
    {
      "epoch": 0.18268582486187362,
      "grad_norm": 3.4683709144592285,
      "learning_rate": 8.174856742556918e-05,
      "loss": 0.8183,
      "num_input_tokens_seen": 41963912,
      "step": 2608
    },
    {
      "epoch": 0.18275587310760286,
      "grad_norm": 4.603201866149902,
      "learning_rate": 8.174156917688267e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 41979976,
      "step": 2609
    },
    {
      "epoch": 0.18282592135333212,
      "grad_norm": 3.9904422760009766,
      "learning_rate": 8.173457092819615e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 41996360,
      "step": 2610
    },
    {
      "epoch": 0.18289596959906135,
      "grad_norm": 6.363785743713379,
      "learning_rate": 8.172757267950963e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 42011712,
      "step": 2611
    },
    {
      "epoch": 0.1829660178447906,
      "grad_norm": 3.7257959842681885,
      "learning_rate": 8.172057443082312e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 42028096,
      "step": 2612
    },
    {
      "epoch": 0.18303606609051984,
      "grad_norm": 3.8486809730529785,
      "learning_rate": 8.171357618213661e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 42044480,
      "step": 2613
    },
    {
      "epoch": 0.1831061143362491,
      "grad_norm": 4.620292663574219,
      "learning_rate": 8.17065779334501e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 42060864,
      "step": 2614
    },
    {
      "epoch": 0.18317616258197833,
      "grad_norm": 3.52644944190979,
      "learning_rate": 8.169957968476359e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 42077072,
      "step": 2615
    },
    {
      "epoch": 0.1832462108277076,
      "grad_norm": 3.800718069076538,
      "learning_rate": 8.169258143607706e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 42093128,
      "step": 2616
    },
    {
      "epoch": 0.18331625907343682,
      "grad_norm": 3.9447405338287354,
      "learning_rate": 8.168558318739054e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 42109512,
      "step": 2617
    },
    {
      "epoch": 0.18338630731916608,
      "grad_norm": 4.109864234924316,
      "learning_rate": 8.167858493870403e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 42125776,
      "step": 2618
    },
    {
      "epoch": 0.1834563555648953,
      "grad_norm": 3.6538870334625244,
      "learning_rate": 8.167158669001752e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 42141760,
      "step": 2619
    },
    {
      "epoch": 0.18352640381062457,
      "grad_norm": 5.139223098754883,
      "learning_rate": 8.1664588441331e-05,
      "loss": 1.2108,
      "num_input_tokens_seen": 42157096,
      "step": 2620
    },
    {
      "epoch": 0.1835964520563538,
      "grad_norm": 4.420098781585693,
      "learning_rate": 8.165759019264449e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 42173480,
      "step": 2621
    },
    {
      "epoch": 0.18366650030208306,
      "grad_norm": 5.559954643249512,
      "learning_rate": 8.165059194395798e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 42188944,
      "step": 2622
    },
    {
      "epoch": 0.1837365485478123,
      "grad_norm": 3.827627420425415,
      "learning_rate": 8.164359369527146e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 42204952,
      "step": 2623
    },
    {
      "epoch": 0.18380659679354155,
      "grad_norm": 7.454338550567627,
      "learning_rate": 8.163659544658494e-05,
      "loss": 1.2441,
      "num_input_tokens_seen": 42221336,
      "step": 2624
    },
    {
      "epoch": 0.1838766450392708,
      "grad_norm": 4.34182071685791,
      "learning_rate": 8.162959719789842e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 42237720,
      "step": 2625
    },
    {
      "epoch": 0.18394669328500005,
      "grad_norm": 3.9157843589782715,
      "learning_rate": 8.162259894921191e-05,
      "loss": 1.1206,
      "num_input_tokens_seen": 42253584,
      "step": 2626
    },
    {
      "epoch": 0.18401674153072928,
      "grad_norm": 3.952451467514038,
      "learning_rate": 8.16156007005254e-05,
      "loss": 1.2001,
      "num_input_tokens_seen": 42269968,
      "step": 2627
    },
    {
      "epoch": 0.18408678977645854,
      "grad_norm": 5.0545148849487305,
      "learning_rate": 8.160860245183889e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 42286232,
      "step": 2628
    },
    {
      "epoch": 0.18415683802218777,
      "grad_norm": 7.176907062530518,
      "learning_rate": 8.160160420315237e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 42302616,
      "step": 2629
    },
    {
      "epoch": 0.18422688626791703,
      "grad_norm": 3.994748830795288,
      "learning_rate": 8.159460595446585e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 42319000,
      "step": 2630
    },
    {
      "epoch": 0.18429693451364626,
      "grad_norm": 3.5744547843933105,
      "learning_rate": 8.158760770577934e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 42335384,
      "step": 2631
    },
    {
      "epoch": 0.18436698275937552,
      "grad_norm": 3.528723955154419,
      "learning_rate": 8.158060945709283e-05,
      "loss": 0.9955,
      "num_input_tokens_seen": 42351768,
      "step": 2632
    },
    {
      "epoch": 0.18443703100510478,
      "grad_norm": 3.9958291053771973,
      "learning_rate": 8.15736112084063e-05,
      "loss": 1.076,
      "num_input_tokens_seen": 42368152,
      "step": 2633
    },
    {
      "epoch": 0.184507079250834,
      "grad_norm": 4.1659650802612305,
      "learning_rate": 8.15666129597198e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 42384536,
      "step": 2634
    },
    {
      "epoch": 0.18457712749656327,
      "grad_norm": 5.116000652313232,
      "learning_rate": 8.155961471103328e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 42399704,
      "step": 2635
    },
    {
      "epoch": 0.1846471757422925,
      "grad_norm": 4.01514196395874,
      "learning_rate": 8.155261646234677e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 42416056,
      "step": 2636
    },
    {
      "epoch": 0.18471722398802176,
      "grad_norm": 4.290152072906494,
      "learning_rate": 8.154561821366024e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 42431544,
      "step": 2637
    },
    {
      "epoch": 0.184787272233751,
      "grad_norm": 4.267684459686279,
      "learning_rate": 8.153861996497373e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 42447928,
      "step": 2638
    },
    {
      "epoch": 0.18485732047948025,
      "grad_norm": 3.593191385269165,
      "learning_rate": 8.153162171628722e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 42464312,
      "step": 2639
    },
    {
      "epoch": 0.18492736872520948,
      "grad_norm": 4.322700023651123,
      "learning_rate": 8.152462346760071e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 42480696,
      "step": 2640
    },
    {
      "epoch": 0.18499741697093874,
      "grad_norm": 4.176753520965576,
      "learning_rate": 8.15176252189142e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 42496520,
      "step": 2641
    },
    {
      "epoch": 0.18506746521666798,
      "grad_norm": 4.405294895172119,
      "learning_rate": 8.151062697022769e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 42512904,
      "step": 2642
    },
    {
      "epoch": 0.18513751346239724,
      "grad_norm": 3.8770217895507812,
      "learning_rate": 8.150362872154116e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 42529288,
      "step": 2643
    },
    {
      "epoch": 0.18520756170812647,
      "grad_norm": 3.5909271240234375,
      "learning_rate": 8.149663047285464e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 42545672,
      "step": 2644
    },
    {
      "epoch": 0.18527760995385573,
      "grad_norm": 3.73958420753479,
      "learning_rate": 8.148963222416812e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 42562056,
      "step": 2645
    },
    {
      "epoch": 0.18534765819958496,
      "grad_norm": 3.6813879013061523,
      "learning_rate": 8.148263397548161e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 42577720,
      "step": 2646
    },
    {
      "epoch": 0.18541770644531422,
      "grad_norm": 4.13958215713501,
      "learning_rate": 8.14756357267951e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 42594104,
      "step": 2647
    },
    {
      "epoch": 0.18548775469104345,
      "grad_norm": 3.757805109024048,
      "learning_rate": 8.146863747810859e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 42610256,
      "step": 2648
    },
    {
      "epoch": 0.1855578029367727,
      "grad_norm": 4.57798957824707,
      "learning_rate": 8.146163922942208e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 42626424,
      "step": 2649
    },
    {
      "epoch": 0.18562785118250194,
      "grad_norm": 3.797257423400879,
      "learning_rate": 8.145464098073555e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 42642200,
      "step": 2650
    },
    {
      "epoch": 0.1856978994282312,
      "grad_norm": 4.258513450622559,
      "learning_rate": 8.144764273204904e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 42657416,
      "step": 2651
    },
    {
      "epoch": 0.18576794767396043,
      "grad_norm": 4.369161605834961,
      "learning_rate": 8.144064448336252e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 42673752,
      "step": 2652
    },
    {
      "epoch": 0.1858379959196897,
      "grad_norm": 4.159987926483154,
      "learning_rate": 8.1433646234676e-05,
      "loss": 1.151,
      "num_input_tokens_seen": 42690136,
      "step": 2653
    },
    {
      "epoch": 0.18590804416541892,
      "grad_norm": 7.164428234100342,
      "learning_rate": 8.14266479859895e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 42706520,
      "step": 2654
    },
    {
      "epoch": 0.18597809241114818,
      "grad_norm": 3.4230172634124756,
      "learning_rate": 8.141964973730298e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 42722904,
      "step": 2655
    },
    {
      "epoch": 0.18604814065687741,
      "grad_norm": 4.316817283630371,
      "learning_rate": 8.141265148861647e-05,
      "loss": 1.2645,
      "num_input_tokens_seen": 42738640,
      "step": 2656
    },
    {
      "epoch": 0.18611818890260667,
      "grad_norm": 3.7894318103790283,
      "learning_rate": 8.140565323992995e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 42754848,
      "step": 2657
    },
    {
      "epoch": 0.1861882371483359,
      "grad_norm": 4.198835849761963,
      "learning_rate": 8.139865499124344e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 42771232,
      "step": 2658
    },
    {
      "epoch": 0.18625828539406517,
      "grad_norm": 3.796414852142334,
      "learning_rate": 8.139165674255692e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 42787344,
      "step": 2659
    },
    {
      "epoch": 0.1863283336397944,
      "grad_norm": 3.6421244144439697,
      "learning_rate": 8.138465849387041e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 42803728,
      "step": 2660
    },
    {
      "epoch": 0.18639838188552366,
      "grad_norm": 4.391780376434326,
      "learning_rate": 8.13776602451839e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 42820112,
      "step": 2661
    },
    {
      "epoch": 0.1864684301312529,
      "grad_norm": 4.187370777130127,
      "learning_rate": 8.137066199649738e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 42836496,
      "step": 2662
    },
    {
      "epoch": 0.18653847837698215,
      "grad_norm": 3.794281244277954,
      "learning_rate": 8.136366374781086e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 42852880,
      "step": 2663
    },
    {
      "epoch": 0.18660852662271138,
      "grad_norm": 4.386116981506348,
      "learning_rate": 8.135666549912434e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 42869264,
      "step": 2664
    },
    {
      "epoch": 0.18667857486844064,
      "grad_norm": 4.223747253417969,
      "learning_rate": 8.134966725043783e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 42885528,
      "step": 2665
    },
    {
      "epoch": 0.1867486231141699,
      "grad_norm": 5.020680904388428,
      "learning_rate": 8.134266900175132e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 42901912,
      "step": 2666
    },
    {
      "epoch": 0.18681867135989913,
      "grad_norm": 8.58757495880127,
      "learning_rate": 8.13356707530648e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 42917072,
      "step": 2667
    },
    {
      "epoch": 0.1868887196056284,
      "grad_norm": 3.6986234188079834,
      "learning_rate": 8.13286725043783e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 42933296,
      "step": 2668
    },
    {
      "epoch": 0.18695876785135762,
      "grad_norm": 4.196423053741455,
      "learning_rate": 8.132167425569178e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 42948968,
      "step": 2669
    },
    {
      "epoch": 0.18702881609708688,
      "grad_norm": 4.019235610961914,
      "learning_rate": 8.131467600700526e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 42965352,
      "step": 2670
    },
    {
      "epoch": 0.1870988643428161,
      "grad_norm": 4.035778045654297,
      "learning_rate": 8.130767775831873e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 42980872,
      "step": 2671
    },
    {
      "epoch": 0.18716891258854537,
      "grad_norm": 4.193873882293701,
      "learning_rate": 8.130067950963222e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 42996848,
      "step": 2672
    },
    {
      "epoch": 0.1872389608342746,
      "grad_norm": 4.011183738708496,
      "learning_rate": 8.129368126094571e-05,
      "loss": 1.1151,
      "num_input_tokens_seen": 43012728,
      "step": 2673
    },
    {
      "epoch": 0.18730900908000386,
      "grad_norm": 5.662332534790039,
      "learning_rate": 8.12866830122592e-05,
      "loss": 1.1238,
      "num_input_tokens_seen": 43028728,
      "step": 2674
    },
    {
      "epoch": 0.1873790573257331,
      "grad_norm": 4.4699387550354,
      "learning_rate": 8.127968476357269e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 43044504,
      "step": 2675
    },
    {
      "epoch": 0.18744910557146235,
      "grad_norm": 3.857011556625366,
      "learning_rate": 8.127268651488618e-05,
      "loss": 0.9866,
      "num_input_tokens_seen": 43060496,
      "step": 2676
    },
    {
      "epoch": 0.1875191538171916,
      "grad_norm": 3.5458414554595947,
      "learning_rate": 8.126568826619965e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 43076880,
      "step": 2677
    },
    {
      "epoch": 0.18758920206292085,
      "grad_norm": 5.4007744789123535,
      "learning_rate": 8.125869001751314e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 43091368,
      "step": 2678
    },
    {
      "epoch": 0.18765925030865008,
      "grad_norm": 5.15717077255249,
      "learning_rate": 8.125169176882661e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 43107752,
      "step": 2679
    },
    {
      "epoch": 0.18772929855437934,
      "grad_norm": 4.891427516937256,
      "learning_rate": 8.124469352014012e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 43122808,
      "step": 2680
    },
    {
      "epoch": 0.18779934680010857,
      "grad_norm": 4.651966571807861,
      "learning_rate": 8.123769527145359e-05,
      "loss": 1.1457,
      "num_input_tokens_seen": 43139056,
      "step": 2681
    },
    {
      "epoch": 0.18786939504583783,
      "grad_norm": 3.844129800796509,
      "learning_rate": 8.123069702276708e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 43155440,
      "step": 2682
    },
    {
      "epoch": 0.18793944329156706,
      "grad_norm": 3.669360876083374,
      "learning_rate": 8.122369877408057e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 43171824,
      "step": 2683
    },
    {
      "epoch": 0.18800949153729632,
      "grad_norm": 3.6102185249328613,
      "learning_rate": 8.121670052539404e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 43188208,
      "step": 2684
    },
    {
      "epoch": 0.18807953978302555,
      "grad_norm": 3.593414783477783,
      "learning_rate": 8.120970227670753e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 43204248,
      "step": 2685
    },
    {
      "epoch": 0.1881495880287548,
      "grad_norm": 5.017848491668701,
      "learning_rate": 8.120270402802102e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 43219808,
      "step": 2686
    },
    {
      "epoch": 0.18821963627448404,
      "grad_norm": 4.083794593811035,
      "learning_rate": 8.119570577933451e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 43235712,
      "step": 2687
    },
    {
      "epoch": 0.1882896845202133,
      "grad_norm": 4.265167713165283,
      "learning_rate": 8.1188707530648e-05,
      "loss": 1.2967,
      "num_input_tokens_seen": 43252048,
      "step": 2688
    },
    {
      "epoch": 0.18835973276594253,
      "grad_norm": 4.24991512298584,
      "learning_rate": 8.118170928196147e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 43268152,
      "step": 2689
    },
    {
      "epoch": 0.1884297810116718,
      "grad_norm": 4.059658050537109,
      "learning_rate": 8.117471103327496e-05,
      "loss": 1.1356,
      "num_input_tokens_seen": 43284240,
      "step": 2690
    },
    {
      "epoch": 0.18849982925740102,
      "grad_norm": 4.807305812835693,
      "learning_rate": 8.116771278458844e-05,
      "loss": 1.0424,
      "num_input_tokens_seen": 43299368,
      "step": 2691
    },
    {
      "epoch": 0.18856987750313028,
      "grad_norm": 5.590726852416992,
      "learning_rate": 8.116071453590193e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 43315648,
      "step": 2692
    },
    {
      "epoch": 0.18863992574885952,
      "grad_norm": 5.114964485168457,
      "learning_rate": 8.115371628721541e-05,
      "loss": 0.8916,
      "num_input_tokens_seen": 43331688,
      "step": 2693
    },
    {
      "epoch": 0.18870997399458878,
      "grad_norm": 4.323836803436279,
      "learning_rate": 8.11467180385289e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 43346672,
      "step": 2694
    },
    {
      "epoch": 0.188780022240318,
      "grad_norm": 4.290014743804932,
      "learning_rate": 8.113971978984239e-05,
      "loss": 1.2498,
      "num_input_tokens_seen": 43362872,
      "step": 2695
    },
    {
      "epoch": 0.18885007048604727,
      "grad_norm": 3.5292484760284424,
      "learning_rate": 8.113272154115588e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 43379256,
      "step": 2696
    },
    {
      "epoch": 0.1889201187317765,
      "grad_norm": 4.21523380279541,
      "learning_rate": 8.112572329246935e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 43395152,
      "step": 2697
    },
    {
      "epoch": 0.18899016697750576,
      "grad_norm": 4.900782108306885,
      "learning_rate": 8.111872504378283e-05,
      "loss": 1.1038,
      "num_input_tokens_seen": 43411536,
      "step": 2698
    },
    {
      "epoch": 0.189060215223235,
      "grad_norm": 3.613231658935547,
      "learning_rate": 8.111172679509632e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 43427920,
      "step": 2699
    },
    {
      "epoch": 0.18913026346896425,
      "grad_norm": 3.681725263595581,
      "learning_rate": 8.110472854640982e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 43444304,
      "step": 2700
    },
    {
      "epoch": 0.1892003117146935,
      "grad_norm": 3.801785707473755,
      "learning_rate": 8.10977302977233e-05,
      "loss": 0.9856,
      "num_input_tokens_seen": 43459960,
      "step": 2701
    },
    {
      "epoch": 0.18927035996042274,
      "grad_norm": 3.4208626747131348,
      "learning_rate": 8.109073204903678e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 43476344,
      "step": 2702
    },
    {
      "epoch": 0.189340408206152,
      "grad_norm": 4.169189453125,
      "learning_rate": 8.108373380035027e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 43492728,
      "step": 2703
    },
    {
      "epoch": 0.18941045645188123,
      "grad_norm": 3.7125117778778076,
      "learning_rate": 8.107673555166375e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 43508168,
      "step": 2704
    },
    {
      "epoch": 0.1894805046976105,
      "grad_norm": 4.550642490386963,
      "learning_rate": 8.106973730297724e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 43524480,
      "step": 2705
    },
    {
      "epoch": 0.18955055294333972,
      "grad_norm": 4.219499588012695,
      "learning_rate": 8.106273905429072e-05,
      "loss": 1.148,
      "num_input_tokens_seen": 43540864,
      "step": 2706
    },
    {
      "epoch": 0.18962060118906898,
      "grad_norm": 4.605996131896973,
      "learning_rate": 8.105574080560421e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 43557248,
      "step": 2707
    },
    {
      "epoch": 0.18969064943479821,
      "grad_norm": 3.740314245223999,
      "learning_rate": 8.104874255691769e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 43573632,
      "step": 2708
    },
    {
      "epoch": 0.18976069768052747,
      "grad_norm": 3.92555832862854,
      "learning_rate": 8.104174430823118e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 43589728,
      "step": 2709
    },
    {
      "epoch": 0.1898307459262567,
      "grad_norm": 3.5653927326202393,
      "learning_rate": 8.103474605954467e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 43606112,
      "step": 2710
    },
    {
      "epoch": 0.18990079417198596,
      "grad_norm": 5.943650245666504,
      "learning_rate": 8.102774781085814e-05,
      "loss": 1.309,
      "num_input_tokens_seen": 43621072,
      "step": 2711
    },
    {
      "epoch": 0.1899708424177152,
      "grad_norm": 3.7632322311401367,
      "learning_rate": 8.102074956217163e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 43636976,
      "step": 2712
    },
    {
      "epoch": 0.19004089066344446,
      "grad_norm": 3.605536699295044,
      "learning_rate": 8.101375131348512e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 43653360,
      "step": 2713
    },
    {
      "epoch": 0.1901109389091737,
      "grad_norm": 3.7717363834381104,
      "learning_rate": 8.10067530647986e-05,
      "loss": 0.9407,
      "num_input_tokens_seen": 43669488,
      "step": 2714
    },
    {
      "epoch": 0.19018098715490295,
      "grad_norm": 4.55484676361084,
      "learning_rate": 8.09997548161121e-05,
      "loss": 0.8501,
      "num_input_tokens_seen": 43684704,
      "step": 2715
    },
    {
      "epoch": 0.19025103540063218,
      "grad_norm": 4.155830383300781,
      "learning_rate": 8.099275656742557e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 43700112,
      "step": 2716
    },
    {
      "epoch": 0.19032108364636144,
      "grad_norm": 5.615505695343018,
      "learning_rate": 8.098575831873906e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 43716136,
      "step": 2717
    },
    {
      "epoch": 0.19039113189209067,
      "grad_norm": 4.60966157913208,
      "learning_rate": 8.097876007005253e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 43731576,
      "step": 2718
    },
    {
      "epoch": 0.19046118013781993,
      "grad_norm": 5.698062896728516,
      "learning_rate": 8.097176182136602e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 43747960,
      "step": 2719
    },
    {
      "epoch": 0.19053122838354916,
      "grad_norm": 3.760756492614746,
      "learning_rate": 8.096476357267952e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 43764304,
      "step": 2720
    },
    {
      "epoch": 0.19060127662927842,
      "grad_norm": 4.084067344665527,
      "learning_rate": 8.0957765323993e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 43780688,
      "step": 2721
    },
    {
      "epoch": 0.19067132487500765,
      "grad_norm": 3.9934301376342773,
      "learning_rate": 8.095076707530649e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 43797072,
      "step": 2722
    },
    {
      "epoch": 0.1907413731207369,
      "grad_norm": 3.915512800216675,
      "learning_rate": 8.094376882661998e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 43813456,
      "step": 2723
    },
    {
      "epoch": 0.19081142136646614,
      "grad_norm": 3.967040777206421,
      "learning_rate": 8.093677057793345e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 43829656,
      "step": 2724
    },
    {
      "epoch": 0.1908814696121954,
      "grad_norm": 3.707667827606201,
      "learning_rate": 8.092977232924693e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 43846040,
      "step": 2725
    },
    {
      "epoch": 0.19095151785792464,
      "grad_norm": 3.3822734355926514,
      "learning_rate": 8.092277408056043e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 43862144,
      "step": 2726
    },
    {
      "epoch": 0.1910215661036539,
      "grad_norm": 3.7703781127929688,
      "learning_rate": 8.091577583187392e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 43878328,
      "step": 2727
    },
    {
      "epoch": 0.19109161434938313,
      "grad_norm": 3.902003049850464,
      "learning_rate": 8.090877758318739e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 43893480,
      "step": 2728
    },
    {
      "epoch": 0.19116166259511239,
      "grad_norm": 3.971395969390869,
      "learning_rate": 8.090177933450088e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 43909752,
      "step": 2729
    },
    {
      "epoch": 0.19123171084084162,
      "grad_norm": 3.4233641624450684,
      "learning_rate": 8.089478108581437e-05,
      "loss": 0.8821,
      "num_input_tokens_seen": 43926136,
      "step": 2730
    },
    {
      "epoch": 0.19130175908657088,
      "grad_norm": 5.967614650726318,
      "learning_rate": 8.088778283712784e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 43941592,
      "step": 2731
    },
    {
      "epoch": 0.1913718073323001,
      "grad_norm": 4.431912899017334,
      "learning_rate": 8.088078458844133e-05,
      "loss": 1.2471,
      "num_input_tokens_seen": 43957784,
      "step": 2732
    },
    {
      "epoch": 0.19144185557802937,
      "grad_norm": 3.659182071685791,
      "learning_rate": 8.087378633975482e-05,
      "loss": 0.9701,
      "num_input_tokens_seen": 43973648,
      "step": 2733
    },
    {
      "epoch": 0.1915119038237586,
      "grad_norm": 4.983634948730469,
      "learning_rate": 8.086678809106831e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 43990032,
      "step": 2734
    },
    {
      "epoch": 0.19158195206948786,
      "grad_norm": 4.236748695373535,
      "learning_rate": 8.085978984238179e-05,
      "loss": 1.0724,
      "num_input_tokens_seen": 44005064,
      "step": 2735
    },
    {
      "epoch": 0.19165200031521712,
      "grad_norm": 3.3617727756500244,
      "learning_rate": 8.085279159369527e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 44021448,
      "step": 2736
    },
    {
      "epoch": 0.19172204856094635,
      "grad_norm": 3.4514083862304688,
      "learning_rate": 8.084579334500876e-05,
      "loss": 0.8738,
      "num_input_tokens_seen": 44037832,
      "step": 2737
    },
    {
      "epoch": 0.1917920968066756,
      "grad_norm": 4.126194000244141,
      "learning_rate": 8.083879509632224e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 44053384,
      "step": 2738
    },
    {
      "epoch": 0.19186214505240484,
      "grad_norm": 5.12385368347168,
      "learning_rate": 8.083179684763573e-05,
      "loss": 1.251,
      "num_input_tokens_seen": 44068728,
      "step": 2739
    },
    {
      "epoch": 0.1919321932981341,
      "grad_norm": 3.457253932952881,
      "learning_rate": 8.082479859894923e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 44085112,
      "step": 2740
    },
    {
      "epoch": 0.19200224154386333,
      "grad_norm": 3.8708858489990234,
      "learning_rate": 8.08178003502627e-05,
      "loss": 1.1838,
      "num_input_tokens_seen": 44101456,
      "step": 2741
    },
    {
      "epoch": 0.1920722897895926,
      "grad_norm": 4.175468921661377,
      "learning_rate": 8.081080210157619e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 44116640,
      "step": 2742
    },
    {
      "epoch": 0.19214233803532182,
      "grad_norm": 4.141748428344727,
      "learning_rate": 8.080380385288967e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 44132328,
      "step": 2743
    },
    {
      "epoch": 0.19221238628105108,
      "grad_norm": 5.1061692237854,
      "learning_rate": 8.079680560420316e-05,
      "loss": 1.172,
      "num_input_tokens_seen": 44148712,
      "step": 2744
    },
    {
      "epoch": 0.19228243452678032,
      "grad_norm": 3.990196704864502,
      "learning_rate": 8.078980735551663e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 44164600,
      "step": 2745
    },
    {
      "epoch": 0.19235248277250958,
      "grad_norm": 4.365367412567139,
      "learning_rate": 8.078280910683013e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 44180984,
      "step": 2746
    },
    {
      "epoch": 0.1924225310182388,
      "grad_norm": 4.092031002044678,
      "learning_rate": 8.077581085814362e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 44196400,
      "step": 2747
    },
    {
      "epoch": 0.19249257926396807,
      "grad_norm": 3.4052438735961914,
      "learning_rate": 8.07688126094571e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 44212736,
      "step": 2748
    },
    {
      "epoch": 0.1925626275096973,
      "grad_norm": 4.703436374664307,
      "learning_rate": 8.076181436077059e-05,
      "loss": 1.2058,
      "num_input_tokens_seen": 44229120,
      "step": 2749
    },
    {
      "epoch": 0.19263267575542656,
      "grad_norm": 3.7579853534698486,
      "learning_rate": 8.075481611208407e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 44245144,
      "step": 2750
    },
    {
      "epoch": 0.1927027240011558,
      "grad_norm": 3.6251869201660156,
      "learning_rate": 8.074781786339755e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 44260920,
      "step": 2751
    },
    {
      "epoch": 0.19277277224688505,
      "grad_norm": 3.4949889183044434,
      "learning_rate": 8.074081961471104e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 44277280,
      "step": 2752
    },
    {
      "epoch": 0.19284282049261428,
      "grad_norm": 4.28520393371582,
      "learning_rate": 8.073382136602453e-05,
      "loss": 1.2536,
      "num_input_tokens_seen": 44293664,
      "step": 2753
    },
    {
      "epoch": 0.19291286873834354,
      "grad_norm": 3.9574859142303467,
      "learning_rate": 8.072682311733801e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 44309328,
      "step": 2754
    },
    {
      "epoch": 0.19298291698407277,
      "grad_norm": 3.6340646743774414,
      "learning_rate": 8.071982486865149e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 44325336,
      "step": 2755
    },
    {
      "epoch": 0.19305296522980203,
      "grad_norm": 5.131178855895996,
      "learning_rate": 8.071282661996498e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 44341264,
      "step": 2756
    },
    {
      "epoch": 0.19312301347553126,
      "grad_norm": 4.273870944976807,
      "learning_rate": 8.070582837127847e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 44357648,
      "step": 2757
    },
    {
      "epoch": 0.19319306172126052,
      "grad_norm": 3.883690118789673,
      "learning_rate": 8.069883012259194e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 44373984,
      "step": 2758
    },
    {
      "epoch": 0.19326310996698975,
      "grad_norm": 4.284129619598389,
      "learning_rate": 8.069183187390543e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 44389160,
      "step": 2759
    },
    {
      "epoch": 0.19333315821271901,
      "grad_norm": 4.517998695373535,
      "learning_rate": 8.068483362521892e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 44405544,
      "step": 2760
    },
    {
      "epoch": 0.19340320645844825,
      "grad_norm": 4.098707675933838,
      "learning_rate": 8.067783537653241e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 44421560,
      "step": 2761
    },
    {
      "epoch": 0.1934732547041775,
      "grad_norm": 3.656461477279663,
      "learning_rate": 8.067083712784588e-05,
      "loss": 0.8462,
      "num_input_tokens_seen": 44437944,
      "step": 2762
    },
    {
      "epoch": 0.19354330294990674,
      "grad_norm": 3.8305914402008057,
      "learning_rate": 8.066383887915937e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 44454208,
      "step": 2763
    },
    {
      "epoch": 0.193613351195636,
      "grad_norm": 4.0582990646362305,
      "learning_rate": 8.065684063047286e-05,
      "loss": 1.2152,
      "num_input_tokens_seen": 44470592,
      "step": 2764
    },
    {
      "epoch": 0.19368339944136523,
      "grad_norm": 4.159184455871582,
      "learning_rate": 8.064984238178633e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 44486976,
      "step": 2765
    },
    {
      "epoch": 0.1937534476870945,
      "grad_norm": 3.7490620613098145,
      "learning_rate": 8.064284413309984e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 44503360,
      "step": 2766
    },
    {
      "epoch": 0.19382349593282372,
      "grad_norm": 4.3000288009643555,
      "learning_rate": 8.063584588441333e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 44519744,
      "step": 2767
    },
    {
      "epoch": 0.19389354417855298,
      "grad_norm": 3.9175477027893066,
      "learning_rate": 8.06288476357268e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 44535664,
      "step": 2768
    },
    {
      "epoch": 0.1939635924242822,
      "grad_norm": 4.4328293800354,
      "learning_rate": 8.062184938704029e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 44550984,
      "step": 2769
    },
    {
      "epoch": 0.19403364067001147,
      "grad_norm": 4.556321620941162,
      "learning_rate": 8.061485113835376e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 44566808,
      "step": 2770
    },
    {
      "epoch": 0.19410368891574073,
      "grad_norm": 4.382159233093262,
      "learning_rate": 8.060785288966725e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 44583192,
      "step": 2771
    },
    {
      "epoch": 0.19417373716146996,
      "grad_norm": 3.920137405395508,
      "learning_rate": 8.060085464098074e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 44599480,
      "step": 2772
    },
    {
      "epoch": 0.19424378540719922,
      "grad_norm": 4.23013162612915,
      "learning_rate": 8.059385639229423e-05,
      "loss": 1.2268,
      "num_input_tokens_seen": 44615240,
      "step": 2773
    },
    {
      "epoch": 0.19431383365292845,
      "grad_norm": 3.7917346954345703,
      "learning_rate": 8.058685814360772e-05,
      "loss": 1.0935,
      "num_input_tokens_seen": 44630952,
      "step": 2774
    },
    {
      "epoch": 0.1943838818986577,
      "grad_norm": 4.798681259155273,
      "learning_rate": 8.05798598949212e-05,
      "loss": 1.1321,
      "num_input_tokens_seen": 44647336,
      "step": 2775
    },
    {
      "epoch": 0.19445393014438694,
      "grad_norm": 3.563124418258667,
      "learning_rate": 8.057286164623468e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 44663720,
      "step": 2776
    },
    {
      "epoch": 0.1945239783901162,
      "grad_norm": 6.6064019203186035,
      "learning_rate": 8.056586339754817e-05,
      "loss": 0.8685,
      "num_input_tokens_seen": 44679616,
      "step": 2777
    },
    {
      "epoch": 0.19459402663584544,
      "grad_norm": 4.1651291847229,
      "learning_rate": 8.055886514886165e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 44695800,
      "step": 2778
    },
    {
      "epoch": 0.1946640748815747,
      "grad_norm": 3.929474353790283,
      "learning_rate": 8.055186690017513e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 44711744,
      "step": 2779
    },
    {
      "epoch": 0.19473412312730393,
      "grad_norm": 3.758721351623535,
      "learning_rate": 8.054486865148862e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 44728128,
      "step": 2780
    },
    {
      "epoch": 0.19480417137303319,
      "grad_norm": 4.988550662994385,
      "learning_rate": 8.053787040280211e-05,
      "loss": 1.222,
      "num_input_tokens_seen": 44744512,
      "step": 2781
    },
    {
      "epoch": 0.19487421961876242,
      "grad_norm": 3.7875940799713135,
      "learning_rate": 8.053087215411559e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 44760896,
      "step": 2782
    },
    {
      "epoch": 0.19494426786449168,
      "grad_norm": 3.877729654312134,
      "learning_rate": 8.052387390542908e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 44777280,
      "step": 2783
    },
    {
      "epoch": 0.1950143161102209,
      "grad_norm": 4.979894161224365,
      "learning_rate": 8.051687565674256e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 44793664,
      "step": 2784
    },
    {
      "epoch": 0.19508436435595017,
      "grad_norm": 4.3148579597473145,
      "learning_rate": 8.050987740805604e-05,
      "loss": 1.1587,
      "num_input_tokens_seen": 44809688,
      "step": 2785
    },
    {
      "epoch": 0.1951544126016794,
      "grad_norm": 4.082404136657715,
      "learning_rate": 8.050287915936954e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 44825600,
      "step": 2786
    },
    {
      "epoch": 0.19522446084740866,
      "grad_norm": 3.6951189041137695,
      "learning_rate": 8.049588091068302e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 44841984,
      "step": 2787
    },
    {
      "epoch": 0.1952945090931379,
      "grad_norm": 3.797136068344116,
      "learning_rate": 8.04888826619965e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 44858368,
      "step": 2788
    },
    {
      "epoch": 0.19536455733886715,
      "grad_norm": 3.8912811279296875,
      "learning_rate": 8.048188441330998e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 44873752,
      "step": 2789
    },
    {
      "epoch": 0.19543460558459638,
      "grad_norm": 4.355793476104736,
      "learning_rate": 8.047488616462347e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 44889336,
      "step": 2790
    },
    {
      "epoch": 0.19550465383032564,
      "grad_norm": 4.216153144836426,
      "learning_rate": 8.046788791593696e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 44905720,
      "step": 2791
    },
    {
      "epoch": 0.19557470207605487,
      "grad_norm": 3.5346696376800537,
      "learning_rate": 8.046088966725045e-05,
      "loss": 1.1628,
      "num_input_tokens_seen": 44921864,
      "step": 2792
    },
    {
      "epoch": 0.19564475032178413,
      "grad_norm": 4.2197489738464355,
      "learning_rate": 8.045389141856393e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 44938248,
      "step": 2793
    },
    {
      "epoch": 0.19571479856751337,
      "grad_norm": 3.66995906829834,
      "learning_rate": 8.044689316987742e-05,
      "loss": 1.0401,
      "num_input_tokens_seen": 44954632,
      "step": 2794
    },
    {
      "epoch": 0.19578484681324262,
      "grad_norm": 5.062297821044922,
      "learning_rate": 8.04398949211909e-05,
      "loss": 1.2106,
      "num_input_tokens_seen": 44971016,
      "step": 2795
    },
    {
      "epoch": 0.19585489505897186,
      "grad_norm": 4.473872661590576,
      "learning_rate": 8.043289667250439e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 44987400,
      "step": 2796
    },
    {
      "epoch": 0.19592494330470112,
      "grad_norm": 4.724556922912598,
      "learning_rate": 8.042589842381786e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 45002968,
      "step": 2797
    },
    {
      "epoch": 0.19599499155043035,
      "grad_norm": 4.324196815490723,
      "learning_rate": 8.041890017513135e-05,
      "loss": 0.8708,
      "num_input_tokens_seen": 45019352,
      "step": 2798
    },
    {
      "epoch": 0.1960650397961596,
      "grad_norm": 4.309204578399658,
      "learning_rate": 8.041190192644484e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 45034960,
      "step": 2799
    },
    {
      "epoch": 0.19613508804188884,
      "grad_norm": 3.4928808212280273,
      "learning_rate": 8.040490367775833e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 45051344,
      "step": 2800
    },
    {
      "epoch": 0.19613508804188884,
      "eval_loss": 1.1401225328445435,
      "eval_runtime": 0.185,
      "eval_samples_per_second": 5.405,
      "eval_steps_per_second": 5.405,
      "num_input_tokens_seen": 45051344,
      "step": 2800
    },
    {
      "epoch": 0.1962051362876181,
      "grad_norm": 6.388762474060059,
      "learning_rate": 8.039790542907182e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 45066712,
      "step": 2801
    },
    {
      "epoch": 0.19627518453334733,
      "grad_norm": 3.8386781215667725,
      "learning_rate": 8.039090718038529e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 45082472,
      "step": 2802
    },
    {
      "epoch": 0.1963452327790766,
      "grad_norm": 3.540064573287964,
      "learning_rate": 8.038390893169878e-05,
      "loss": 0.846,
      "num_input_tokens_seen": 45098072,
      "step": 2803
    },
    {
      "epoch": 0.19641528102480582,
      "grad_norm": 3.9858322143554688,
      "learning_rate": 8.037691068301227e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 45114456,
      "step": 2804
    },
    {
      "epoch": 0.19648532927053508,
      "grad_norm": 4.418299674987793,
      "learning_rate": 8.036991243432574e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 45130416,
      "step": 2805
    },
    {
      "epoch": 0.19655537751626434,
      "grad_norm": 4.6108880043029785,
      "learning_rate": 8.036291418563923e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 45146800,
      "step": 2806
    },
    {
      "epoch": 0.19662542576199357,
      "grad_norm": 3.686781883239746,
      "learning_rate": 8.035591593695272e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 45163016,
      "step": 2807
    },
    {
      "epoch": 0.19669547400772283,
      "grad_norm": 3.7459771633148193,
      "learning_rate": 8.034891768826621e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 45178912,
      "step": 2808
    },
    {
      "epoch": 0.19676552225345206,
      "grad_norm": 4.955589771270752,
      "learning_rate": 8.034191943957968e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 45193928,
      "step": 2809
    },
    {
      "epoch": 0.19683557049918132,
      "grad_norm": 4.901642322540283,
      "learning_rate": 8.033492119089317e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 45209080,
      "step": 2810
    },
    {
      "epoch": 0.19690561874491055,
      "grad_norm": 3.685493230819702,
      "learning_rate": 8.032792294220666e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 45225400,
      "step": 2811
    },
    {
      "epoch": 0.19697566699063981,
      "grad_norm": 4.731873512268066,
      "learning_rate": 8.032092469352015e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 45241152,
      "step": 2812
    },
    {
      "epoch": 0.19704571523636905,
      "grad_norm": 4.52595853805542,
      "learning_rate": 8.031392644483364e-05,
      "loss": 1.142,
      "num_input_tokens_seen": 45256976,
      "step": 2813
    },
    {
      "epoch": 0.1971157634820983,
      "grad_norm": 4.4693074226379395,
      "learning_rate": 8.030692819614711e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 45273360,
      "step": 2814
    },
    {
      "epoch": 0.19718581172782754,
      "grad_norm": 5.599058151245117,
      "learning_rate": 8.02999299474606e-05,
      "loss": 1.1544,
      "num_input_tokens_seen": 45289744,
      "step": 2815
    },
    {
      "epoch": 0.1972558599735568,
      "grad_norm": 3.758751153945923,
      "learning_rate": 8.029293169877408e-05,
      "loss": 1.1877,
      "num_input_tokens_seen": 45305960,
      "step": 2816
    },
    {
      "epoch": 0.19732590821928603,
      "grad_norm": 4.059335231781006,
      "learning_rate": 8.028593345008757e-05,
      "loss": 1.0294,
      "num_input_tokens_seen": 45321536,
      "step": 2817
    },
    {
      "epoch": 0.1973959564650153,
      "grad_norm": 3.8090553283691406,
      "learning_rate": 8.027893520140105e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 45337920,
      "step": 2818
    },
    {
      "epoch": 0.19746600471074452,
      "grad_norm": 3.7900006771087646,
      "learning_rate": 8.027193695271454e-05,
      "loss": 1.2042,
      "num_input_tokens_seen": 45353632,
      "step": 2819
    },
    {
      "epoch": 0.19753605295647378,
      "grad_norm": 4.279977321624756,
      "learning_rate": 8.026493870402803e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 45369712,
      "step": 2820
    },
    {
      "epoch": 0.197606101202203,
      "grad_norm": 3.7417356967926025,
      "learning_rate": 8.025794045534152e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 45384816,
      "step": 2821
    },
    {
      "epoch": 0.19767614944793227,
      "grad_norm": 4.084759712219238,
      "learning_rate": 8.0250942206655e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 45401200,
      "step": 2822
    },
    {
      "epoch": 0.1977461976936615,
      "grad_norm": 4.963731288909912,
      "learning_rate": 8.024394395796848e-05,
      "loss": 1.2548,
      "num_input_tokens_seen": 45417096,
      "step": 2823
    },
    {
      "epoch": 0.19781624593939076,
      "grad_norm": 4.115303993225098,
      "learning_rate": 8.023694570928196e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 45433480,
      "step": 2824
    },
    {
      "epoch": 0.19788629418512,
      "grad_norm": 3.908439874649048,
      "learning_rate": 8.022994746059545e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 45448984,
      "step": 2825
    },
    {
      "epoch": 0.19795634243084925,
      "grad_norm": 4.0723090171813965,
      "learning_rate": 8.022294921190894e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 45465192,
      "step": 2826
    },
    {
      "epoch": 0.19802639067657848,
      "grad_norm": 4.219478607177734,
      "learning_rate": 8.021595096322242e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 45480904,
      "step": 2827
    },
    {
      "epoch": 0.19809643892230774,
      "grad_norm": 4.246188163757324,
      "learning_rate": 8.020895271453591e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 45497288,
      "step": 2828
    },
    {
      "epoch": 0.19816648716803698,
      "grad_norm": 4.898525714874268,
      "learning_rate": 8.020195446584939e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 45513456,
      "step": 2829
    },
    {
      "epoch": 0.19823653541376623,
      "grad_norm": 4.1247239112854,
      "learning_rate": 8.019495621716288e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 45528752,
      "step": 2830
    },
    {
      "epoch": 0.19830658365949547,
      "grad_norm": 4.352110385894775,
      "learning_rate": 8.018795796847636e-05,
      "loss": 1.3602,
      "num_input_tokens_seen": 45545136,
      "step": 2831
    },
    {
      "epoch": 0.19837663190522473,
      "grad_norm": 3.731719732284546,
      "learning_rate": 8.018095971978985e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 45561160,
      "step": 2832
    },
    {
      "epoch": 0.19844668015095396,
      "grad_norm": 4.234768867492676,
      "learning_rate": 8.017396147110333e-05,
      "loss": 1.2279,
      "num_input_tokens_seen": 45577288,
      "step": 2833
    },
    {
      "epoch": 0.19851672839668322,
      "grad_norm": 4.682285308837891,
      "learning_rate": 8.016696322241682e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 45593152,
      "step": 2834
    },
    {
      "epoch": 0.19858677664241245,
      "grad_norm": 4.576408863067627,
      "learning_rate": 8.01599649737303e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 45609408,
      "step": 2835
    },
    {
      "epoch": 0.1986568248881417,
      "grad_norm": 4.209808826446533,
      "learning_rate": 8.015296672504378e-05,
      "loss": 1.0308,
      "num_input_tokens_seen": 45625792,
      "step": 2836
    },
    {
      "epoch": 0.19872687313387094,
      "grad_norm": 4.383143901824951,
      "learning_rate": 8.014596847635727e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 45642176,
      "step": 2837
    },
    {
      "epoch": 0.1987969213796002,
      "grad_norm": 4.105413913726807,
      "learning_rate": 8.013897022767076e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 45657480,
      "step": 2838
    },
    {
      "epoch": 0.19886696962532946,
      "grad_norm": 5.339532852172852,
      "learning_rate": 8.013197197898425e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 45673168,
      "step": 2839
    },
    {
      "epoch": 0.1989370178710587,
      "grad_norm": 3.8160016536712646,
      "learning_rate": 8.012497373029774e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 45689088,
      "step": 2840
    },
    {
      "epoch": 0.19900706611678795,
      "grad_norm": 3.763986587524414,
      "learning_rate": 8.011797548161121e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 45705472,
      "step": 2841
    },
    {
      "epoch": 0.19907711436251718,
      "grad_norm": 4.034756183624268,
      "learning_rate": 8.01109772329247e-05,
      "loss": 1.1856,
      "num_input_tokens_seen": 45721168,
      "step": 2842
    },
    {
      "epoch": 0.19914716260824644,
      "grad_norm": 3.971479654312134,
      "learning_rate": 8.010397898423817e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 45737312,
      "step": 2843
    },
    {
      "epoch": 0.19921721085397567,
      "grad_norm": 4.118296146392822,
      "learning_rate": 8.009698073555166e-05,
      "loss": 0.9964,
      "num_input_tokens_seen": 45752792,
      "step": 2844
    },
    {
      "epoch": 0.19928725909970493,
      "grad_norm": 3.628143310546875,
      "learning_rate": 8.008998248686515e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 45769008,
      "step": 2845
    },
    {
      "epoch": 0.19935730734543416,
      "grad_norm": 3.9946494102478027,
      "learning_rate": 8.008298423817864e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 45785392,
      "step": 2846
    },
    {
      "epoch": 0.19942735559116342,
      "grad_norm": 3.7445459365844727,
      "learning_rate": 8.007598598949213e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 45801320,
      "step": 2847
    },
    {
      "epoch": 0.19949740383689266,
      "grad_norm": 3.745481491088867,
      "learning_rate": 8.006898774080562e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 45817504,
      "step": 2848
    },
    {
      "epoch": 0.19956745208262192,
      "grad_norm": 4.1305766105651855,
      "learning_rate": 8.006198949211909e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 45833888,
      "step": 2849
    },
    {
      "epoch": 0.19963750032835115,
      "grad_norm": 3.7843470573425293,
      "learning_rate": 8.005499124343258e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 45850272,
      "step": 2850
    },
    {
      "epoch": 0.1997075485740804,
      "grad_norm": 3.9884989261627197,
      "learning_rate": 8.004799299474606e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 45866656,
      "step": 2851
    },
    {
      "epoch": 0.19977759681980964,
      "grad_norm": 3.7280545234680176,
      "learning_rate": 8.004099474605956e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 45882776,
      "step": 2852
    },
    {
      "epoch": 0.1998476450655389,
      "grad_norm": 5.151428699493408,
      "learning_rate": 8.003399649737303e-05,
      "loss": 1.2988,
      "num_input_tokens_seen": 45898520,
      "step": 2853
    },
    {
      "epoch": 0.19991769331126813,
      "grad_norm": 6.738519191741943,
      "learning_rate": 8.002699824868652e-05,
      "loss": 1.1934,
      "num_input_tokens_seen": 45914904,
      "step": 2854
    },
    {
      "epoch": 0.1999877415569974,
      "grad_norm": 4.689775466918945,
      "learning_rate": 8.002000000000001e-05,
      "loss": 1.3534,
      "num_input_tokens_seen": 45931288,
      "step": 2855
    },
    {
      "epoch": 0.20005778980272662,
      "grad_norm": 4.047792911529541,
      "learning_rate": 8.001300175131348e-05,
      "loss": 1.2926,
      "num_input_tokens_seen": 45947672,
      "step": 2856
    },
    {
      "epoch": 0.20012783804845588,
      "grad_norm": 4.609661102294922,
      "learning_rate": 8.000600350262697e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 45964056,
      "step": 2857
    },
    {
      "epoch": 0.2001978862941851,
      "grad_norm": 4.188840389251709,
      "learning_rate": 7.999900525394046e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 45980152,
      "step": 2858
    },
    {
      "epoch": 0.20026793453991437,
      "grad_norm": 3.558335781097412,
      "learning_rate": 7.999200700525395e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 45996536,
      "step": 2859
    },
    {
      "epoch": 0.2003379827856436,
      "grad_norm": 10.145834922790527,
      "learning_rate": 7.998500875656743e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 46011616,
      "step": 2860
    },
    {
      "epoch": 0.20040803103137286,
      "grad_norm": 5.534536838531494,
      "learning_rate": 7.997801050788091e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 46027016,
      "step": 2861
    },
    {
      "epoch": 0.2004780792771021,
      "grad_norm": 4.258336544036865,
      "learning_rate": 7.99710122591944e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 46043400,
      "step": 2862
    },
    {
      "epoch": 0.20054812752283135,
      "grad_norm": 5.266301155090332,
      "learning_rate": 7.996401401050788e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 46059784,
      "step": 2863
    },
    {
      "epoch": 0.20061817576856059,
      "grad_norm": 4.502764701843262,
      "learning_rate": 7.995701576182137e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 46075584,
      "step": 2864
    },
    {
      "epoch": 0.20068822401428985,
      "grad_norm": 4.39752197265625,
      "learning_rate": 7.995001751313485e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 46091520,
      "step": 2865
    },
    {
      "epoch": 0.20075827226001908,
      "grad_norm": 3.9562480449676514,
      "learning_rate": 7.994301926444834e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 46107568,
      "step": 2866
    },
    {
      "epoch": 0.20082832050574834,
      "grad_norm": 4.466681957244873,
      "learning_rate": 7.993602101576183e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 46123952,
      "step": 2867
    },
    {
      "epoch": 0.20089836875147757,
      "grad_norm": 3.9317095279693604,
      "learning_rate": 7.992902276707531e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 46140336,
      "step": 2868
    },
    {
      "epoch": 0.20096841699720683,
      "grad_norm": 5.025266170501709,
      "learning_rate": 7.99220245183888e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 46155504,
      "step": 2869
    },
    {
      "epoch": 0.20103846524293606,
      "grad_norm": 3.82340931892395,
      "learning_rate": 7.991502626970227e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 46171888,
      "step": 2870
    },
    {
      "epoch": 0.20110851348866532,
      "grad_norm": 4.017914295196533,
      "learning_rate": 7.990802802101576e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 46187712,
      "step": 2871
    },
    {
      "epoch": 0.20117856173439455,
      "grad_norm": 4.053089618682861,
      "learning_rate": 7.990102977232926e-05,
      "loss": 0.9687,
      "num_input_tokens_seen": 46202912,
      "step": 2872
    },
    {
      "epoch": 0.2012486099801238,
      "grad_norm": 3.5664076805114746,
      "learning_rate": 7.989403152364274e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 46219296,
      "step": 2873
    },
    {
      "epoch": 0.20131865822585307,
      "grad_norm": 4.039318084716797,
      "learning_rate": 7.988703327495623e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 46235680,
      "step": 2874
    },
    {
      "epoch": 0.2013887064715823,
      "grad_norm": 3.8851678371429443,
      "learning_rate": 7.988003502626971e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 46251408,
      "step": 2875
    },
    {
      "epoch": 0.20145875471731156,
      "grad_norm": 3.581632137298584,
      "learning_rate": 7.987303677758319e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 46267696,
      "step": 2876
    },
    {
      "epoch": 0.2015288029630408,
      "grad_norm": 4.135960102081299,
      "learning_rate": 7.986603852889668e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 46284080,
      "step": 2877
    },
    {
      "epoch": 0.20159885120877005,
      "grad_norm": 3.649959087371826,
      "learning_rate": 7.985904028021017e-05,
      "loss": 0.8932,
      "num_input_tokens_seen": 46300456,
      "step": 2878
    },
    {
      "epoch": 0.20166889945449928,
      "grad_norm": 4.564159393310547,
      "learning_rate": 7.985204203152365e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 46315928,
      "step": 2879
    },
    {
      "epoch": 0.20173894770022854,
      "grad_norm": 3.806626796722412,
      "learning_rate": 7.984504378283713e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 46331520,
      "step": 2880
    },
    {
      "epoch": 0.20180899594595778,
      "grad_norm": 6.621458530426025,
      "learning_rate": 7.983804553415062e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 46347904,
      "step": 2881
    },
    {
      "epoch": 0.20187904419168703,
      "grad_norm": 4.554089546203613,
      "learning_rate": 7.983104728546411e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 46364288,
      "step": 2882
    },
    {
      "epoch": 0.20194909243741627,
      "grad_norm": 4.206694602966309,
      "learning_rate": 7.982404903677758e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 46380672,
      "step": 2883
    },
    {
      "epoch": 0.20201914068314553,
      "grad_norm": 6.333064079284668,
      "learning_rate": 7.981705078809107e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 46396384,
      "step": 2884
    },
    {
      "epoch": 0.20208918892887476,
      "grad_norm": 3.6293835639953613,
      "learning_rate": 7.981005253940456e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 46412712,
      "step": 2885
    },
    {
      "epoch": 0.20215923717460402,
      "grad_norm": 6.282841682434082,
      "learning_rate": 7.980305429071805e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 46429096,
      "step": 2886
    },
    {
      "epoch": 0.20222928542033325,
      "grad_norm": 3.661564350128174,
      "learning_rate": 7.979605604203152e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 46445480,
      "step": 2887
    },
    {
      "epoch": 0.2022993336660625,
      "grad_norm": 4.232359409332275,
      "learning_rate": 7.978905779334501e-05,
      "loss": 1.3196,
      "num_input_tokens_seen": 46461344,
      "step": 2888
    },
    {
      "epoch": 0.20236938191179174,
      "grad_norm": 3.9777348041534424,
      "learning_rate": 7.97820595446585e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 46477728,
      "step": 2889
    },
    {
      "epoch": 0.202439430157521,
      "grad_norm": 4.221210479736328,
      "learning_rate": 7.977506129597197e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 46493680,
      "step": 2890
    },
    {
      "epoch": 0.20250947840325023,
      "grad_norm": 4.210818767547607,
      "learning_rate": 7.976806304728546e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 46510064,
      "step": 2891
    },
    {
      "epoch": 0.2025795266489795,
      "grad_norm": 5.012551307678223,
      "learning_rate": 7.976106479859895e-05,
      "loss": 0.9933,
      "num_input_tokens_seen": 46526448,
      "step": 2892
    },
    {
      "epoch": 0.20264957489470872,
      "grad_norm": 3.4867520332336426,
      "learning_rate": 7.975406654991244e-05,
      "loss": 0.8495,
      "num_input_tokens_seen": 46542832,
      "step": 2893
    },
    {
      "epoch": 0.20271962314043798,
      "grad_norm": 4.74222993850708,
      "learning_rate": 7.974706830122593e-05,
      "loss": 1.1398,
      "num_input_tokens_seen": 46559048,
      "step": 2894
    },
    {
      "epoch": 0.20278967138616721,
      "grad_norm": 5.358060359954834,
      "learning_rate": 7.97400700525394e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 46575400,
      "step": 2895
    },
    {
      "epoch": 0.20285971963189647,
      "grad_norm": 4.2599053382873535,
      "learning_rate": 7.973307180385289e-05,
      "loss": 1.0021,
      "num_input_tokens_seen": 46591064,
      "step": 2896
    },
    {
      "epoch": 0.2029297678776257,
      "grad_norm": 5.993118762969971,
      "learning_rate": 7.972607355516637e-05,
      "loss": 1.2017,
      "num_input_tokens_seen": 46606504,
      "step": 2897
    },
    {
      "epoch": 0.20299981612335496,
      "grad_norm": 4.129568576812744,
      "learning_rate": 7.971907530647987e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 46622400,
      "step": 2898
    },
    {
      "epoch": 0.2030698643690842,
      "grad_norm": 3.8486111164093018,
      "learning_rate": 7.971207705779336e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 46638752,
      "step": 2899
    },
    {
      "epoch": 0.20313991261481346,
      "grad_norm": 4.262311935424805,
      "learning_rate": 7.970507880910683e-05,
      "loss": 1.1222,
      "num_input_tokens_seen": 46655136,
      "step": 2900
    },
    {
      "epoch": 0.2032099608605427,
      "grad_norm": 4.065335750579834,
      "learning_rate": 7.969808056042032e-05,
      "loss": 1.2965,
      "num_input_tokens_seen": 46671520,
      "step": 2901
    },
    {
      "epoch": 0.20328000910627195,
      "grad_norm": 3.8313064575195312,
      "learning_rate": 7.969108231173381e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 46687904,
      "step": 2902
    },
    {
      "epoch": 0.20335005735200118,
      "grad_norm": 3.711580276489258,
      "learning_rate": 7.968408406304729e-05,
      "loss": 1.1688,
      "num_input_tokens_seen": 46704088,
      "step": 2903
    },
    {
      "epoch": 0.20342010559773044,
      "grad_norm": 4.172581672668457,
      "learning_rate": 7.967708581436077e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 46720360,
      "step": 2904
    },
    {
      "epoch": 0.20349015384345967,
      "grad_norm": 4.7567267417907715,
      "learning_rate": 7.967008756567426e-05,
      "loss": 1.169,
      "num_input_tokens_seen": 46735560,
      "step": 2905
    },
    {
      "epoch": 0.20356020208918893,
      "grad_norm": 4.304897308349609,
      "learning_rate": 7.966308931698775e-05,
      "loss": 0.9359,
      "num_input_tokens_seen": 46751720,
      "step": 2906
    },
    {
      "epoch": 0.20363025033491816,
      "grad_norm": 4.0556864738464355,
      "learning_rate": 7.965609106830123e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 46767432,
      "step": 2907
    },
    {
      "epoch": 0.20370029858064742,
      "grad_norm": 3.7381911277770996,
      "learning_rate": 7.964909281961472e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 46783488,
      "step": 2908
    },
    {
      "epoch": 0.20377034682637668,
      "grad_norm": 4.363048553466797,
      "learning_rate": 7.96420945709282e-05,
      "loss": 0.9627,
      "num_input_tokens_seen": 46799016,
      "step": 2909
    },
    {
      "epoch": 0.2038403950721059,
      "grad_norm": 4.04617166519165,
      "learning_rate": 7.963509632224168e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 46815400,
      "step": 2910
    },
    {
      "epoch": 0.20391044331783517,
      "grad_norm": 3.8854830265045166,
      "learning_rate": 7.962809807355517e-05,
      "loss": 1.0525,
      "num_input_tokens_seen": 46831784,
      "step": 2911
    },
    {
      "epoch": 0.2039804915635644,
      "grad_norm": 4.197749614715576,
      "learning_rate": 7.962109982486866e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 46848168,
      "step": 2912
    },
    {
      "epoch": 0.20405053980929366,
      "grad_norm": 4.414098739624023,
      "learning_rate": 7.961410157618214e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 46864552,
      "step": 2913
    },
    {
      "epoch": 0.2041205880550229,
      "grad_norm": 3.7771573066711426,
      "learning_rate": 7.960710332749562e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 46880936,
      "step": 2914
    },
    {
      "epoch": 0.20419063630075215,
      "grad_norm": 4.179026126861572,
      "learning_rate": 7.960010507880911e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 46897192,
      "step": 2915
    },
    {
      "epoch": 0.20426068454648139,
      "grad_norm": 4.017509460449219,
      "learning_rate": 7.95931068301226e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 46913576,
      "step": 2916
    },
    {
      "epoch": 0.20433073279221065,
      "grad_norm": 5.863056182861328,
      "learning_rate": 7.958610858143607e-05,
      "loss": 1.235,
      "num_input_tokens_seen": 46929960,
      "step": 2917
    },
    {
      "epoch": 0.20440078103793988,
      "grad_norm": 5.267307281494141,
      "learning_rate": 7.957911033274956e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 46946344,
      "step": 2918
    },
    {
      "epoch": 0.20447082928366914,
      "grad_norm": 5.20788049697876,
      "learning_rate": 7.957211208406306e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 46961712,
      "step": 2919
    },
    {
      "epoch": 0.20454087752939837,
      "grad_norm": 4.609791278839111,
      "learning_rate": 7.956511383537654e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 46977752,
      "step": 2920
    },
    {
      "epoch": 0.20461092577512763,
      "grad_norm": 3.9752824306488037,
      "learning_rate": 7.955811558669003e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 46994136,
      "step": 2921
    },
    {
      "epoch": 0.20468097402085686,
      "grad_norm": 3.8456339836120605,
      "learning_rate": 7.95511173380035e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 47010520,
      "step": 2922
    },
    {
      "epoch": 0.20475102226658612,
      "grad_norm": 4.087759494781494,
      "learning_rate": 7.954411908931699e-05,
      "loss": 1.0535,
      "num_input_tokens_seen": 47026904,
      "step": 2923
    },
    {
      "epoch": 0.20482107051231535,
      "grad_norm": 3.9754104614257812,
      "learning_rate": 7.953712084063048e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 47043288,
      "step": 2924
    },
    {
      "epoch": 0.2048911187580446,
      "grad_norm": 3.61798357963562,
      "learning_rate": 7.953012259194397e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 47059672,
      "step": 2925
    },
    {
      "epoch": 0.20496116700377384,
      "grad_norm": 4.015439510345459,
      "learning_rate": 7.952312434325746e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 47074232,
      "step": 2926
    },
    {
      "epoch": 0.2050312152495031,
      "grad_norm": 5.869642734527588,
      "learning_rate": 7.951612609457093e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 47090616,
      "step": 2927
    },
    {
      "epoch": 0.20510126349523233,
      "grad_norm": 4.0500922203063965,
      "learning_rate": 7.950912784588442e-05,
      "loss": 1.1142,
      "num_input_tokens_seen": 47106656,
      "step": 2928
    },
    {
      "epoch": 0.2051713117409616,
      "grad_norm": 5.468737602233887,
      "learning_rate": 7.950212959719791e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 47122648,
      "step": 2929
    },
    {
      "epoch": 0.20524135998669082,
      "grad_norm": 3.842905282974243,
      "learning_rate": 7.949513134851138e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 47139032,
      "step": 2930
    },
    {
      "epoch": 0.20531140823242008,
      "grad_norm": 4.24273681640625,
      "learning_rate": 7.948813309982487e-05,
      "loss": 1.0533,
      "num_input_tokens_seen": 47154344,
      "step": 2931
    },
    {
      "epoch": 0.20538145647814932,
      "grad_norm": 3.977433443069458,
      "learning_rate": 7.948113485113836e-05,
      "loss": 0.9184,
      "num_input_tokens_seen": 47170728,
      "step": 2932
    },
    {
      "epoch": 0.20545150472387858,
      "grad_norm": 3.8441646099090576,
      "learning_rate": 7.947413660245185e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 47187112,
      "step": 2933
    },
    {
      "epoch": 0.2055215529696078,
      "grad_norm": 3.3789381980895996,
      "learning_rate": 7.946713835376532e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 47203400,
      "step": 2934
    },
    {
      "epoch": 0.20559160121533707,
      "grad_norm": 3.817631483078003,
      "learning_rate": 7.946014010507881e-05,
      "loss": 1.198,
      "num_input_tokens_seen": 47219784,
      "step": 2935
    },
    {
      "epoch": 0.2056616494610663,
      "grad_norm": 3.788300037384033,
      "learning_rate": 7.94531418563923e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 47236168,
      "step": 2936
    },
    {
      "epoch": 0.20573169770679556,
      "grad_norm": 3.852132558822632,
      "learning_rate": 7.944614360770578e-05,
      "loss": 1.1259,
      "num_input_tokens_seen": 47252288,
      "step": 2937
    },
    {
      "epoch": 0.2058017459525248,
      "grad_norm": 3.8631093502044678,
      "learning_rate": 7.943914535901926e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 47267000,
      "step": 2938
    },
    {
      "epoch": 0.20587179419825405,
      "grad_norm": 3.72165846824646,
      "learning_rate": 7.943214711033275e-05,
      "loss": 0.7975,
      "num_input_tokens_seen": 47282832,
      "step": 2939
    },
    {
      "epoch": 0.20594184244398328,
      "grad_norm": 4.04188871383667,
      "learning_rate": 7.942514886164624e-05,
      "loss": 1.0953,
      "num_input_tokens_seen": 47298320,
      "step": 2940
    },
    {
      "epoch": 0.20601189068971254,
      "grad_norm": 3.5907206535339355,
      "learning_rate": 7.941815061295972e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 47314704,
      "step": 2941
    },
    {
      "epoch": 0.20608193893544177,
      "grad_norm": 5.023667335510254,
      "learning_rate": 7.94111523642732e-05,
      "loss": 1.2083,
      "num_input_tokens_seen": 47331088,
      "step": 2942
    },
    {
      "epoch": 0.20615198718117103,
      "grad_norm": 3.8885724544525146,
      "learning_rate": 7.94041541155867e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 47347424,
      "step": 2943
    },
    {
      "epoch": 0.2062220354269003,
      "grad_norm": 4.289493560791016,
      "learning_rate": 7.939715586690017e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 47363808,
      "step": 2944
    },
    {
      "epoch": 0.20629208367262952,
      "grad_norm": 4.976572513580322,
      "learning_rate": 7.939015761821367e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 47379152,
      "step": 2945
    },
    {
      "epoch": 0.20636213191835878,
      "grad_norm": 4.0893425941467285,
      "learning_rate": 7.938315936952716e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 47395536,
      "step": 2946
    },
    {
      "epoch": 0.206432180164088,
      "grad_norm": 3.799873113632202,
      "learning_rate": 7.937616112084063e-05,
      "loss": 1.1433,
      "num_input_tokens_seen": 47410968,
      "step": 2947
    },
    {
      "epoch": 0.20650222840981727,
      "grad_norm": 4.688945293426514,
      "learning_rate": 7.936916287215412e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 47427352,
      "step": 2948
    },
    {
      "epoch": 0.2065722766555465,
      "grad_norm": 3.6503846645355225,
      "learning_rate": 7.93621646234676e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 47443736,
      "step": 2949
    },
    {
      "epoch": 0.20664232490127576,
      "grad_norm": 4.2314324378967285,
      "learning_rate": 7.935516637478109e-05,
      "loss": 1.2795,
      "num_input_tokens_seen": 47460120,
      "step": 2950
    },
    {
      "epoch": 0.206712373147005,
      "grad_norm": 5.159674644470215,
      "learning_rate": 7.934816812609458e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 47476256,
      "step": 2951
    },
    {
      "epoch": 0.20678242139273426,
      "grad_norm": 3.798804759979248,
      "learning_rate": 7.934116987740806e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 47492208,
      "step": 2952
    },
    {
      "epoch": 0.2068524696384635,
      "grad_norm": 4.233975887298584,
      "learning_rate": 7.933417162872155e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 47507728,
      "step": 2953
    },
    {
      "epoch": 0.20692251788419275,
      "grad_norm": 3.38350772857666,
      "learning_rate": 7.932717338003503e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 47523992,
      "step": 2954
    },
    {
      "epoch": 0.20699256612992198,
      "grad_norm": 3.94380521774292,
      "learning_rate": 7.932017513134852e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 47540376,
      "step": 2955
    },
    {
      "epoch": 0.20706261437565124,
      "grad_norm": 3.887354612350464,
      "learning_rate": 7.9313176882662e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 47555336,
      "step": 2956
    },
    {
      "epoch": 0.20713266262138047,
      "grad_norm": 4.271602153778076,
      "learning_rate": 7.930617863397548e-05,
      "loss": 1.3089,
      "num_input_tokens_seen": 47570520,
      "step": 2957
    },
    {
      "epoch": 0.20720271086710973,
      "grad_norm": 4.119933605194092,
      "learning_rate": 7.929918038528897e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 47586904,
      "step": 2958
    },
    {
      "epoch": 0.20727275911283896,
      "grad_norm": 6.137136936187744,
      "learning_rate": 7.929218213660246e-05,
      "loss": 0.7847,
      "num_input_tokens_seen": 47602424,
      "step": 2959
    },
    {
      "epoch": 0.20734280735856822,
      "grad_norm": 3.5264923572540283,
      "learning_rate": 7.928518388791595e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 47618808,
      "step": 2960
    },
    {
      "epoch": 0.20741285560429745,
      "grad_norm": 4.183988094329834,
      "learning_rate": 7.927818563922942e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 47634576,
      "step": 2961
    },
    {
      "epoch": 0.2074829038500267,
      "grad_norm": 3.486311197280884,
      "learning_rate": 7.927118739054291e-05,
      "loss": 0.8559,
      "num_input_tokens_seen": 47649920,
      "step": 2962
    },
    {
      "epoch": 0.20755295209575594,
      "grad_norm": 4.561336994171143,
      "learning_rate": 7.92641891418564e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 47666304,
      "step": 2963
    },
    {
      "epoch": 0.2076230003414852,
      "grad_norm": 4.002289295196533,
      "learning_rate": 7.925719089316987e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 47682688,
      "step": 2964
    },
    {
      "epoch": 0.20769304858721443,
      "grad_norm": 3.694175958633423,
      "learning_rate": 7.925019264448338e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 47699072,
      "step": 2965
    },
    {
      "epoch": 0.2077630968329437,
      "grad_norm": 3.7827298641204834,
      "learning_rate": 7.924319439579685e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 47714720,
      "step": 2966
    },
    {
      "epoch": 0.20783314507867293,
      "grad_norm": 3.8371527194976807,
      "learning_rate": 7.923619614711034e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 47730904,
      "step": 2967
    },
    {
      "epoch": 0.20790319332440219,
      "grad_norm": 4.20089054107666,
      "learning_rate": 7.922919789842381e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 47747288,
      "step": 2968
    },
    {
      "epoch": 0.20797324157013142,
      "grad_norm": 3.978065252304077,
      "learning_rate": 7.92221996497373e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 47763672,
      "step": 2969
    },
    {
      "epoch": 0.20804328981586068,
      "grad_norm": 4.882012844085693,
      "learning_rate": 7.921520140105079e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 47778888,
      "step": 2970
    },
    {
      "epoch": 0.2081133380615899,
      "grad_norm": 4.202088356018066,
      "learning_rate": 7.920820315236428e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 47795272,
      "step": 2971
    },
    {
      "epoch": 0.20818338630731917,
      "grad_norm": 3.623647928237915,
      "learning_rate": 7.920120490367777e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 47811656,
      "step": 2972
    },
    {
      "epoch": 0.2082534345530484,
      "grad_norm": 4.158148765563965,
      "learning_rate": 7.919420665499126e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 47827520,
      "step": 2973
    },
    {
      "epoch": 0.20832348279877766,
      "grad_norm": 4.016353130340576,
      "learning_rate": 7.918720840630473e-05,
      "loss": 1.0176,
      "num_input_tokens_seen": 47843904,
      "step": 2974
    },
    {
      "epoch": 0.2083935310445069,
      "grad_norm": 5.862729072570801,
      "learning_rate": 7.918021015761822e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 47860288,
      "step": 2975
    },
    {
      "epoch": 0.20846357929023615,
      "grad_norm": 4.194519519805908,
      "learning_rate": 7.91732119089317e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 47876536,
      "step": 2976
    },
    {
      "epoch": 0.20853362753596538,
      "grad_norm": 3.925144672393799,
      "learning_rate": 7.916621366024518e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 47892216,
      "step": 2977
    },
    {
      "epoch": 0.20860367578169464,
      "grad_norm": 4.005881309509277,
      "learning_rate": 7.915921541155867e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 47907840,
      "step": 2978
    },
    {
      "epoch": 0.2086737240274239,
      "grad_norm": 3.6061627864837646,
      "learning_rate": 7.915221716287216e-05,
      "loss": 0.8235,
      "num_input_tokens_seen": 47923832,
      "step": 2979
    },
    {
      "epoch": 0.20874377227315313,
      "grad_norm": 4.407896041870117,
      "learning_rate": 7.914521891418565e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 47940216,
      "step": 2980
    },
    {
      "epoch": 0.2088138205188824,
      "grad_norm": 4.089472770690918,
      "learning_rate": 7.913822066549912e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 47956600,
      "step": 2981
    },
    {
      "epoch": 0.20888386876461162,
      "grad_norm": 4.384250640869141,
      "learning_rate": 7.913122241681261e-05,
      "loss": 1.1681,
      "num_input_tokens_seen": 47972984,
      "step": 2982
    },
    {
      "epoch": 0.20895391701034088,
      "grad_norm": 3.881756544113159,
      "learning_rate": 7.91242241681261e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 47989368,
      "step": 2983
    },
    {
      "epoch": 0.20902396525607012,
      "grad_norm": 3.9435884952545166,
      "learning_rate": 7.911722591943958e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 48005608,
      "step": 2984
    },
    {
      "epoch": 0.20909401350179938,
      "grad_norm": 4.1196794509887695,
      "learning_rate": 7.911022767075308e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 48021992,
      "step": 2985
    },
    {
      "epoch": 0.2091640617475286,
      "grad_norm": 4.482571125030518,
      "learning_rate": 7.910322942206655e-05,
      "loss": 1.0663,
      "num_input_tokens_seen": 48037816,
      "step": 2986
    },
    {
      "epoch": 0.20923410999325787,
      "grad_norm": 5.359109401702881,
      "learning_rate": 7.909623117338004e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 48054200,
      "step": 2987
    },
    {
      "epoch": 0.2093041582389871,
      "grad_norm": 5.712708950042725,
      "learning_rate": 7.908923292469352e-05,
      "loss": 1.1454,
      "num_input_tokens_seen": 48070008,
      "step": 2988
    },
    {
      "epoch": 0.20937420648471636,
      "grad_norm": 3.980526924133301,
      "learning_rate": 7.9082234676007e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 48084864,
      "step": 2989
    },
    {
      "epoch": 0.2094442547304456,
      "grad_norm": 4.963679790496826,
      "learning_rate": 7.90752364273205e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 48101248,
      "step": 2990
    },
    {
      "epoch": 0.20951430297617485,
      "grad_norm": 6.20939302444458,
      "learning_rate": 7.906823817863398e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 48114984,
      "step": 2991
    },
    {
      "epoch": 0.20958435122190408,
      "grad_norm": 13.218465805053711,
      "learning_rate": 7.906123992994747e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 48129704,
      "step": 2992
    },
    {
      "epoch": 0.20965439946763334,
      "grad_norm": 6.285522937774658,
      "learning_rate": 7.905424168126095e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 48144280,
      "step": 2993
    },
    {
      "epoch": 0.20972444771336257,
      "grad_norm": 5.113750457763672,
      "learning_rate": 7.904724343257444e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 48160664,
      "step": 2994
    },
    {
      "epoch": 0.20979449595909183,
      "grad_norm": 3.5571322441101074,
      "learning_rate": 7.904024518388791e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 48177048,
      "step": 2995
    },
    {
      "epoch": 0.20986454420482106,
      "grad_norm": 4.965229511260986,
      "learning_rate": 7.90332469352014e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 48193400,
      "step": 2996
    },
    {
      "epoch": 0.20993459245055032,
      "grad_norm": 4.466450214385986,
      "learning_rate": 7.902624868651489e-05,
      "loss": 1.2786,
      "num_input_tokens_seen": 48209784,
      "step": 2997
    },
    {
      "epoch": 0.21000464069627955,
      "grad_norm": 3.556642770767212,
      "learning_rate": 7.901925043782838e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 48226096,
      "step": 2998
    },
    {
      "epoch": 0.2100746889420088,
      "grad_norm": 5.175073146820068,
      "learning_rate": 7.901225218914187e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 48242384,
      "step": 2999
    },
    {
      "epoch": 0.21014473718773805,
      "grad_norm": 4.901797771453857,
      "learning_rate": 7.900525394045535e-05,
      "loss": 0.9413,
      "num_input_tokens_seen": 48257944,
      "step": 3000
    },
    {
      "epoch": 0.21014473718773805,
      "eval_loss": 1.137844204902649,
      "eval_runtime": 0.2151,
      "eval_samples_per_second": 4.649,
      "eval_steps_per_second": 4.649,
      "num_input_tokens_seen": 48257944,
      "step": 3000
    },
    {
      "epoch": 0.2102147854334673,
      "grad_norm": 3.8474860191345215,
      "learning_rate": 7.899825569176883e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 48273144,
      "step": 3001
    },
    {
      "epoch": 0.21028483367919654,
      "grad_norm": 4.4164347648620605,
      "learning_rate": 7.899125744308232e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 48288896,
      "step": 3002
    },
    {
      "epoch": 0.2103548819249258,
      "grad_norm": 4.560143947601318,
      "learning_rate": 7.898425919439579e-05,
      "loss": 1.1129,
      "num_input_tokens_seen": 48305168,
      "step": 3003
    },
    {
      "epoch": 0.21042493017065503,
      "grad_norm": 4.310809135437012,
      "learning_rate": 7.897726094570928e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 48320936,
      "step": 3004
    },
    {
      "epoch": 0.2104949784163843,
      "grad_norm": 5.8606367111206055,
      "learning_rate": 7.897026269702277e-05,
      "loss": 0.7859,
      "num_input_tokens_seen": 48334752,
      "step": 3005
    },
    {
      "epoch": 0.21056502666211352,
      "grad_norm": 4.533644676208496,
      "learning_rate": 7.896326444833626e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 48351136,
      "step": 3006
    },
    {
      "epoch": 0.21063507490784278,
      "grad_norm": 3.955151081085205,
      "learning_rate": 7.895626619964975e-05,
      "loss": 1.3093,
      "num_input_tokens_seen": 48367520,
      "step": 3007
    },
    {
      "epoch": 0.210705123153572,
      "grad_norm": 4.857527732849121,
      "learning_rate": 7.894926795096322e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 48383584,
      "step": 3008
    },
    {
      "epoch": 0.21077517139930127,
      "grad_norm": 4.2091593742370605,
      "learning_rate": 7.894226970227671e-05,
      "loss": 0.9278,
      "num_input_tokens_seen": 48399968,
      "step": 3009
    },
    {
      "epoch": 0.2108452196450305,
      "grad_norm": 4.02255916595459,
      "learning_rate": 7.89352714535902e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 48416016,
      "step": 3010
    },
    {
      "epoch": 0.21091526789075976,
      "grad_norm": 4.021467208862305,
      "learning_rate": 7.892827320490369e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 48432400,
      "step": 3011
    },
    {
      "epoch": 0.21098531613648902,
      "grad_norm": 4.211849212646484,
      "learning_rate": 7.892127495621716e-05,
      "loss": 1.1698,
      "num_input_tokens_seen": 48448784,
      "step": 3012
    },
    {
      "epoch": 0.21105536438221825,
      "grad_norm": 3.890512704849243,
      "learning_rate": 7.891427670753065e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 48465168,
      "step": 3013
    },
    {
      "epoch": 0.2111254126279475,
      "grad_norm": 3.9605376720428467,
      "learning_rate": 7.890727845884414e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 48481024,
      "step": 3014
    },
    {
      "epoch": 0.21119546087367674,
      "grad_norm": 3.6985483169555664,
      "learning_rate": 7.890028021015761e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 48497408,
      "step": 3015
    },
    {
      "epoch": 0.211265509119406,
      "grad_norm": 4.245354652404785,
      "learning_rate": 7.88932819614711e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 48513640,
      "step": 3016
    },
    {
      "epoch": 0.21133555736513523,
      "grad_norm": 4.163609027862549,
      "learning_rate": 7.888628371278459e-05,
      "loss": 1.2399,
      "num_input_tokens_seen": 48529704,
      "step": 3017
    },
    {
      "epoch": 0.2114056056108645,
      "grad_norm": 4.139742374420166,
      "learning_rate": 7.887928546409808e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 48545808,
      "step": 3018
    },
    {
      "epoch": 0.21147565385659373,
      "grad_norm": 4.119020938873291,
      "learning_rate": 7.887228721541157e-05,
      "loss": 1.233,
      "num_input_tokens_seen": 48561584,
      "step": 3019
    },
    {
      "epoch": 0.21154570210232299,
      "grad_norm": 3.467578172683716,
      "learning_rate": 7.886528896672504e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 48577912,
      "step": 3020
    },
    {
      "epoch": 0.21161575034805222,
      "grad_norm": 4.891791820526123,
      "learning_rate": 7.885829071803853e-05,
      "loss": 1.0507,
      "num_input_tokens_seen": 48591792,
      "step": 3021
    },
    {
      "epoch": 0.21168579859378148,
      "grad_norm": 3.8184545040130615,
      "learning_rate": 7.885129246935201e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 48606656,
      "step": 3022
    },
    {
      "epoch": 0.2117558468395107,
      "grad_norm": 3.909607410430908,
      "learning_rate": 7.88442942206655e-05,
      "loss": 1.2735,
      "num_input_tokens_seen": 48622608,
      "step": 3023
    },
    {
      "epoch": 0.21182589508523997,
      "grad_norm": 3.780740737915039,
      "learning_rate": 7.883729597197899e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 48638992,
      "step": 3024
    },
    {
      "epoch": 0.2118959433309692,
      "grad_norm": 3.95491099357605,
      "learning_rate": 7.883029772329247e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 48654344,
      "step": 3025
    },
    {
      "epoch": 0.21196599157669846,
      "grad_norm": 3.724346876144409,
      "learning_rate": 7.882329947460596e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 48670728,
      "step": 3026
    },
    {
      "epoch": 0.2120360398224277,
      "grad_norm": 4.314544200897217,
      "learning_rate": 7.881630122591945e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 48685424,
      "step": 3027
    },
    {
      "epoch": 0.21210608806815695,
      "grad_norm": 3.9340150356292725,
      "learning_rate": 7.880930297723293e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 48700416,
      "step": 3028
    },
    {
      "epoch": 0.21217613631388618,
      "grad_norm": 4.978084087371826,
      "learning_rate": 7.880230472854641e-05,
      "loss": 1.2418,
      "num_input_tokens_seen": 48716800,
      "step": 3029
    },
    {
      "epoch": 0.21224618455961544,
      "grad_norm": 3.7038094997406006,
      "learning_rate": 7.879530647985989e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 48732616,
      "step": 3030
    },
    {
      "epoch": 0.21231623280534467,
      "grad_norm": 5.743021011352539,
      "learning_rate": 7.878830823117339e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 48748656,
      "step": 3031
    },
    {
      "epoch": 0.21238628105107393,
      "grad_norm": 5.655540943145752,
      "learning_rate": 7.878130998248687e-05,
      "loss": 1.3541,
      "num_input_tokens_seen": 48765040,
      "step": 3032
    },
    {
      "epoch": 0.21245632929680316,
      "grad_norm": 4.291803359985352,
      "learning_rate": 7.877431173380036e-05,
      "loss": 1.1966,
      "num_input_tokens_seen": 48781424,
      "step": 3033
    },
    {
      "epoch": 0.21252637754253242,
      "grad_norm": 5.103096961975098,
      "learning_rate": 7.876731348511384e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 48797808,
      "step": 3034
    },
    {
      "epoch": 0.21259642578826166,
      "grad_norm": 5.048161029815674,
      "learning_rate": 7.876031523642732e-05,
      "loss": 0.9595,
      "num_input_tokens_seen": 48814192,
      "step": 3035
    },
    {
      "epoch": 0.21266647403399092,
      "grad_norm": 4.086791038513184,
      "learning_rate": 7.875331698774081e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 48829816,
      "step": 3036
    },
    {
      "epoch": 0.21273652227972015,
      "grad_norm": 3.8422605991363525,
      "learning_rate": 7.87463187390543e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 48846200,
      "step": 3037
    },
    {
      "epoch": 0.2128065705254494,
      "grad_norm": 3.7120776176452637,
      "learning_rate": 7.873932049036778e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 48862584,
      "step": 3038
    },
    {
      "epoch": 0.21287661877117864,
      "grad_norm": 5.051353454589844,
      "learning_rate": 7.873232224168126e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 48878368,
      "step": 3039
    },
    {
      "epoch": 0.2129466670169079,
      "grad_norm": 3.9874653816223145,
      "learning_rate": 7.872532399299475e-05,
      "loss": 1.256,
      "num_input_tokens_seen": 48894696,
      "step": 3040
    },
    {
      "epoch": 0.21301671526263713,
      "grad_norm": 4.455258369445801,
      "learning_rate": 7.871832574430824e-05,
      "loss": 1.2226,
      "num_input_tokens_seen": 48911080,
      "step": 3041
    },
    {
      "epoch": 0.2130867635083664,
      "grad_norm": 5.521103382110596,
      "learning_rate": 7.871132749562171e-05,
      "loss": 1.2116,
      "num_input_tokens_seen": 48927464,
      "step": 3042
    },
    {
      "epoch": 0.21315681175409562,
      "grad_norm": 3.80818510055542,
      "learning_rate": 7.87043292469352e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 48943848,
      "step": 3043
    },
    {
      "epoch": 0.21322685999982488,
      "grad_norm": 4.319914817810059,
      "learning_rate": 7.869733099824869e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 48960232,
      "step": 3044
    },
    {
      "epoch": 0.2132969082455541,
      "grad_norm": 4.196371078491211,
      "learning_rate": 7.869033274956218e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 48976616,
      "step": 3045
    },
    {
      "epoch": 0.21336695649128337,
      "grad_norm": 3.988114595413208,
      "learning_rate": 7.868333450087567e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 48992248,
      "step": 3046
    },
    {
      "epoch": 0.21343700473701263,
      "grad_norm": 3.887589454650879,
      "learning_rate": 7.867633625218914e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 49008632,
      "step": 3047
    },
    {
      "epoch": 0.21350705298274186,
      "grad_norm": 3.7942206859588623,
      "learning_rate": 7.866933800350263e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 49024560,
      "step": 3048
    },
    {
      "epoch": 0.21357710122847112,
      "grad_norm": 4.464767932891846,
      "learning_rate": 7.86623397548161e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 49040200,
      "step": 3049
    },
    {
      "epoch": 0.21364714947420035,
      "grad_norm": 4.411591529846191,
      "learning_rate": 7.86553415061296e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 49056328,
      "step": 3050
    },
    {
      "epoch": 0.2137171977199296,
      "grad_norm": 4.895592212677002,
      "learning_rate": 7.86483432574431e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 49072696,
      "step": 3051
    },
    {
      "epoch": 0.21378724596565885,
      "grad_norm": 4.46630859375,
      "learning_rate": 7.864134500875657e-05,
      "loss": 0.837,
      "num_input_tokens_seen": 49087256,
      "step": 3052
    },
    {
      "epoch": 0.2138572942113881,
      "grad_norm": 4.975766658782959,
      "learning_rate": 7.863434676007006e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 49103640,
      "step": 3053
    },
    {
      "epoch": 0.21392734245711734,
      "grad_norm": 4.441544532775879,
      "learning_rate": 7.862734851138355e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 49119032,
      "step": 3054
    },
    {
      "epoch": 0.2139973907028466,
      "grad_norm": 3.797757148742676,
      "learning_rate": 7.862035026269702e-05,
      "loss": 0.8701,
      "num_input_tokens_seen": 49134960,
      "step": 3055
    },
    {
      "epoch": 0.21406743894857583,
      "grad_norm": 4.021834373474121,
      "learning_rate": 7.861335201401051e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 49151344,
      "step": 3056
    },
    {
      "epoch": 0.2141374871943051,
      "grad_norm": 3.772587537765503,
      "learning_rate": 7.8606353765324e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 49167424,
      "step": 3057
    },
    {
      "epoch": 0.21420753544003432,
      "grad_norm": 5.356356143951416,
      "learning_rate": 7.859935551663749e-05,
      "loss": 1.027,
      "num_input_tokens_seen": 49183504,
      "step": 3058
    },
    {
      "epoch": 0.21427758368576358,
      "grad_norm": 4.314568042755127,
      "learning_rate": 7.859235726795096e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 49199320,
      "step": 3059
    },
    {
      "epoch": 0.2143476319314928,
      "grad_norm": 3.777794122695923,
      "learning_rate": 7.858535901926445e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 49215032,
      "step": 3060
    },
    {
      "epoch": 0.21441768017722207,
      "grad_norm": 3.788496732711792,
      "learning_rate": 7.857836077057794e-05,
      "loss": 0.9121,
      "num_input_tokens_seen": 49230248,
      "step": 3061
    },
    {
      "epoch": 0.2144877284229513,
      "grad_norm": 3.776698589324951,
      "learning_rate": 7.857136252189142e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 49246264,
      "step": 3062
    },
    {
      "epoch": 0.21455777666868056,
      "grad_norm": 3.8229172229766846,
      "learning_rate": 7.85643642732049e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 49262648,
      "step": 3063
    },
    {
      "epoch": 0.2146278249144098,
      "grad_norm": 3.7620902061462402,
      "learning_rate": 7.85573660245184e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 49278640,
      "step": 3064
    },
    {
      "epoch": 0.21469787316013905,
      "grad_norm": 3.953148126602173,
      "learning_rate": 7.855036777583188e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 49295024,
      "step": 3065
    },
    {
      "epoch": 0.21476792140586828,
      "grad_norm": 4.1923441886901855,
      "learning_rate": 7.854336952714536e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 49311408,
      "step": 3066
    },
    {
      "epoch": 0.21483796965159754,
      "grad_norm": 4.922461986541748,
      "learning_rate": 7.853637127845885e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 49327120,
      "step": 3067
    },
    {
      "epoch": 0.21490801789732678,
      "grad_norm": 3.7414777278900146,
      "learning_rate": 7.852937302977233e-05,
      "loss": 0.8628,
      "num_input_tokens_seen": 49343504,
      "step": 3068
    },
    {
      "epoch": 0.21497806614305603,
      "grad_norm": 6.1490912437438965,
      "learning_rate": 7.852237478108581e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 49359336,
      "step": 3069
    },
    {
      "epoch": 0.21504811438878527,
      "grad_norm": 4.232786178588867,
      "learning_rate": 7.85153765323993e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 49374888,
      "step": 3070
    },
    {
      "epoch": 0.21511816263451453,
      "grad_norm": 4.170281887054443,
      "learning_rate": 7.85083782837128e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 49391272,
      "step": 3071
    },
    {
      "epoch": 0.21518821088024376,
      "grad_norm": 4.096348285675049,
      "learning_rate": 7.850138003502627e-05,
      "loss": 1.1574,
      "num_input_tokens_seen": 49407656,
      "step": 3072
    },
    {
      "epoch": 0.21525825912597302,
      "grad_norm": 4.523014068603516,
      "learning_rate": 7.849438178633976e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 49424040,
      "step": 3073
    },
    {
      "epoch": 0.21532830737170225,
      "grad_norm": 5.029306888580322,
      "learning_rate": 7.848738353765324e-05,
      "loss": 1.2744,
      "num_input_tokens_seen": 49440208,
      "step": 3074
    },
    {
      "epoch": 0.2153983556174315,
      "grad_norm": 3.5349771976470947,
      "learning_rate": 7.848038528896673e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 49456520,
      "step": 3075
    },
    {
      "epoch": 0.21546840386316074,
      "grad_norm": 3.544787645339966,
      "learning_rate": 7.84733870402802e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 49472904,
      "step": 3076
    },
    {
      "epoch": 0.21553845210889,
      "grad_norm": 4.602756500244141,
      "learning_rate": 7.84663887915937e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 49489264,
      "step": 3077
    },
    {
      "epoch": 0.21560850035461923,
      "grad_norm": 6.479659080505371,
      "learning_rate": 7.845939054290719e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 49505232,
      "step": 3078
    },
    {
      "epoch": 0.2156785486003485,
      "grad_norm": 4.584348201751709,
      "learning_rate": 7.845239229422067e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 49521616,
      "step": 3079
    },
    {
      "epoch": 0.21574859684607772,
      "grad_norm": 4.339470386505127,
      "learning_rate": 7.844539404553416e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 49537376,
      "step": 3080
    },
    {
      "epoch": 0.21581864509180698,
      "grad_norm": 4.098686218261719,
      "learning_rate": 7.843839579684765e-05,
      "loss": 0.9376,
      "num_input_tokens_seen": 49552256,
      "step": 3081
    },
    {
      "epoch": 0.21588869333753624,
      "grad_norm": 4.619485855102539,
      "learning_rate": 7.843139754816112e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 49568640,
      "step": 3082
    },
    {
      "epoch": 0.21595874158326547,
      "grad_norm": 4.018712997436523,
      "learning_rate": 7.842439929947461e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 49584816,
      "step": 3083
    },
    {
      "epoch": 0.21602878982899473,
      "grad_norm": 5.898901462554932,
      "learning_rate": 7.84174010507881e-05,
      "loss": 1.1956,
      "num_input_tokens_seen": 49600872,
      "step": 3084
    },
    {
      "epoch": 0.21609883807472396,
      "grad_norm": 4.794529438018799,
      "learning_rate": 7.841040280210159e-05,
      "loss": 1.0035,
      "num_input_tokens_seen": 49616840,
      "step": 3085
    },
    {
      "epoch": 0.21616888632045322,
      "grad_norm": 4.934964656829834,
      "learning_rate": 7.840340455341506e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 49633224,
      "step": 3086
    },
    {
      "epoch": 0.21623893456618246,
      "grad_norm": 3.6171560287475586,
      "learning_rate": 7.839640630472855e-05,
      "loss": 1.1165,
      "num_input_tokens_seen": 49649056,
      "step": 3087
    },
    {
      "epoch": 0.21630898281191172,
      "grad_norm": 4.032123565673828,
      "learning_rate": 7.838940805604204e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 49665440,
      "step": 3088
    },
    {
      "epoch": 0.21637903105764095,
      "grad_norm": 3.4669382572174072,
      "learning_rate": 7.838240980735551e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 49681824,
      "step": 3089
    },
    {
      "epoch": 0.2164490793033702,
      "grad_norm": 3.6899688243865967,
      "learning_rate": 7.8375411558669e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 49698208,
      "step": 3090
    },
    {
      "epoch": 0.21651912754909944,
      "grad_norm": 4.231171131134033,
      "learning_rate": 7.83684133099825e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 49713664,
      "step": 3091
    },
    {
      "epoch": 0.2165891757948287,
      "grad_norm": 4.792253017425537,
      "learning_rate": 7.836141506129598e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 49730048,
      "step": 3092
    },
    {
      "epoch": 0.21665922404055793,
      "grad_norm": 5.7171478271484375,
      "learning_rate": 7.835441681260945e-05,
      "loss": 1.189,
      "num_input_tokens_seen": 49746432,
      "step": 3093
    },
    {
      "epoch": 0.2167292722862872,
      "grad_norm": 4.393872261047363,
      "learning_rate": 7.834741856392294e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 49762816,
      "step": 3094
    },
    {
      "epoch": 0.21679932053201642,
      "grad_norm": 6.388276100158691,
      "learning_rate": 7.834042031523643e-05,
      "loss": 1.2192,
      "num_input_tokens_seen": 49778680,
      "step": 3095
    },
    {
      "epoch": 0.21686936877774568,
      "grad_norm": 3.8204843997955322,
      "learning_rate": 7.83334220665499e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 49794344,
      "step": 3096
    },
    {
      "epoch": 0.2169394170234749,
      "grad_norm": 10.573785781860352,
      "learning_rate": 7.832642381786341e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 49810208,
      "step": 3097
    },
    {
      "epoch": 0.21700946526920417,
      "grad_norm": 3.437734603881836,
      "learning_rate": 7.83194255691769e-05,
      "loss": 0.8757,
      "num_input_tokens_seen": 49826448,
      "step": 3098
    },
    {
      "epoch": 0.2170795135149334,
      "grad_norm": 3.476918935775757,
      "learning_rate": 7.831242732049037e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 49842832,
      "step": 3099
    },
    {
      "epoch": 0.21714956176066266,
      "grad_norm": 4.037630558013916,
      "learning_rate": 7.830542907180386e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 49859216,
      "step": 3100
    },
    {
      "epoch": 0.2172196100063919,
      "grad_norm": 3.7424814701080322,
      "learning_rate": 7.829843082311734e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 49875528,
      "step": 3101
    },
    {
      "epoch": 0.21728965825212115,
      "grad_norm": 4.222198486328125,
      "learning_rate": 7.829143257443082e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 49891912,
      "step": 3102
    },
    {
      "epoch": 0.21735970649785039,
      "grad_norm": 4.064510822296143,
      "learning_rate": 7.828443432574431e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 49908064,
      "step": 3103
    },
    {
      "epoch": 0.21742975474357965,
      "grad_norm": 3.822498083114624,
      "learning_rate": 7.82774360770578e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 49923776,
      "step": 3104
    },
    {
      "epoch": 0.21749980298930888,
      "grad_norm": 4.368459224700928,
      "learning_rate": 7.827043782837129e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 49940104,
      "step": 3105
    },
    {
      "epoch": 0.21756985123503814,
      "grad_norm": 3.722587823867798,
      "learning_rate": 7.826343957968477e-05,
      "loss": 1.0286,
      "num_input_tokens_seen": 49955624,
      "step": 3106
    },
    {
      "epoch": 0.21763989948076737,
      "grad_norm": 4.277473449707031,
      "learning_rate": 7.825644133099825e-05,
      "loss": 1.1797,
      "num_input_tokens_seen": 49971784,
      "step": 3107
    },
    {
      "epoch": 0.21770994772649663,
      "grad_norm": 4.586781024932861,
      "learning_rate": 7.824944308231174e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 49988168,
      "step": 3108
    },
    {
      "epoch": 0.21777999597222586,
      "grad_norm": 4.456960201263428,
      "learning_rate": 7.824244483362522e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 50003392,
      "step": 3109
    },
    {
      "epoch": 0.21785004421795512,
      "grad_norm": 4.115220069885254,
      "learning_rate": 7.82354465849387e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 50019776,
      "step": 3110
    },
    {
      "epoch": 0.21792009246368435,
      "grad_norm": 3.5760059356689453,
      "learning_rate": 7.82284483362522e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 50036160,
      "step": 3111
    },
    {
      "epoch": 0.2179901407094136,
      "grad_norm": 5.757627487182617,
      "learning_rate": 7.822145008756568e-05,
      "loss": 1.1382,
      "num_input_tokens_seen": 50052544,
      "step": 3112
    },
    {
      "epoch": 0.21806018895514284,
      "grad_norm": 3.4349796772003174,
      "learning_rate": 7.821445183887916e-05,
      "loss": 0.8474,
      "num_input_tokens_seen": 50068872,
      "step": 3113
    },
    {
      "epoch": 0.2181302372008721,
      "grad_norm": 5.546512603759766,
      "learning_rate": 7.820745359019265e-05,
      "loss": 1.4564,
      "num_input_tokens_seen": 50085256,
      "step": 3114
    },
    {
      "epoch": 0.21820028544660133,
      "grad_norm": 3.4954123497009277,
      "learning_rate": 7.820045534150614e-05,
      "loss": 1.0479,
      "num_input_tokens_seen": 50101640,
      "step": 3115
    },
    {
      "epoch": 0.2182703336923306,
      "grad_norm": 5.396134376525879,
      "learning_rate": 7.819345709281961e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 50117040,
      "step": 3116
    },
    {
      "epoch": 0.21834038193805985,
      "grad_norm": 3.7895803451538086,
      "learning_rate": 7.818645884413311e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 50133424,
      "step": 3117
    },
    {
      "epoch": 0.21841043018378908,
      "grad_norm": 5.321155548095703,
      "learning_rate": 7.81794605954466e-05,
      "loss": 1.2246,
      "num_input_tokens_seen": 50148520,
      "step": 3118
    },
    {
      "epoch": 0.21848047842951834,
      "grad_norm": 4.993834495544434,
      "learning_rate": 7.817246234676008e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 50164904,
      "step": 3119
    },
    {
      "epoch": 0.21855052667524758,
      "grad_norm": 3.69236159324646,
      "learning_rate": 7.816546409807355e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 50181288,
      "step": 3120
    },
    {
      "epoch": 0.21862057492097683,
      "grad_norm": 3.689748764038086,
      "learning_rate": 7.815846584938704e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 50197672,
      "step": 3121
    },
    {
      "epoch": 0.21869062316670607,
      "grad_norm": 5.904501914978027,
      "learning_rate": 7.815146760070053e-05,
      "loss": 1.2348,
      "num_input_tokens_seen": 50214016,
      "step": 3122
    },
    {
      "epoch": 0.21876067141243533,
      "grad_norm": 4.018721103668213,
      "learning_rate": 7.814446935201402e-05,
      "loss": 1.2688,
      "num_input_tokens_seen": 50229984,
      "step": 3123
    },
    {
      "epoch": 0.21883071965816456,
      "grad_norm": 3.6527509689331055,
      "learning_rate": 7.81374711033275e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 50246368,
      "step": 3124
    },
    {
      "epoch": 0.21890076790389382,
      "grad_norm": 3.6600260734558105,
      "learning_rate": 7.8130472854641e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 50262208,
      "step": 3125
    },
    {
      "epoch": 0.21897081614962305,
      "grad_norm": 3.415969133377075,
      "learning_rate": 7.812347460595447e-05,
      "loss": 1.1055,
      "num_input_tokens_seen": 50278592,
      "step": 3126
    },
    {
      "epoch": 0.2190408643953523,
      "grad_norm": 5.027013301849365,
      "learning_rate": 7.811647635726796e-05,
      "loss": 1.2189,
      "num_input_tokens_seen": 50294976,
      "step": 3127
    },
    {
      "epoch": 0.21911091264108154,
      "grad_norm": 3.806324005126953,
      "learning_rate": 7.810947810858143e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 50311360,
      "step": 3128
    },
    {
      "epoch": 0.2191809608868108,
      "grad_norm": 5.208338260650635,
      "learning_rate": 7.810247985989492e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 50327744,
      "step": 3129
    },
    {
      "epoch": 0.21925100913254003,
      "grad_norm": 3.5902316570281982,
      "learning_rate": 7.809548161120841e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 50343616,
      "step": 3130
    },
    {
      "epoch": 0.2193210573782693,
      "grad_norm": 3.7159717082977295,
      "learning_rate": 7.80884833625219e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 50360000,
      "step": 3131
    },
    {
      "epoch": 0.21939110562399852,
      "grad_norm": 4.03640079498291,
      "learning_rate": 7.808148511383539e-05,
      "loss": 0.8906,
      "num_input_tokens_seen": 50376384,
      "step": 3132
    },
    {
      "epoch": 0.21946115386972778,
      "grad_norm": 3.763805627822876,
      "learning_rate": 7.807448686514886e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 50392328,
      "step": 3133
    },
    {
      "epoch": 0.219531202115457,
      "grad_norm": 4.242026329040527,
      "learning_rate": 7.806748861646235e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 50408712,
      "step": 3134
    },
    {
      "epoch": 0.21960125036118627,
      "grad_norm": 3.5783863067626953,
      "learning_rate": 7.806049036777584e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 50424816,
      "step": 3135
    },
    {
      "epoch": 0.2196712986069155,
      "grad_norm": 3.8409011363983154,
      "learning_rate": 7.805349211908931e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 50440464,
      "step": 3136
    },
    {
      "epoch": 0.21974134685264476,
      "grad_norm": 3.642411231994629,
      "learning_rate": 7.804649387040282e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 50456552,
      "step": 3137
    },
    {
      "epoch": 0.219811395098374,
      "grad_norm": 3.702481985092163,
      "learning_rate": 7.803949562171629e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 50472936,
      "step": 3138
    },
    {
      "epoch": 0.21988144334410326,
      "grad_norm": 3.776094913482666,
      "learning_rate": 7.803249737302978e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 50488760,
      "step": 3139
    },
    {
      "epoch": 0.2199514915898325,
      "grad_norm": 4.29668664932251,
      "learning_rate": 7.802549912434326e-05,
      "loss": 1.3095,
      "num_input_tokens_seen": 50505144,
      "step": 3140
    },
    {
      "epoch": 0.22002153983556175,
      "grad_norm": 3.8290088176727295,
      "learning_rate": 7.801850087565674e-05,
      "loss": 0.9331,
      "num_input_tokens_seen": 50521520,
      "step": 3141
    },
    {
      "epoch": 0.22009158808129098,
      "grad_norm": 3.9471163749694824,
      "learning_rate": 7.801150262697023e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 50537688,
      "step": 3142
    },
    {
      "epoch": 0.22016163632702024,
      "grad_norm": 4.3500657081604,
      "learning_rate": 7.800450437828372e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 50554072,
      "step": 3143
    },
    {
      "epoch": 0.22023168457274947,
      "grad_norm": 4.257317066192627,
      "learning_rate": 7.799750612959721e-05,
      "loss": 1.2822,
      "num_input_tokens_seen": 50570456,
      "step": 3144
    },
    {
      "epoch": 0.22030173281847873,
      "grad_norm": 3.881340265274048,
      "learning_rate": 7.79905078809107e-05,
      "loss": 1.2797,
      "num_input_tokens_seen": 50586840,
      "step": 3145
    },
    {
      "epoch": 0.22037178106420796,
      "grad_norm": 4.07082986831665,
      "learning_rate": 7.798350963222417e-05,
      "loss": 1.1659,
      "num_input_tokens_seen": 50603224,
      "step": 3146
    },
    {
      "epoch": 0.22044182930993722,
      "grad_norm": 3.740081310272217,
      "learning_rate": 7.797651138353765e-05,
      "loss": 1.1638,
      "num_input_tokens_seen": 50619608,
      "step": 3147
    },
    {
      "epoch": 0.22051187755566645,
      "grad_norm": 3.9368820190429688,
      "learning_rate": 7.796951313485114e-05,
      "loss": 1.2122,
      "num_input_tokens_seen": 50635240,
      "step": 3148
    },
    {
      "epoch": 0.2205819258013957,
      "grad_norm": 4.027481555938721,
      "learning_rate": 7.796251488616463e-05,
      "loss": 1.1479,
      "num_input_tokens_seen": 50651144,
      "step": 3149
    },
    {
      "epoch": 0.22065197404712494,
      "grad_norm": 3.53271222114563,
      "learning_rate": 7.795551663747811e-05,
      "loss": 0.8352,
      "num_input_tokens_seen": 50667320,
      "step": 3150
    },
    {
      "epoch": 0.2207220222928542,
      "grad_norm": 3.9494400024414062,
      "learning_rate": 7.79485183887916e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 50683704,
      "step": 3151
    },
    {
      "epoch": 0.22079207053858346,
      "grad_norm": 4.944929122924805,
      "learning_rate": 7.794152014010509e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 50699544,
      "step": 3152
    },
    {
      "epoch": 0.2208621187843127,
      "grad_norm": 4.625114440917969,
      "learning_rate": 7.793452189141857e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 50715096,
      "step": 3153
    },
    {
      "epoch": 0.22093216703004195,
      "grad_norm": 4.543829441070557,
      "learning_rate": 7.792752364273205e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 50731480,
      "step": 3154
    },
    {
      "epoch": 0.22100221527577119,
      "grad_norm": 4.038646221160889,
      "learning_rate": 7.792052539404553e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 50747864,
      "step": 3155
    },
    {
      "epoch": 0.22107226352150045,
      "grad_norm": 5.269920825958252,
      "learning_rate": 7.791352714535902e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 50764248,
      "step": 3156
    },
    {
      "epoch": 0.22114231176722968,
      "grad_norm": 3.661792278289795,
      "learning_rate": 7.790652889667251e-05,
      "loss": 0.8881,
      "num_input_tokens_seen": 50780296,
      "step": 3157
    },
    {
      "epoch": 0.22121236001295894,
      "grad_norm": 5.119567394256592,
      "learning_rate": 7.7899530647986e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 50796680,
      "step": 3158
    },
    {
      "epoch": 0.22128240825868817,
      "grad_norm": 4.011631965637207,
      "learning_rate": 7.789253239929948e-05,
      "loss": 1.0854,
      "num_input_tokens_seen": 50812648,
      "step": 3159
    },
    {
      "epoch": 0.22135245650441743,
      "grad_norm": 4.292233467102051,
      "learning_rate": 7.788553415061296e-05,
      "loss": 0.8441,
      "num_input_tokens_seen": 50829032,
      "step": 3160
    },
    {
      "epoch": 0.22142250475014666,
      "grad_norm": 3.9228122234344482,
      "learning_rate": 7.787853590192645e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 50844776,
      "step": 3161
    },
    {
      "epoch": 0.22149255299587592,
      "grad_norm": 4.396078109741211,
      "learning_rate": 7.787153765323994e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 50860792,
      "step": 3162
    },
    {
      "epoch": 0.22156260124160515,
      "grad_norm": 3.6809213161468506,
      "learning_rate": 7.786453940455342e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 50877176,
      "step": 3163
    },
    {
      "epoch": 0.2216326494873344,
      "grad_norm": 3.7879207134246826,
      "learning_rate": 7.785754115586691e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 50893560,
      "step": 3164
    },
    {
      "epoch": 0.22170269773306364,
      "grad_norm": 5.248175621032715,
      "learning_rate": 7.785054290718039e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 50909944,
      "step": 3165
    },
    {
      "epoch": 0.2217727459787929,
      "grad_norm": 5.437406539916992,
      "learning_rate": 7.784354465849388e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 50925848,
      "step": 3166
    },
    {
      "epoch": 0.22184279422452213,
      "grad_norm": 4.390413761138916,
      "learning_rate": 7.783654640980735e-05,
      "loss": 1.143,
      "num_input_tokens_seen": 50941488,
      "step": 3167
    },
    {
      "epoch": 0.2219128424702514,
      "grad_norm": 3.6923863887786865,
      "learning_rate": 7.782954816112084e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 50956984,
      "step": 3168
    },
    {
      "epoch": 0.22198289071598062,
      "grad_norm": 4.338325023651123,
      "learning_rate": 7.782254991243433e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 50973096,
      "step": 3169
    },
    {
      "epoch": 0.22205293896170988,
      "grad_norm": 5.631222248077393,
      "learning_rate": 7.781555166374782e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 50989480,
      "step": 3170
    },
    {
      "epoch": 0.22212298720743912,
      "grad_norm": 3.852337598800659,
      "learning_rate": 7.78085534150613e-05,
      "loss": 1.0279,
      "num_input_tokens_seen": 51005864,
      "step": 3171
    },
    {
      "epoch": 0.22219303545316837,
      "grad_norm": 3.6684298515319824,
      "learning_rate": 7.78015551663748e-05,
      "loss": 0.9906,
      "num_input_tokens_seen": 51022248,
      "step": 3172
    },
    {
      "epoch": 0.2222630836988976,
      "grad_norm": 3.7521257400512695,
      "learning_rate": 7.779455691768827e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 51038632,
      "step": 3173
    },
    {
      "epoch": 0.22233313194462687,
      "grad_norm": 4.571293354034424,
      "learning_rate": 7.778755866900175e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 51055016,
      "step": 3174
    },
    {
      "epoch": 0.2224031801903561,
      "grad_norm": 3.921743154525757,
      "learning_rate": 7.778056042031523e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 51071288,
      "step": 3175
    },
    {
      "epoch": 0.22247322843608536,
      "grad_norm": 6.1666083335876465,
      "learning_rate": 7.777356217162872e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 51086712,
      "step": 3176
    },
    {
      "epoch": 0.2225432766818146,
      "grad_norm": 4.170863151550293,
      "learning_rate": 7.776656392294221e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 51102904,
      "step": 3177
    },
    {
      "epoch": 0.22261332492754385,
      "grad_norm": 4.218405246734619,
      "learning_rate": 7.77595656742557e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 51119288,
      "step": 3178
    },
    {
      "epoch": 0.22268337317327308,
      "grad_norm": 4.158823490142822,
      "learning_rate": 7.775256742556919e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 51135672,
      "step": 3179
    },
    {
      "epoch": 0.22275342141900234,
      "grad_norm": 3.900827407836914,
      "learning_rate": 7.774556917688266e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 51151880,
      "step": 3180
    },
    {
      "epoch": 0.22282346966473157,
      "grad_norm": 3.6363813877105713,
      "learning_rate": 7.773857092819615e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 51167712,
      "step": 3181
    },
    {
      "epoch": 0.22289351791046083,
      "grad_norm": 6.452186584472656,
      "learning_rate": 7.773157267950963e-05,
      "loss": 1.3543,
      "num_input_tokens_seen": 51184096,
      "step": 3182
    },
    {
      "epoch": 0.22296356615619006,
      "grad_norm": 4.324470043182373,
      "learning_rate": 7.772457443082313e-05,
      "loss": 1.3328,
      "num_input_tokens_seen": 51200480,
      "step": 3183
    },
    {
      "epoch": 0.22303361440191932,
      "grad_norm": 4.093019485473633,
      "learning_rate": 7.77175761821366e-05,
      "loss": 1.2647,
      "num_input_tokens_seen": 51216864,
      "step": 3184
    },
    {
      "epoch": 0.22310366264764858,
      "grad_norm": 3.923771619796753,
      "learning_rate": 7.771057793345009e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 51233248,
      "step": 3185
    },
    {
      "epoch": 0.2231737108933778,
      "grad_norm": 3.3340275287628174,
      "learning_rate": 7.770357968476358e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 51249400,
      "step": 3186
    },
    {
      "epoch": 0.22324375913910707,
      "grad_norm": 5.360925197601318,
      "learning_rate": 7.769658143607706e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 51264920,
      "step": 3187
    },
    {
      "epoch": 0.2233138073848363,
      "grad_norm": 4.377450466156006,
      "learning_rate": 7.768958318739054e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 51280528,
      "step": 3188
    },
    {
      "epoch": 0.22338385563056556,
      "grad_norm": 4.01370906829834,
      "learning_rate": 7.768258493870403e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 51296912,
      "step": 3189
    },
    {
      "epoch": 0.2234539038762948,
      "grad_norm": 5.112427711486816,
      "learning_rate": 7.767558669001752e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 51313296,
      "step": 3190
    },
    {
      "epoch": 0.22352395212202406,
      "grad_norm": 3.5889225006103516,
      "learning_rate": 7.766858844133101e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 51329680,
      "step": 3191
    },
    {
      "epoch": 0.2235940003677533,
      "grad_norm": 3.6924920082092285,
      "learning_rate": 7.766159019264449e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 51346064,
      "step": 3192
    },
    {
      "epoch": 0.22366404861348255,
      "grad_norm": 3.9349400997161865,
      "learning_rate": 7.765459194395797e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 51361200,
      "step": 3193
    },
    {
      "epoch": 0.22373409685921178,
      "grad_norm": 3.6980738639831543,
      "learning_rate": 7.764759369527145e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 51377584,
      "step": 3194
    },
    {
      "epoch": 0.22380414510494104,
      "grad_norm": 4.400575637817383,
      "learning_rate": 7.764059544658494e-05,
      "loss": 1.2927,
      "num_input_tokens_seen": 51393968,
      "step": 3195
    },
    {
      "epoch": 0.22387419335067027,
      "grad_norm": 3.758664846420288,
      "learning_rate": 7.763359719789843e-05,
      "loss": 0.8743,
      "num_input_tokens_seen": 51410160,
      "step": 3196
    },
    {
      "epoch": 0.22394424159639953,
      "grad_norm": 4.376255512237549,
      "learning_rate": 7.762659894921192e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 51426192,
      "step": 3197
    },
    {
      "epoch": 0.22401428984212876,
      "grad_norm": 4.371212959289551,
      "learning_rate": 7.76196007005254e-05,
      "loss": 1.4918,
      "num_input_tokens_seen": 51442576,
      "step": 3198
    },
    {
      "epoch": 0.22408433808785802,
      "grad_norm": 3.5152950286865234,
      "learning_rate": 7.761260245183889e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 51458648,
      "step": 3199
    },
    {
      "epoch": 0.22415438633358725,
      "grad_norm": 4.100535869598389,
      "learning_rate": 7.760560420315237e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 51475032,
      "step": 3200
    },
    {
      "epoch": 0.22415438633358725,
      "eval_loss": 1.1358542442321777,
      "eval_runtime": 0.2073,
      "eval_samples_per_second": 4.825,
      "eval_steps_per_second": 4.825,
      "num_input_tokens_seen": 51475032,
      "step": 3200
    },
    {
      "epoch": 0.2242244345793165,
      "grad_norm": 4.394073486328125,
      "learning_rate": 7.759860595446584e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 51490544,
      "step": 3201
    },
    {
      "epoch": 0.22429448282504574,
      "grad_norm": 4.041582107543945,
      "learning_rate": 7.759160770577933e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 51506928,
      "step": 3202
    },
    {
      "epoch": 0.224364531070775,
      "grad_norm": 4.268798351287842,
      "learning_rate": 7.758460945709282e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 51523232,
      "step": 3203
    },
    {
      "epoch": 0.22443457931650423,
      "grad_norm": 4.080141067504883,
      "learning_rate": 7.757761120840631e-05,
      "loss": 0.9809,
      "num_input_tokens_seen": 51539616,
      "step": 3204
    },
    {
      "epoch": 0.2245046275622335,
      "grad_norm": 7.690321445465088,
      "learning_rate": 7.75706129597198e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 51556000,
      "step": 3205
    },
    {
      "epoch": 0.22457467580796273,
      "grad_norm": 4.161118507385254,
      "learning_rate": 7.756361471103329e-05,
      "loss": 0.9672,
      "num_input_tokens_seen": 51572384,
      "step": 3206
    },
    {
      "epoch": 0.22464472405369199,
      "grad_norm": 3.922683000564575,
      "learning_rate": 7.755661646234676e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 51588768,
      "step": 3207
    },
    {
      "epoch": 0.22471477229942122,
      "grad_norm": 3.7474617958068848,
      "learning_rate": 7.754961821366025e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 51604792,
      "step": 3208
    },
    {
      "epoch": 0.22478482054515048,
      "grad_norm": 3.856959819793701,
      "learning_rate": 7.754261996497374e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 51621176,
      "step": 3209
    },
    {
      "epoch": 0.2248548687908797,
      "grad_norm": 4.130929470062256,
      "learning_rate": 7.753562171628723e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 51636864,
      "step": 3210
    },
    {
      "epoch": 0.22492491703660897,
      "grad_norm": 3.5023388862609863,
      "learning_rate": 7.75286234676007e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 51653248,
      "step": 3211
    },
    {
      "epoch": 0.2249949652823382,
      "grad_norm": 3.736415386199951,
      "learning_rate": 7.752162521891419e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 51669632,
      "step": 3212
    },
    {
      "epoch": 0.22506501352806746,
      "grad_norm": 4.355846881866455,
      "learning_rate": 7.751462697022768e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 51684632,
      "step": 3213
    },
    {
      "epoch": 0.2251350617737967,
      "grad_norm": 4.165436744689941,
      "learning_rate": 7.750762872154115e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 51701016,
      "step": 3214
    },
    {
      "epoch": 0.22520511001952595,
      "grad_norm": 4.4387946128845215,
      "learning_rate": 7.750063047285464e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 51716176,
      "step": 3215
    },
    {
      "epoch": 0.22527515826525518,
      "grad_norm": 4.749145030975342,
      "learning_rate": 7.749363222416813e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 51732560,
      "step": 3216
    },
    {
      "epoch": 0.22534520651098444,
      "grad_norm": 4.321863651275635,
      "learning_rate": 7.748663397548162e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 51748944,
      "step": 3217
    },
    {
      "epoch": 0.22541525475671367,
      "grad_norm": 5.319899082183838,
      "learning_rate": 7.747963572679511e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 51765328,
      "step": 3218
    },
    {
      "epoch": 0.22548530300244293,
      "grad_norm": 3.5695643424987793,
      "learning_rate": 7.747263747810858e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 51781712,
      "step": 3219
    },
    {
      "epoch": 0.2255553512481722,
      "grad_norm": 3.725698947906494,
      "learning_rate": 7.746563922942207e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 51798096,
      "step": 3220
    },
    {
      "epoch": 0.22562539949390142,
      "grad_norm": 3.795003652572632,
      "learning_rate": 7.745864098073555e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 51814480,
      "step": 3221
    },
    {
      "epoch": 0.22569544773963068,
      "grad_norm": 3.817578077316284,
      "learning_rate": 7.745164273204903e-05,
      "loss": 1.1218,
      "num_input_tokens_seen": 51830864,
      "step": 3222
    },
    {
      "epoch": 0.22576549598535992,
      "grad_norm": 5.982937812805176,
      "learning_rate": 7.744464448336252e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 51846104,
      "step": 3223
    },
    {
      "epoch": 0.22583554423108917,
      "grad_norm": 5.063079833984375,
      "learning_rate": 7.743764623467601e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 51862488,
      "step": 3224
    },
    {
      "epoch": 0.2259055924768184,
      "grad_norm": 3.620837450027466,
      "learning_rate": 7.74306479859895e-05,
      "loss": 1.0484,
      "num_input_tokens_seen": 51878784,
      "step": 3225
    },
    {
      "epoch": 0.22597564072254767,
      "grad_norm": 3.578369617462158,
      "learning_rate": 7.742364973730299e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 51894832,
      "step": 3226
    },
    {
      "epoch": 0.2260456889682769,
      "grad_norm": 4.0356974601745605,
      "learning_rate": 7.741665148861646e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 51911216,
      "step": 3227
    },
    {
      "epoch": 0.22611573721400616,
      "grad_norm": 4.133927822113037,
      "learning_rate": 7.740965323992994e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 51927600,
      "step": 3228
    },
    {
      "epoch": 0.2261857854597354,
      "grad_norm": 4.2958879470825195,
      "learning_rate": 7.740265499124343e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 51943688,
      "step": 3229
    },
    {
      "epoch": 0.22625583370546465,
      "grad_norm": 6.211035251617432,
      "learning_rate": 7.739565674255693e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 51960072,
      "step": 3230
    },
    {
      "epoch": 0.22632588195119388,
      "grad_norm": 4.073126316070557,
      "learning_rate": 7.73886584938704e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 51976456,
      "step": 3231
    },
    {
      "epoch": 0.22639593019692314,
      "grad_norm": 3.605041980743408,
      "learning_rate": 7.73816602451839e-05,
      "loss": 0.817,
      "num_input_tokens_seen": 51992840,
      "step": 3232
    },
    {
      "epoch": 0.22646597844265237,
      "grad_norm": 4.341184139251709,
      "learning_rate": 7.737466199649738e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 52008696,
      "step": 3233
    },
    {
      "epoch": 0.22653602668838163,
      "grad_norm": 4.676966667175293,
      "learning_rate": 7.736766374781086e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 52024944,
      "step": 3234
    },
    {
      "epoch": 0.22660607493411086,
      "grad_norm": 4.6688032150268555,
      "learning_rate": 7.736066549912435e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 52041104,
      "step": 3235
    },
    {
      "epoch": 0.22667612317984012,
      "grad_norm": 4.6416916847229,
      "learning_rate": 7.735366725043783e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 52055864,
      "step": 3236
    },
    {
      "epoch": 0.22674617142556935,
      "grad_norm": 3.713846206665039,
      "learning_rate": 7.734666900175132e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 52071992,
      "step": 3237
    },
    {
      "epoch": 0.2268162196712986,
      "grad_norm": 3.694094657897949,
      "learning_rate": 7.73396707530648e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 52088376,
      "step": 3238
    },
    {
      "epoch": 0.22688626791702785,
      "grad_norm": 4.250162601470947,
      "learning_rate": 7.733267250437829e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 52104320,
      "step": 3239
    },
    {
      "epoch": 0.2269563161627571,
      "grad_norm": 3.8184008598327637,
      "learning_rate": 7.732567425569178e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 52120416,
      "step": 3240
    },
    {
      "epoch": 0.22702636440848634,
      "grad_norm": 3.9957122802734375,
      "learning_rate": 7.731867600700525e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 52136704,
      "step": 3241
    },
    {
      "epoch": 0.2270964126542156,
      "grad_norm": 4.153292655944824,
      "learning_rate": 7.731167775831874e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 52153088,
      "step": 3242
    },
    {
      "epoch": 0.22716646089994483,
      "grad_norm": 3.628377914428711,
      "learning_rate": 7.730467950963223e-05,
      "loss": 0.9826,
      "num_input_tokens_seen": 52169032,
      "step": 3243
    },
    {
      "epoch": 0.2272365091456741,
      "grad_norm": 3.45796275138855,
      "learning_rate": 7.729768126094572e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 52185416,
      "step": 3244
    },
    {
      "epoch": 0.22730655739140332,
      "grad_norm": 3.9128968715667725,
      "learning_rate": 7.72906830122592e-05,
      "loss": 1.2954,
      "num_input_tokens_seen": 52201504,
      "step": 3245
    },
    {
      "epoch": 0.22737660563713258,
      "grad_norm": 4.4097394943237305,
      "learning_rate": 7.728368476357268e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 52217184,
      "step": 3246
    },
    {
      "epoch": 0.2274466538828618,
      "grad_norm": 4.110626220703125,
      "learning_rate": 7.727668651488617e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 52233432,
      "step": 3247
    },
    {
      "epoch": 0.22751670212859107,
      "grad_norm": 4.161354064941406,
      "learning_rate": 7.726968826619964e-05,
      "loss": 0.9371,
      "num_input_tokens_seen": 52249816,
      "step": 3248
    },
    {
      "epoch": 0.2275867503743203,
      "grad_norm": 5.910977363586426,
      "learning_rate": 7.726269001751313e-05,
      "loss": 0.8993,
      "num_input_tokens_seen": 52266200,
      "step": 3249
    },
    {
      "epoch": 0.22765679862004956,
      "grad_norm": 3.8264660835266113,
      "learning_rate": 7.725569176882663e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 52282136,
      "step": 3250
    },
    {
      "epoch": 0.2277268468657788,
      "grad_norm": 3.9992623329162598,
      "learning_rate": 7.724869352014011e-05,
      "loss": 0.9256,
      "num_input_tokens_seen": 52297368,
      "step": 3251
    },
    {
      "epoch": 0.22779689511150805,
      "grad_norm": 4.263967990875244,
      "learning_rate": 7.72416952714536e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 52313200,
      "step": 3252
    },
    {
      "epoch": 0.22786694335723728,
      "grad_norm": 3.8846871852874756,
      "learning_rate": 7.723469702276709e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 52329584,
      "step": 3253
    },
    {
      "epoch": 0.22793699160296654,
      "grad_norm": 4.3504533767700195,
      "learning_rate": 7.722769877408056e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 52345968,
      "step": 3254
    },
    {
      "epoch": 0.2280070398486958,
      "grad_norm": 3.9775991439819336,
      "learning_rate": 7.722070052539404e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 52362352,
      "step": 3255
    },
    {
      "epoch": 0.22807708809442503,
      "grad_norm": 4.098363399505615,
      "learning_rate": 7.721370227670754e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 52378736,
      "step": 3256
    },
    {
      "epoch": 0.2281471363401543,
      "grad_norm": 3.7094836235046387,
      "learning_rate": 7.720670402802103e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 52394896,
      "step": 3257
    },
    {
      "epoch": 0.22821718458588353,
      "grad_norm": 4.042232036590576,
      "learning_rate": 7.71997057793345e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 52410952,
      "step": 3258
    },
    {
      "epoch": 0.22828723283161279,
      "grad_norm": 3.725853443145752,
      "learning_rate": 7.719270753064799e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 52427200,
      "step": 3259
    },
    {
      "epoch": 0.22835728107734202,
      "grad_norm": 5.186229705810547,
      "learning_rate": 7.718570928196148e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 52443584,
      "step": 3260
    },
    {
      "epoch": 0.22842732932307128,
      "grad_norm": 3.8725364208221436,
      "learning_rate": 7.717871103327495e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 52458272,
      "step": 3261
    },
    {
      "epoch": 0.2284973775688005,
      "grad_norm": 5.006584644317627,
      "learning_rate": 7.717171278458844e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 52474456,
      "step": 3262
    },
    {
      "epoch": 0.22856742581452977,
      "grad_norm": 5.102536201477051,
      "learning_rate": 7.716471453590193e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 52490464,
      "step": 3263
    },
    {
      "epoch": 0.228637474060259,
      "grad_norm": 3.741029977798462,
      "learning_rate": 7.715771628721542e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 52506112,
      "step": 3264
    },
    {
      "epoch": 0.22870752230598826,
      "grad_norm": 5.656842231750488,
      "learning_rate": 7.71507180385289e-05,
      "loss": 1.1569,
      "num_input_tokens_seen": 52522496,
      "step": 3265
    },
    {
      "epoch": 0.2287775705517175,
      "grad_norm": 3.882403612136841,
      "learning_rate": 7.714371978984238e-05,
      "loss": 1.163,
      "num_input_tokens_seen": 52538240,
      "step": 3266
    },
    {
      "epoch": 0.22884761879744675,
      "grad_norm": 4.812796592712402,
      "learning_rate": 7.713672154115587e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 52554024,
      "step": 3267
    },
    {
      "epoch": 0.22891766704317598,
      "grad_norm": 3.9040687084198,
      "learning_rate": 7.712972329246935e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 52570408,
      "step": 3268
    },
    {
      "epoch": 0.22898771528890524,
      "grad_norm": 3.8387644290924072,
      "learning_rate": 7.712272504378284e-05,
      "loss": 0.9401,
      "num_input_tokens_seen": 52586512,
      "step": 3269
    },
    {
      "epoch": 0.22905776353463447,
      "grad_norm": 4.602542877197266,
      "learning_rate": 7.711572679509634e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 52602896,
      "step": 3270
    },
    {
      "epoch": 0.22912781178036373,
      "grad_norm": 4.209007263183594,
      "learning_rate": 7.710872854640981e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 52619080,
      "step": 3271
    },
    {
      "epoch": 0.22919786002609296,
      "grad_norm": 3.5082032680511475,
      "learning_rate": 7.71017302977233e-05,
      "loss": 0.9979,
      "num_input_tokens_seen": 52635464,
      "step": 3272
    },
    {
      "epoch": 0.22926790827182222,
      "grad_norm": 4.123980522155762,
      "learning_rate": 7.709473204903678e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 52651848,
      "step": 3273
    },
    {
      "epoch": 0.22933795651755146,
      "grad_norm": 4.267751216888428,
      "learning_rate": 7.708773380035027e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 52668232,
      "step": 3274
    },
    {
      "epoch": 0.22940800476328072,
      "grad_norm": 4.1165666580200195,
      "learning_rate": 7.708073555166374e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 52684616,
      "step": 3275
    },
    {
      "epoch": 0.22947805300900995,
      "grad_norm": 4.810427665710449,
      "learning_rate": 7.707373730297724e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 52701000,
      "step": 3276
    },
    {
      "epoch": 0.2295481012547392,
      "grad_norm": 6.566617488861084,
      "learning_rate": 7.706673905429073e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 52715920,
      "step": 3277
    },
    {
      "epoch": 0.22961814950046844,
      "grad_norm": 4.456092834472656,
      "learning_rate": 7.70597408056042e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 52732304,
      "step": 3278
    },
    {
      "epoch": 0.2296881977461977,
      "grad_norm": 4.063642501831055,
      "learning_rate": 7.70527425569177e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 52748688,
      "step": 3279
    },
    {
      "epoch": 0.22975824599192693,
      "grad_norm": 3.337742567062378,
      "learning_rate": 7.704574430823118e-05,
      "loss": 0.8497,
      "num_input_tokens_seen": 52764800,
      "step": 3280
    },
    {
      "epoch": 0.2298282942376562,
      "grad_norm": 4.36488151550293,
      "learning_rate": 7.703874605954466e-05,
      "loss": 1.0851,
      "num_input_tokens_seen": 52780952,
      "step": 3281
    },
    {
      "epoch": 0.22989834248338542,
      "grad_norm": 4.948200702667236,
      "learning_rate": 7.703174781085815e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 52795728,
      "step": 3282
    },
    {
      "epoch": 0.22996839072911468,
      "grad_norm": 4.977625370025635,
      "learning_rate": 7.702474956217164e-05,
      "loss": 1.2094,
      "num_input_tokens_seen": 52812112,
      "step": 3283
    },
    {
      "epoch": 0.2300384389748439,
      "grad_norm": 3.7551944255828857,
      "learning_rate": 7.701775131348512e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 52828184,
      "step": 3284
    },
    {
      "epoch": 0.23010848722057317,
      "grad_norm": 3.700916051864624,
      "learning_rate": 7.70107530647986e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 52844568,
      "step": 3285
    },
    {
      "epoch": 0.2301785354663024,
      "grad_norm": 4.135788917541504,
      "learning_rate": 7.700375481611209e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 52860952,
      "step": 3286
    },
    {
      "epoch": 0.23024858371203166,
      "grad_norm": 4.018477916717529,
      "learning_rate": 7.699675656742558e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 52876808,
      "step": 3287
    },
    {
      "epoch": 0.2303186319577609,
      "grad_norm": 5.230745315551758,
      "learning_rate": 7.698975831873905e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 52893192,
      "step": 3288
    },
    {
      "epoch": 0.23038868020349015,
      "grad_norm": 4.192041873931885,
      "learning_rate": 7.698276007005254e-05,
      "loss": 1.1476,
      "num_input_tokens_seen": 52909576,
      "step": 3289
    },
    {
      "epoch": 0.2304587284492194,
      "grad_norm": 4.28109073638916,
      "learning_rate": 7.697576182136603e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 52925592,
      "step": 3290
    },
    {
      "epoch": 0.23052877669494864,
      "grad_norm": 4.673538684844971,
      "learning_rate": 7.696876357267952e-05,
      "loss": 1.2104,
      "num_input_tokens_seen": 52941784,
      "step": 3291
    },
    {
      "epoch": 0.2305988249406779,
      "grad_norm": 3.791339159011841,
      "learning_rate": 7.696176532399299e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 52958168,
      "step": 3292
    },
    {
      "epoch": 0.23066887318640714,
      "grad_norm": 5.353015899658203,
      "learning_rate": 7.695476707530648e-05,
      "loss": 1.346,
      "num_input_tokens_seen": 52974552,
      "step": 3293
    },
    {
      "epoch": 0.2307389214321364,
      "grad_norm": 6.66793966293335,
      "learning_rate": 7.694776882661997e-05,
      "loss": 1.127,
      "num_input_tokens_seen": 52990512,
      "step": 3294
    },
    {
      "epoch": 0.23080896967786563,
      "grad_norm": 5.462240695953369,
      "learning_rate": 7.694077057793344e-05,
      "loss": 1.2397,
      "num_input_tokens_seen": 53006768,
      "step": 3295
    },
    {
      "epoch": 0.2308790179235949,
      "grad_norm": 4.212863445281982,
      "learning_rate": 7.693377232924695e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 53023152,
      "step": 3296
    },
    {
      "epoch": 0.23094906616932412,
      "grad_norm": 3.623929977416992,
      "learning_rate": 7.692677408056044e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 53039536,
      "step": 3297
    },
    {
      "epoch": 0.23101911441505338,
      "grad_norm": 4.791571617126465,
      "learning_rate": 7.691977583187391e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 53055920,
      "step": 3298
    },
    {
      "epoch": 0.2310891626607826,
      "grad_norm": 3.733243465423584,
      "learning_rate": 7.69127775831874e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 53072304,
      "step": 3299
    },
    {
      "epoch": 0.23115921090651187,
      "grad_norm": 3.916738986968994,
      "learning_rate": 7.690577933450087e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 53088568,
      "step": 3300
    },
    {
      "epoch": 0.2312292591522411,
      "grad_norm": 4.0346856117248535,
      "learning_rate": 7.689878108581436e-05,
      "loss": 1.0858,
      "num_input_tokens_seen": 53103656,
      "step": 3301
    },
    {
      "epoch": 0.23129930739797036,
      "grad_norm": 4.834316730499268,
      "learning_rate": 7.689178283712785e-05,
      "loss": 0.9328,
      "num_input_tokens_seen": 53120040,
      "step": 3302
    },
    {
      "epoch": 0.2313693556436996,
      "grad_norm": 4.5966291427612305,
      "learning_rate": 7.688478458844134e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 53136424,
      "step": 3303
    },
    {
      "epoch": 0.23143940388942885,
      "grad_norm": 5.17268705368042,
      "learning_rate": 7.687778633975483e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 53152080,
      "step": 3304
    },
    {
      "epoch": 0.23150945213515808,
      "grad_norm": 3.6322672367095947,
      "learning_rate": 7.68707880910683e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 53168464,
      "step": 3305
    },
    {
      "epoch": 0.23157950038088734,
      "grad_norm": 4.761613368988037,
      "learning_rate": 7.686378984238179e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 53184848,
      "step": 3306
    },
    {
      "epoch": 0.23164954862661657,
      "grad_norm": 3.4870493412017822,
      "learning_rate": 7.685679159369528e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 53201232,
      "step": 3307
    },
    {
      "epoch": 0.23171959687234583,
      "grad_norm": 4.122028827667236,
      "learning_rate": 7.684979334500876e-05,
      "loss": 1.2103,
      "num_input_tokens_seen": 53217616,
      "step": 3308
    },
    {
      "epoch": 0.23178964511807507,
      "grad_norm": 3.4486751556396484,
      "learning_rate": 7.684279509632224e-05,
      "loss": 0.6654,
      "num_input_tokens_seen": 53233936,
      "step": 3309
    },
    {
      "epoch": 0.23185969336380433,
      "grad_norm": 4.321650981903076,
      "learning_rate": 7.683579684763573e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 53250320,
      "step": 3310
    },
    {
      "epoch": 0.23192974160953356,
      "grad_norm": 5.820108413696289,
      "learning_rate": 7.682879859894922e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 53266592,
      "step": 3311
    },
    {
      "epoch": 0.23199978985526282,
      "grad_norm": 5.5514912605285645,
      "learning_rate": 7.68218003502627e-05,
      "loss": 1.1083,
      "num_input_tokens_seen": 53282976,
      "step": 3312
    },
    {
      "epoch": 0.23206983810099205,
      "grad_norm": 4.108302116394043,
      "learning_rate": 7.681480210157618e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 53299184,
      "step": 3313
    },
    {
      "epoch": 0.2321398863467213,
      "grad_norm": 4.037779331207275,
      "learning_rate": 7.680780385288967e-05,
      "loss": 1.2858,
      "num_input_tokens_seen": 53315000,
      "step": 3314
    },
    {
      "epoch": 0.23220993459245054,
      "grad_norm": 4.5398383140563965,
      "learning_rate": 7.680080560420315e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 53331104,
      "step": 3315
    },
    {
      "epoch": 0.2322799828381798,
      "grad_norm": 4.2399067878723145,
      "learning_rate": 7.679380735551665e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 53347488,
      "step": 3316
    },
    {
      "epoch": 0.23235003108390903,
      "grad_norm": 5.6600775718688965,
      "learning_rate": 7.678680910683013e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 53363872,
      "step": 3317
    },
    {
      "epoch": 0.2324200793296383,
      "grad_norm": 4.462069511413574,
      "learning_rate": 7.677981085814361e-05,
      "loss": 0.9313,
      "num_input_tokens_seen": 53379424,
      "step": 3318
    },
    {
      "epoch": 0.23249012757536752,
      "grad_norm": 4.644591808319092,
      "learning_rate": 7.677281260945709e-05,
      "loss": 1.3155,
      "num_input_tokens_seen": 53395728,
      "step": 3319
    },
    {
      "epoch": 0.23256017582109678,
      "grad_norm": 3.860954523086548,
      "learning_rate": 7.676581436077058e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 53412112,
      "step": 3320
    },
    {
      "epoch": 0.232630224066826,
      "grad_norm": 4.625146389007568,
      "learning_rate": 7.675881611208407e-05,
      "loss": 0.9253,
      "num_input_tokens_seen": 53427992,
      "step": 3321
    },
    {
      "epoch": 0.23270027231255527,
      "grad_norm": 6.473335266113281,
      "learning_rate": 7.675181786339756e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 53444376,
      "step": 3322
    },
    {
      "epoch": 0.2327703205582845,
      "grad_norm": 3.6846091747283936,
      "learning_rate": 7.674481961471104e-05,
      "loss": 0.9976,
      "num_input_tokens_seen": 53460760,
      "step": 3323
    },
    {
      "epoch": 0.23284036880401376,
      "grad_norm": 3.784900188446045,
      "learning_rate": 7.673782136602453e-05,
      "loss": 0.8865,
      "num_input_tokens_seen": 53477144,
      "step": 3324
    },
    {
      "epoch": 0.23291041704974302,
      "grad_norm": 4.175132751464844,
      "learning_rate": 7.673082311733801e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 53493496,
      "step": 3325
    },
    {
      "epoch": 0.23298046529547226,
      "grad_norm": 4.355600833892822,
      "learning_rate": 7.67238248686515e-05,
      "loss": 0.8686,
      "num_input_tokens_seen": 53509560,
      "step": 3326
    },
    {
      "epoch": 0.23305051354120151,
      "grad_norm": 4.32242488861084,
      "learning_rate": 7.671682661996497e-05,
      "loss": 0.9493,
      "num_input_tokens_seen": 53525944,
      "step": 3327
    },
    {
      "epoch": 0.23312056178693075,
      "grad_norm": 4.937814235687256,
      "learning_rate": 7.670982837127846e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 53541312,
      "step": 3328
    },
    {
      "epoch": 0.23319061003266,
      "grad_norm": 3.1939101219177246,
      "learning_rate": 7.670283012259195e-05,
      "loss": 0.8866,
      "num_input_tokens_seen": 53557696,
      "step": 3329
    },
    {
      "epoch": 0.23326065827838924,
      "grad_norm": 5.137113094329834,
      "learning_rate": 7.669583187390544e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 53573600,
      "step": 3330
    },
    {
      "epoch": 0.2333307065241185,
      "grad_norm": 3.777954578399658,
      "learning_rate": 7.668883362521893e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 53588808,
      "step": 3331
    },
    {
      "epoch": 0.23340075476984773,
      "grad_norm": 4.229750633239746,
      "learning_rate": 7.66818353765324e-05,
      "loss": 1.3247,
      "num_input_tokens_seen": 53603416,
      "step": 3332
    },
    {
      "epoch": 0.233470803015577,
      "grad_norm": 4.248676776885986,
      "learning_rate": 7.667483712784589e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 53618896,
      "step": 3333
    },
    {
      "epoch": 0.23354085126130622,
      "grad_norm": 3.7393991947174072,
      "learning_rate": 7.666783887915938e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 53635280,
      "step": 3334
    },
    {
      "epoch": 0.23361089950703548,
      "grad_norm": 3.6224875450134277,
      "learning_rate": 7.666084063047285e-05,
      "loss": 0.8727,
      "num_input_tokens_seen": 53651664,
      "step": 3335
    },
    {
      "epoch": 0.2336809477527647,
      "grad_norm": 4.2722063064575195,
      "learning_rate": 7.665384238178634e-05,
      "loss": 1.1982,
      "num_input_tokens_seen": 53668048,
      "step": 3336
    },
    {
      "epoch": 0.23375099599849397,
      "grad_norm": 3.4717535972595215,
      "learning_rate": 7.664684413309983e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 53684432,
      "step": 3337
    },
    {
      "epoch": 0.2338210442442232,
      "grad_norm": 3.6640021800994873,
      "learning_rate": 7.663984588441332e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 53700816,
      "step": 3338
    },
    {
      "epoch": 0.23389109248995246,
      "grad_norm": 5.14633321762085,
      "learning_rate": 7.66328476357268e-05,
      "loss": 1.1954,
      "num_input_tokens_seen": 53717200,
      "step": 3339
    },
    {
      "epoch": 0.2339611407356817,
      "grad_norm": 4.479960918426514,
      "learning_rate": 7.662584938704028e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 53733584,
      "step": 3340
    },
    {
      "epoch": 0.23403118898141095,
      "grad_norm": 5.33896017074585,
      "learning_rate": 7.661885113835377e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 53749072,
      "step": 3341
    },
    {
      "epoch": 0.23410123722714019,
      "grad_norm": 4.407443046569824,
      "learning_rate": 7.661185288966726e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 53765088,
      "step": 3342
    },
    {
      "epoch": 0.23417128547286944,
      "grad_norm": 3.8250956535339355,
      "learning_rate": 7.660485464098075e-05,
      "loss": 0.9243,
      "num_input_tokens_seen": 53781000,
      "step": 3343
    },
    {
      "epoch": 0.23424133371859868,
      "grad_norm": 4.316215515136719,
      "learning_rate": 7.659785639229422e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 53796744,
      "step": 3344
    },
    {
      "epoch": 0.23431138196432794,
      "grad_norm": 4.291647434234619,
      "learning_rate": 7.659085814360771e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 53813128,
      "step": 3345
    },
    {
      "epoch": 0.23438143021005717,
      "grad_norm": 3.704899787902832,
      "learning_rate": 7.658385989492119e-05,
      "loss": 1.2117,
      "num_input_tokens_seen": 53829512,
      "step": 3346
    },
    {
      "epoch": 0.23445147845578643,
      "grad_norm": 3.5979909896850586,
      "learning_rate": 7.657686164623468e-05,
      "loss": 0.9604,
      "num_input_tokens_seen": 53845536,
      "step": 3347
    },
    {
      "epoch": 0.23452152670151566,
      "grad_norm": 3.8820247650146484,
      "learning_rate": 7.656986339754816e-05,
      "loss": 1.2439,
      "num_input_tokens_seen": 53861920,
      "step": 3348
    },
    {
      "epoch": 0.23459157494724492,
      "grad_norm": 4.226894855499268,
      "learning_rate": 7.656286514886165e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 53878304,
      "step": 3349
    },
    {
      "epoch": 0.23466162319297415,
      "grad_norm": 4.507336616516113,
      "learning_rate": 7.655586690017514e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 53894688,
      "step": 3350
    },
    {
      "epoch": 0.2347316714387034,
      "grad_norm": 3.86645245552063,
      "learning_rate": 7.654886865148863e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 53910736,
      "step": 3351
    },
    {
      "epoch": 0.23480171968443264,
      "grad_norm": 3.8789820671081543,
      "learning_rate": 7.65418704028021e-05,
      "loss": 1.0078,
      "num_input_tokens_seen": 53926688,
      "step": 3352
    },
    {
      "epoch": 0.2348717679301619,
      "grad_norm": 3.893564462661743,
      "learning_rate": 7.653487215411559e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 53942904,
      "step": 3353
    },
    {
      "epoch": 0.23494181617589113,
      "grad_norm": 4.6554412841796875,
      "learning_rate": 7.652787390542907e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 53957976,
      "step": 3354
    },
    {
      "epoch": 0.2350118644216204,
      "grad_norm": 4.118137359619141,
      "learning_rate": 7.652087565674256e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 53973520,
      "step": 3355
    },
    {
      "epoch": 0.23508191266734962,
      "grad_norm": 5.099210262298584,
      "learning_rate": 7.651387740805605e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 53989280,
      "step": 3356
    },
    {
      "epoch": 0.23515196091307888,
      "grad_norm": 3.868797779083252,
      "learning_rate": 7.650687915936953e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 54005664,
      "step": 3357
    },
    {
      "epoch": 0.23522200915880812,
      "grad_norm": 4.032477378845215,
      "learning_rate": 7.649988091068302e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 54022048,
      "step": 3358
    },
    {
      "epoch": 0.23529205740453737,
      "grad_norm": 3.907238483428955,
      "learning_rate": 7.64928826619965e-05,
      "loss": 1.0925,
      "num_input_tokens_seen": 54038432,
      "step": 3359
    },
    {
      "epoch": 0.23536210565026663,
      "grad_norm": 3.6504223346710205,
      "learning_rate": 7.648588441330999e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 54054272,
      "step": 3360
    },
    {
      "epoch": 0.23543215389599587,
      "grad_norm": 4.614812850952148,
      "learning_rate": 7.647888616462347e-05,
      "loss": 1.136,
      "num_input_tokens_seen": 54070656,
      "step": 3361
    },
    {
      "epoch": 0.23550220214172513,
      "grad_norm": 4.812591552734375,
      "learning_rate": 7.647188791593696e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 54086416,
      "step": 3362
    },
    {
      "epoch": 0.23557225038745436,
      "grad_norm": 3.709543466567993,
      "learning_rate": 7.646488966725044e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 54102800,
      "step": 3363
    },
    {
      "epoch": 0.23564229863318362,
      "grad_norm": 3.9850802421569824,
      "learning_rate": 7.645789141856393e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 54119184,
      "step": 3364
    },
    {
      "epoch": 0.23571234687891285,
      "grad_norm": 4.59740686416626,
      "learning_rate": 7.645089316987742e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 54135568,
      "step": 3365
    },
    {
      "epoch": 0.2357823951246421,
      "grad_norm": 4.118459224700928,
      "learning_rate": 7.644389492119089e-05,
      "loss": 1.2196,
      "num_input_tokens_seen": 54151952,
      "step": 3366
    },
    {
      "epoch": 0.23585244337037134,
      "grad_norm": 4.172552108764648,
      "learning_rate": 7.643689667250438e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 54167776,
      "step": 3367
    },
    {
      "epoch": 0.2359224916161006,
      "grad_norm": 3.9671120643615723,
      "learning_rate": 7.642989842381787e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 54184160,
      "step": 3368
    },
    {
      "epoch": 0.23599253986182983,
      "grad_norm": 3.7376415729522705,
      "learning_rate": 7.642290017513136e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 54200280,
      "step": 3369
    },
    {
      "epoch": 0.2360625881075591,
      "grad_norm": 4.665002346038818,
      "learning_rate": 7.641590192644484e-05,
      "loss": 1.3347,
      "num_input_tokens_seen": 54216664,
      "step": 3370
    },
    {
      "epoch": 0.23613263635328832,
      "grad_norm": 3.669015884399414,
      "learning_rate": 7.640890367775832e-05,
      "loss": 0.8359,
      "num_input_tokens_seen": 54232320,
      "step": 3371
    },
    {
      "epoch": 0.23620268459901758,
      "grad_norm": 3.993393659591675,
      "learning_rate": 7.640190542907181e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 54248704,
      "step": 3372
    },
    {
      "epoch": 0.2362727328447468,
      "grad_norm": 3.808516263961792,
      "learning_rate": 7.639490718038528e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 54265088,
      "step": 3373
    },
    {
      "epoch": 0.23634278109047607,
      "grad_norm": 5.25230073928833,
      "learning_rate": 7.638790893169877e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 54281256,
      "step": 3374
    },
    {
      "epoch": 0.2364128293362053,
      "grad_norm": 5.724976062774658,
      "learning_rate": 7.638091068301226e-05,
      "loss": 1.3176,
      "num_input_tokens_seen": 54296832,
      "step": 3375
    },
    {
      "epoch": 0.23648287758193456,
      "grad_norm": 3.553737163543701,
      "learning_rate": 7.637391243432575e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 54313120,
      "step": 3376
    },
    {
      "epoch": 0.2365529258276638,
      "grad_norm": 6.614949703216553,
      "learning_rate": 7.636691418563924e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 54328184,
      "step": 3377
    },
    {
      "epoch": 0.23662297407339306,
      "grad_norm": 3.76234769821167,
      "learning_rate": 7.635991593695273e-05,
      "loss": 1.149,
      "num_input_tokens_seen": 54344568,
      "step": 3378
    },
    {
      "epoch": 0.2366930223191223,
      "grad_norm": 3.4564521312713623,
      "learning_rate": 7.63529176882662e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 54360952,
      "step": 3379
    },
    {
      "epoch": 0.23676307056485155,
      "grad_norm": 3.735978841781616,
      "learning_rate": 7.634591943957969e-05,
      "loss": 1.2159,
      "num_input_tokens_seen": 54377336,
      "step": 3380
    },
    {
      "epoch": 0.23683311881058078,
      "grad_norm": 4.106653690338135,
      "learning_rate": 7.633892119089317e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 54393232,
      "step": 3381
    },
    {
      "epoch": 0.23690316705631004,
      "grad_norm": 3.9169600009918213,
      "learning_rate": 7.633192294220667e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 54409616,
      "step": 3382
    },
    {
      "epoch": 0.23697321530203927,
      "grad_norm": 3.8265388011932373,
      "learning_rate": 7.632492469352014e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 54425312,
      "step": 3383
    },
    {
      "epoch": 0.23704326354776853,
      "grad_norm": 3.6288204193115234,
      "learning_rate": 7.631792644483363e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 54441696,
      "step": 3384
    },
    {
      "epoch": 0.23711331179349776,
      "grad_norm": 4.207483291625977,
      "learning_rate": 7.631092819614712e-05,
      "loss": 1.2068,
      "num_input_tokens_seen": 54457720,
      "step": 3385
    },
    {
      "epoch": 0.23718336003922702,
      "grad_norm": 3.880786895751953,
      "learning_rate": 7.63039299474606e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 54474104,
      "step": 3386
    },
    {
      "epoch": 0.23725340828495625,
      "grad_norm": 4.493243217468262,
      "learning_rate": 7.629693169877408e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 54490080,
      "step": 3387
    },
    {
      "epoch": 0.2373234565306855,
      "grad_norm": 4.432561874389648,
      "learning_rate": 7.628993345008757e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 54506464,
      "step": 3388
    },
    {
      "epoch": 0.23739350477641474,
      "grad_norm": 4.210158824920654,
      "learning_rate": 7.628293520140106e-05,
      "loss": 1.1567,
      "num_input_tokens_seen": 54522848,
      "step": 3389
    },
    {
      "epoch": 0.237463553022144,
      "grad_norm": 4.561443328857422,
      "learning_rate": 7.627593695271454e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 54538192,
      "step": 3390
    },
    {
      "epoch": 0.23753360126787323,
      "grad_norm": 3.6792140007019043,
      "learning_rate": 7.626893870402802e-05,
      "loss": 0.9692,
      "num_input_tokens_seen": 54554576,
      "step": 3391
    },
    {
      "epoch": 0.2376036495136025,
      "grad_norm": 4.3415141105651855,
      "learning_rate": 7.626194045534151e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 54570960,
      "step": 3392
    },
    {
      "epoch": 0.23767369775933175,
      "grad_norm": 3.770224094390869,
      "learning_rate": 7.625494220665499e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 54587344,
      "step": 3393
    },
    {
      "epoch": 0.23774374600506099,
      "grad_norm": 3.7803759574890137,
      "learning_rate": 7.624794395796848e-05,
      "loss": 1.1631,
      "num_input_tokens_seen": 54603728,
      "step": 3394
    },
    {
      "epoch": 0.23781379425079024,
      "grad_norm": 4.559312343597412,
      "learning_rate": 7.624094570928196e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 54619760,
      "step": 3395
    },
    {
      "epoch": 0.23788384249651948,
      "grad_norm": 4.215981483459473,
      "learning_rate": 7.623394746059545e-05,
      "loss": 1.2803,
      "num_input_tokens_seen": 54636144,
      "step": 3396
    },
    {
      "epoch": 0.23795389074224874,
      "grad_norm": 4.108291149139404,
      "learning_rate": 7.622694921190894e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 54652136,
      "step": 3397
    },
    {
      "epoch": 0.23802393898797797,
      "grad_norm": 4.4075093269348145,
      "learning_rate": 7.621995096322242e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 54668520,
      "step": 3398
    },
    {
      "epoch": 0.23809398723370723,
      "grad_norm": 4.002575874328613,
      "learning_rate": 7.62129527145359e-05,
      "loss": 1.1793,
      "num_input_tokens_seen": 54684544,
      "step": 3399
    },
    {
      "epoch": 0.23816403547943646,
      "grad_norm": 3.5264174938201904,
      "learning_rate": 7.620595446584938e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 54700680,
      "step": 3400
    },
    {
      "epoch": 0.23816403547943646,
      "eval_loss": 1.1361509561538696,
      "eval_runtime": 0.1856,
      "eval_samples_per_second": 5.389,
      "eval_steps_per_second": 5.389,
      "num_input_tokens_seen": 54700680,
      "step": 3400
    },
    {
      "epoch": 0.23823408372516572,
      "grad_norm": 3.585204839706421,
      "learning_rate": 7.619895621716287e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 54717064,
      "step": 3401
    },
    {
      "epoch": 0.23830413197089495,
      "grad_norm": 4.442777633666992,
      "learning_rate": 7.619195796847637e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 54732648,
      "step": 3402
    },
    {
      "epoch": 0.2383741802166242,
      "grad_norm": 3.807063102722168,
      "learning_rate": 7.618495971978985e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 54749032,
      "step": 3403
    },
    {
      "epoch": 0.23844422846235344,
      "grad_norm": 4.984583854675293,
      "learning_rate": 7.617796147110333e-05,
      "loss": 1.349,
      "num_input_tokens_seen": 54764192,
      "step": 3404
    },
    {
      "epoch": 0.2385142767080827,
      "grad_norm": 4.326750755310059,
      "learning_rate": 7.617096322241682e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 54780120,
      "step": 3405
    },
    {
      "epoch": 0.23858432495381193,
      "grad_norm": 5.707291126251221,
      "learning_rate": 7.61639649737303e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 54796168,
      "step": 3406
    },
    {
      "epoch": 0.2386543731995412,
      "grad_norm": 4.450499534606934,
      "learning_rate": 7.615696672504379e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 54812056,
      "step": 3407
    },
    {
      "epoch": 0.23872442144527042,
      "grad_norm": 4.253554821014404,
      "learning_rate": 7.614996847635728e-05,
      "loss": 1.1798,
      "num_input_tokens_seen": 54828248,
      "step": 3408
    },
    {
      "epoch": 0.23879446969099968,
      "grad_norm": 5.04890251159668,
      "learning_rate": 7.614297022767076e-05,
      "loss": 0.9968,
      "num_input_tokens_seen": 54844632,
      "step": 3409
    },
    {
      "epoch": 0.23886451793672892,
      "grad_norm": 3.24513578414917,
      "learning_rate": 7.613597197898424e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 54861016,
      "step": 3410
    },
    {
      "epoch": 0.23893456618245817,
      "grad_norm": 4.008625507354736,
      "learning_rate": 7.612897373029773e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 54877168,
      "step": 3411
    },
    {
      "epoch": 0.2390046144281874,
      "grad_norm": 5.393536567687988,
      "learning_rate": 7.612197548161122e-05,
      "loss": 1.1554,
      "num_input_tokens_seen": 54892720,
      "step": 3412
    },
    {
      "epoch": 0.23907466267391667,
      "grad_norm": 4.388333797454834,
      "learning_rate": 7.611497723292469e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 54909104,
      "step": 3413
    },
    {
      "epoch": 0.2391447109196459,
      "grad_norm": 3.8056883811950684,
      "learning_rate": 7.610797898423818e-05,
      "loss": 0.9235,
      "num_input_tokens_seen": 54925280,
      "step": 3414
    },
    {
      "epoch": 0.23921475916537516,
      "grad_norm": 6.9983062744140625,
      "learning_rate": 7.610098073555167e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 54941384,
      "step": 3415
    },
    {
      "epoch": 0.2392848074111044,
      "grad_norm": 3.485119581222534,
      "learning_rate": 7.609398248686516e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 54957592,
      "step": 3416
    },
    {
      "epoch": 0.23935485565683365,
      "grad_norm": 4.450938701629639,
      "learning_rate": 7.608698423817863e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 54973976,
      "step": 3417
    },
    {
      "epoch": 0.23942490390256288,
      "grad_norm": 4.142702579498291,
      "learning_rate": 7.607998598949212e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 54990360,
      "step": 3418
    },
    {
      "epoch": 0.23949495214829214,
      "grad_norm": 4.341495513916016,
      "learning_rate": 7.607298774080561e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 55006744,
      "step": 3419
    },
    {
      "epoch": 0.23956500039402137,
      "grad_norm": 4.355419158935547,
      "learning_rate": 7.606598949211908e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 55022816,
      "step": 3420
    },
    {
      "epoch": 0.23963504863975063,
      "grad_norm": 4.295046806335449,
      "learning_rate": 7.605899124343257e-05,
      "loss": 1.1881,
      "num_input_tokens_seen": 55039200,
      "step": 3421
    },
    {
      "epoch": 0.23970509688547986,
      "grad_norm": 3.9299042224884033,
      "learning_rate": 7.605199299474608e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 55055552,
      "step": 3422
    },
    {
      "epoch": 0.23977514513120912,
      "grad_norm": 3.7252607345581055,
      "learning_rate": 7.604499474605955e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 55071936,
      "step": 3423
    },
    {
      "epoch": 0.23984519337693835,
      "grad_norm": 4.723415851593018,
      "learning_rate": 7.603799649737304e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 55088320,
      "step": 3424
    },
    {
      "epoch": 0.2399152416226676,
      "grad_norm": 3.9923605918884277,
      "learning_rate": 7.603099824868651e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 55104416,
      "step": 3425
    },
    {
      "epoch": 0.23998528986839684,
      "grad_norm": 4.510697364807129,
      "learning_rate": 7.6024e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 55120800,
      "step": 3426
    },
    {
      "epoch": 0.2400553381141261,
      "grad_norm": 4.161818027496338,
      "learning_rate": 7.601700175131348e-05,
      "loss": 1.0915,
      "num_input_tokens_seen": 55137184,
      "step": 3427
    },
    {
      "epoch": 0.24012538635985536,
      "grad_norm": 5.871128082275391,
      "learning_rate": 7.601000350262698e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 55152528,
      "step": 3428
    },
    {
      "epoch": 0.2401954346055846,
      "grad_norm": 4.180598258972168,
      "learning_rate": 7.600300525394047e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 55168552,
      "step": 3429
    },
    {
      "epoch": 0.24026548285131386,
      "grad_norm": 5.575338363647461,
      "learning_rate": 7.599600700525394e-05,
      "loss": 1.2578,
      "num_input_tokens_seen": 55184104,
      "step": 3430
    },
    {
      "epoch": 0.2403355310970431,
      "grad_norm": 4.503122329711914,
      "learning_rate": 7.598900875656743e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 55199768,
      "step": 3431
    },
    {
      "epoch": 0.24040557934277235,
      "grad_norm": 3.6931769847869873,
      "learning_rate": 7.598201050788092e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 55216016,
      "step": 3432
    },
    {
      "epoch": 0.24047562758850158,
      "grad_norm": 4.138489723205566,
      "learning_rate": 7.59750122591944e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 55232400,
      "step": 3433
    },
    {
      "epoch": 0.24054567583423084,
      "grad_norm": 3.603297710418701,
      "learning_rate": 7.596801401050788e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 55248784,
      "step": 3434
    },
    {
      "epoch": 0.24061572407996007,
      "grad_norm": 4.072240352630615,
      "learning_rate": 7.596101576182137e-05,
      "loss": 1.3073,
      "num_input_tokens_seen": 55264320,
      "step": 3435
    },
    {
      "epoch": 0.24068577232568933,
      "grad_norm": 5.015305519104004,
      "learning_rate": 7.595401751313486e-05,
      "loss": 1.3236,
      "num_input_tokens_seen": 55280528,
      "step": 3436
    },
    {
      "epoch": 0.24075582057141856,
      "grad_norm": 5.135364055633545,
      "learning_rate": 7.594701926444834e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 55296912,
      "step": 3437
    },
    {
      "epoch": 0.24082586881714782,
      "grad_norm": 4.737668991088867,
      "learning_rate": 7.594002101576183e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 55313296,
      "step": 3438
    },
    {
      "epoch": 0.24089591706287705,
      "grad_norm": 4.380087375640869,
      "learning_rate": 7.593302276707531e-05,
      "loss": 1.267,
      "num_input_tokens_seen": 55329152,
      "step": 3439
    },
    {
      "epoch": 0.2409659653086063,
      "grad_norm": 4.472866535186768,
      "learning_rate": 7.592602451838879e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 55345536,
      "step": 3440
    },
    {
      "epoch": 0.24103601355433554,
      "grad_norm": 4.323402404785156,
      "learning_rate": 7.591902626970228e-05,
      "loss": 1.1872,
      "num_input_tokens_seen": 55361920,
      "step": 3441
    },
    {
      "epoch": 0.2411060618000648,
      "grad_norm": 3.7247276306152344,
      "learning_rate": 7.591202802101578e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 55377344,
      "step": 3442
    },
    {
      "epoch": 0.24117611004579403,
      "grad_norm": 6.503116607666016,
      "learning_rate": 7.590502977232925e-05,
      "loss": 1.2304,
      "num_input_tokens_seen": 55393728,
      "step": 3443
    },
    {
      "epoch": 0.2412461582915233,
      "grad_norm": 4.590184688568115,
      "learning_rate": 7.589803152364273e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 55410112,
      "step": 3444
    },
    {
      "epoch": 0.24131620653725253,
      "grad_norm": 3.718323230743408,
      "learning_rate": 7.589103327495622e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 55426392,
      "step": 3445
    },
    {
      "epoch": 0.24138625478298179,
      "grad_norm": 4.8696465492248535,
      "learning_rate": 7.58840350262697e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 55442632,
      "step": 3446
    },
    {
      "epoch": 0.24145630302871102,
      "grad_norm": 3.7620716094970703,
      "learning_rate": 7.587703677758318e-05,
      "loss": 0.9411,
      "num_input_tokens_seen": 55459016,
      "step": 3447
    },
    {
      "epoch": 0.24152635127444028,
      "grad_norm": 3.8696882724761963,
      "learning_rate": 7.587003852889668e-05,
      "loss": 0.992,
      "num_input_tokens_seen": 55474944,
      "step": 3448
    },
    {
      "epoch": 0.2415963995201695,
      "grad_norm": 4.628901481628418,
      "learning_rate": 7.586304028021017e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 55490416,
      "step": 3449
    },
    {
      "epoch": 0.24166644776589877,
      "grad_norm": 4.1568121910095215,
      "learning_rate": 7.585604203152365e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 55505760,
      "step": 3450
    },
    {
      "epoch": 0.241736496011628,
      "grad_norm": 4.089991569519043,
      "learning_rate": 7.584904378283714e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 55521528,
      "step": 3451
    },
    {
      "epoch": 0.24180654425735726,
      "grad_norm": 7.870755195617676,
      "learning_rate": 7.584204553415061e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 55536256,
      "step": 3452
    },
    {
      "epoch": 0.2418765925030865,
      "grad_norm": 4.898053169250488,
      "learning_rate": 7.58350472854641e-05,
      "loss": 1.0912,
      "num_input_tokens_seen": 55551904,
      "step": 3453
    },
    {
      "epoch": 0.24194664074881575,
      "grad_norm": 4.515797138214111,
      "learning_rate": 7.582804903677759e-05,
      "loss": 1.2266,
      "num_input_tokens_seen": 55567240,
      "step": 3454
    },
    {
      "epoch": 0.24201668899454498,
      "grad_norm": 3.7202370166778564,
      "learning_rate": 7.582105078809108e-05,
      "loss": 1.0118,
      "num_input_tokens_seen": 55583176,
      "step": 3455
    },
    {
      "epoch": 0.24208673724027424,
      "grad_norm": 5.834963321685791,
      "learning_rate": 7.581405253940457e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 55599144,
      "step": 3456
    },
    {
      "epoch": 0.24215678548600347,
      "grad_norm": 4.450705528259277,
      "learning_rate": 7.580705429071804e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 55615528,
      "step": 3457
    },
    {
      "epoch": 0.24222683373173273,
      "grad_norm": 3.796229839324951,
      "learning_rate": 7.580005604203153e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 55631912,
      "step": 3458
    },
    {
      "epoch": 0.24229688197746196,
      "grad_norm": 4.004448413848877,
      "learning_rate": 7.579305779334502e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 55647896,
      "step": 3459
    },
    {
      "epoch": 0.24236693022319122,
      "grad_norm": 4.511063098907471,
      "learning_rate": 7.578605954465849e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 55664280,
      "step": 3460
    },
    {
      "epoch": 0.24243697846892046,
      "grad_norm": 6.866496562957764,
      "learning_rate": 7.577906129597198e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 55679720,
      "step": 3461
    },
    {
      "epoch": 0.24250702671464971,
      "grad_norm": 5.447164058685303,
      "learning_rate": 7.577206304728547e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 55696104,
      "step": 3462
    },
    {
      "epoch": 0.24257707496037897,
      "grad_norm": 6.401725769042969,
      "learning_rate": 7.576506479859896e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 55712488,
      "step": 3463
    },
    {
      "epoch": 0.2426471232061082,
      "grad_norm": 3.7833733558654785,
      "learning_rate": 7.575806654991243e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 55728504,
      "step": 3464
    },
    {
      "epoch": 0.24271717145183747,
      "grad_norm": 4.157958984375,
      "learning_rate": 7.575106830122592e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 55743904,
      "step": 3465
    },
    {
      "epoch": 0.2427872196975667,
      "grad_norm": 4.657470703125,
      "learning_rate": 7.574407005253941e-05,
      "loss": 1.2033,
      "num_input_tokens_seen": 55759920,
      "step": 3466
    },
    {
      "epoch": 0.24285726794329596,
      "grad_norm": 5.129040718078613,
      "learning_rate": 7.573707180385289e-05,
      "loss": 1.2959,
      "num_input_tokens_seen": 55776304,
      "step": 3467
    },
    {
      "epoch": 0.2429273161890252,
      "grad_norm": 4.887351036071777,
      "learning_rate": 7.573007355516639e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 55792688,
      "step": 3468
    },
    {
      "epoch": 0.24299736443475445,
      "grad_norm": 4.042501926422119,
      "learning_rate": 7.572307530647988e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 55809072,
      "step": 3469
    },
    {
      "epoch": 0.24306741268048368,
      "grad_norm": 4.162355422973633,
      "learning_rate": 7.571607705779335e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 55825456,
      "step": 3470
    },
    {
      "epoch": 0.24313746092621294,
      "grad_norm": 3.8478844165802,
      "learning_rate": 7.570907880910683e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 55841840,
      "step": 3471
    },
    {
      "epoch": 0.24320750917194217,
      "grad_norm": 6.2065815925598145,
      "learning_rate": 7.570208056042032e-05,
      "loss": 1.3644,
      "num_input_tokens_seen": 55858224,
      "step": 3472
    },
    {
      "epoch": 0.24327755741767143,
      "grad_norm": 4.8233642578125,
      "learning_rate": 7.56950823117338e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 55874608,
      "step": 3473
    },
    {
      "epoch": 0.24334760566340066,
      "grad_norm": 3.534205198287964,
      "learning_rate": 7.568808406304729e-05,
      "loss": 1.0236,
      "num_input_tokens_seen": 55890992,
      "step": 3474
    },
    {
      "epoch": 0.24341765390912992,
      "grad_norm": 4.218345642089844,
      "learning_rate": 7.568108581436078e-05,
      "loss": 1.0921,
      "num_input_tokens_seen": 55907376,
      "step": 3475
    },
    {
      "epoch": 0.24348770215485915,
      "grad_norm": 3.7262325286865234,
      "learning_rate": 7.567408756567427e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 55923752,
      "step": 3476
    },
    {
      "epoch": 0.2435577504005884,
      "grad_norm": 3.7530906200408936,
      "learning_rate": 7.566708931698774e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 55939176,
      "step": 3477
    },
    {
      "epoch": 0.24362779864631764,
      "grad_norm": 4.452608585357666,
      "learning_rate": 7.566009106830123e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 55955200,
      "step": 3478
    },
    {
      "epoch": 0.2436978468920469,
      "grad_norm": 4.049906253814697,
      "learning_rate": 7.565309281961471e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 55971584,
      "step": 3479
    },
    {
      "epoch": 0.24376789513777614,
      "grad_norm": 4.956455230712891,
      "learning_rate": 7.56460945709282e-05,
      "loss": 1.2684,
      "num_input_tokens_seen": 55987968,
      "step": 3480
    },
    {
      "epoch": 0.2438379433835054,
      "grad_norm": 4.846863746643066,
      "learning_rate": 7.563909632224169e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 56003000,
      "step": 3481
    },
    {
      "epoch": 0.24390799162923463,
      "grad_norm": 4.678101539611816,
      "learning_rate": 7.563209807355517e-05,
      "loss": 0.8577,
      "num_input_tokens_seen": 56019384,
      "step": 3482
    },
    {
      "epoch": 0.2439780398749639,
      "grad_norm": 4.13012170791626,
      "learning_rate": 7.562509982486866e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 56035768,
      "step": 3483
    },
    {
      "epoch": 0.24404808812069312,
      "grad_norm": 3.7858669757843018,
      "learning_rate": 7.561810157618214e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 56052152,
      "step": 3484
    },
    {
      "epoch": 0.24411813636642238,
      "grad_norm": 3.7132198810577393,
      "learning_rate": 7.561110332749563e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 56068536,
      "step": 3485
    },
    {
      "epoch": 0.2441881846121516,
      "grad_norm": 4.093675136566162,
      "learning_rate": 7.560410507880911e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 56084888,
      "step": 3486
    },
    {
      "epoch": 0.24425823285788087,
      "grad_norm": 3.8601670265197754,
      "learning_rate": 7.559710683012259e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 56101272,
      "step": 3487
    },
    {
      "epoch": 0.2443282811036101,
      "grad_norm": 3.9332170486450195,
      "learning_rate": 7.559010858143608e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 56117352,
      "step": 3488
    },
    {
      "epoch": 0.24439832934933936,
      "grad_norm": 3.7619807720184326,
      "learning_rate": 7.558311033274957e-05,
      "loss": 1.0948,
      "num_input_tokens_seen": 56133736,
      "step": 3489
    },
    {
      "epoch": 0.2444683775950686,
      "grad_norm": 3.9035656452178955,
      "learning_rate": 7.557611208406306e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 56149624,
      "step": 3490
    },
    {
      "epoch": 0.24453842584079785,
      "grad_norm": 5.9505486488342285,
      "learning_rate": 7.556911383537653e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 56163752,
      "step": 3491
    },
    {
      "epoch": 0.24460847408652708,
      "grad_norm": 4.006525039672852,
      "learning_rate": 7.556211558669002e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 56180136,
      "step": 3492
    },
    {
      "epoch": 0.24467852233225634,
      "grad_norm": 5.28178071975708,
      "learning_rate": 7.555511733800351e-05,
      "loss": 1.3704,
      "num_input_tokens_seen": 56196152,
      "step": 3493
    },
    {
      "epoch": 0.24474857057798557,
      "grad_norm": 3.577709674835205,
      "learning_rate": 7.5548119089317e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 56212528,
      "step": 3494
    },
    {
      "epoch": 0.24481861882371483,
      "grad_norm": 4.6045002937316895,
      "learning_rate": 7.554112084063048e-05,
      "loss": 1.1895,
      "num_input_tokens_seen": 56228912,
      "step": 3495
    },
    {
      "epoch": 0.24488866706944407,
      "grad_norm": 4.160959720611572,
      "learning_rate": 7.553412259194397e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 56244944,
      "step": 3496
    },
    {
      "epoch": 0.24495871531517333,
      "grad_norm": 4.380669593811035,
      "learning_rate": 7.552712434325745e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 56261072,
      "step": 3497
    },
    {
      "epoch": 0.24502876356090258,
      "grad_norm": 3.568679094314575,
      "learning_rate": 7.552012609457092e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 56277456,
      "step": 3498
    },
    {
      "epoch": 0.24509881180663182,
      "grad_norm": 4.006386756896973,
      "learning_rate": 7.551312784588441e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 56293840,
      "step": 3499
    },
    {
      "epoch": 0.24516886005236108,
      "grad_norm": 4.180081844329834,
      "learning_rate": 7.55061295971979e-05,
      "loss": 1.303,
      "num_input_tokens_seen": 56310224,
      "step": 3500
    },
    {
      "epoch": 0.2452389082980903,
      "grad_norm": 5.228555202484131,
      "learning_rate": 7.549913134851139e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 56326608,
      "step": 3501
    },
    {
      "epoch": 0.24530895654381957,
      "grad_norm": 3.60235595703125,
      "learning_rate": 7.549213309982488e-05,
      "loss": 0.9258,
      "num_input_tokens_seen": 56342752,
      "step": 3502
    },
    {
      "epoch": 0.2453790047895488,
      "grad_norm": 5.305295467376709,
      "learning_rate": 7.548513485113837e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 56358208,
      "step": 3503
    },
    {
      "epoch": 0.24544905303527806,
      "grad_norm": 4.5955939292907715,
      "learning_rate": 7.547813660245184e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 56374592,
      "step": 3504
    },
    {
      "epoch": 0.2455191012810073,
      "grad_norm": 4.503798961639404,
      "learning_rate": 7.547113835376533e-05,
      "loss": 0.9045,
      "num_input_tokens_seen": 56390320,
      "step": 3505
    },
    {
      "epoch": 0.24558914952673655,
      "grad_norm": 3.79156231880188,
      "learning_rate": 7.54641401050788e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 56406176,
      "step": 3506
    },
    {
      "epoch": 0.24565919777246578,
      "grad_norm": 4.054116249084473,
      "learning_rate": 7.54571418563923e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 56421120,
      "step": 3507
    },
    {
      "epoch": 0.24572924601819504,
      "grad_norm": 3.57210636138916,
      "learning_rate": 7.545014360770578e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 56437504,
      "step": 3508
    },
    {
      "epoch": 0.24579929426392427,
      "grad_norm": 4.02463960647583,
      "learning_rate": 7.544314535901927e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 56452680,
      "step": 3509
    },
    {
      "epoch": 0.24586934250965353,
      "grad_norm": 3.712689161300659,
      "learning_rate": 7.543614711033276e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 56468424,
      "step": 3510
    },
    {
      "epoch": 0.24593939075538276,
      "grad_norm": 4.943066596984863,
      "learning_rate": 7.542914886164623e-05,
      "loss": 1.2289,
      "num_input_tokens_seen": 56484784,
      "step": 3511
    },
    {
      "epoch": 0.24600943900111202,
      "grad_norm": 4.94294548034668,
      "learning_rate": 7.542215061295972e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 56500992,
      "step": 3512
    },
    {
      "epoch": 0.24607948724684126,
      "grad_norm": 4.003958225250244,
      "learning_rate": 7.541515236427321e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 56516392,
      "step": 3513
    },
    {
      "epoch": 0.24614953549257051,
      "grad_norm": 4.173887252807617,
      "learning_rate": 7.540815411558669e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 56532776,
      "step": 3514
    },
    {
      "epoch": 0.24621958373829975,
      "grad_norm": 5.028538227081299,
      "learning_rate": 7.540115586690019e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 56549056,
      "step": 3515
    },
    {
      "epoch": 0.246289631984029,
      "grad_norm": 3.781219959259033,
      "learning_rate": 7.539415761821366e-05,
      "loss": 0.9056,
      "num_input_tokens_seen": 56564768,
      "step": 3516
    },
    {
      "epoch": 0.24635968022975824,
      "grad_norm": 4.070143222808838,
      "learning_rate": 7.538715936952715e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 56580856,
      "step": 3517
    },
    {
      "epoch": 0.2464297284754875,
      "grad_norm": 4.322885036468506,
      "learning_rate": 7.538016112084063e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 56596664,
      "step": 3518
    },
    {
      "epoch": 0.24649977672121673,
      "grad_norm": 7.525569438934326,
      "learning_rate": 7.537316287215412e-05,
      "loss": 1.278,
      "num_input_tokens_seen": 56611608,
      "step": 3519
    },
    {
      "epoch": 0.246569824966946,
      "grad_norm": 3.6520745754241943,
      "learning_rate": 7.53661646234676e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 56627992,
      "step": 3520
    },
    {
      "epoch": 0.24663987321267522,
      "grad_norm": 6.346038341522217,
      "learning_rate": 7.53591663747811e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 56644224,
      "step": 3521
    },
    {
      "epoch": 0.24670992145840448,
      "grad_norm": 5.049849987030029,
      "learning_rate": 7.535216812609458e-05,
      "loss": 1.1843,
      "num_input_tokens_seen": 56660464,
      "step": 3522
    },
    {
      "epoch": 0.2467799697041337,
      "grad_norm": 5.948208332061768,
      "learning_rate": 7.534516987740807e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 56676848,
      "step": 3523
    },
    {
      "epoch": 0.24685001794986297,
      "grad_norm": 4.2648210525512695,
      "learning_rate": 7.533817162872155e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 56692800,
      "step": 3524
    },
    {
      "epoch": 0.2469200661955922,
      "grad_norm": 4.268098831176758,
      "learning_rate": 7.533117338003502e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 56709184,
      "step": 3525
    },
    {
      "epoch": 0.24699011444132146,
      "grad_norm": 3.987408399581909,
      "learning_rate": 7.532417513134851e-05,
      "loss": 1.1312,
      "num_input_tokens_seen": 56724744,
      "step": 3526
    },
    {
      "epoch": 0.2470601626870507,
      "grad_norm": 4.087530612945557,
      "learning_rate": 7.5317176882662e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 56741128,
      "step": 3527
    },
    {
      "epoch": 0.24713021093277995,
      "grad_norm": 3.484837770462036,
      "learning_rate": 7.531017863397549e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 56757512,
      "step": 3528
    },
    {
      "epoch": 0.24720025917850919,
      "grad_norm": 4.382214546203613,
      "learning_rate": 7.530318038528898e-05,
      "loss": 1.1713,
      "num_input_tokens_seen": 56773896,
      "step": 3529
    },
    {
      "epoch": 0.24727030742423844,
      "grad_norm": 4.361959457397461,
      "learning_rate": 7.529618213660246e-05,
      "loss": 1.2548,
      "num_input_tokens_seen": 56789184,
      "step": 3530
    },
    {
      "epoch": 0.24734035566996768,
      "grad_norm": 4.029329776763916,
      "learning_rate": 7.528918388791594e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 56804912,
      "step": 3531
    },
    {
      "epoch": 0.24741040391569694,
      "grad_norm": 4.577064514160156,
      "learning_rate": 7.528218563922943e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 56821296,
      "step": 3532
    },
    {
      "epoch": 0.2474804521614262,
      "grad_norm": 3.6799368858337402,
      "learning_rate": 7.52751873905429e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 56837680,
      "step": 3533
    },
    {
      "epoch": 0.24755050040715543,
      "grad_norm": 3.962989568710327,
      "learning_rate": 7.526818914185639e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 56853984,
      "step": 3534
    },
    {
      "epoch": 0.2476205486528847,
      "grad_norm": 3.610877275466919,
      "learning_rate": 7.526119089316988e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 56870368,
      "step": 3535
    },
    {
      "epoch": 0.24769059689861392,
      "grad_norm": 4.16568660736084,
      "learning_rate": 7.525419264448337e-05,
      "loss": 0.9326,
      "num_input_tokens_seen": 56886032,
      "step": 3536
    },
    {
      "epoch": 0.24776064514434318,
      "grad_norm": 4.645627021789551,
      "learning_rate": 7.524719439579686e-05,
      "loss": 1.0221,
      "num_input_tokens_seen": 56900928,
      "step": 3537
    },
    {
      "epoch": 0.2478306933900724,
      "grad_norm": 3.652317523956299,
      "learning_rate": 7.524019614711033e-05,
      "loss": 1.1641,
      "num_input_tokens_seen": 56917048,
      "step": 3538
    },
    {
      "epoch": 0.24790074163580167,
      "grad_norm": 5.583502769470215,
      "learning_rate": 7.523319789842382e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 56933432,
      "step": 3539
    },
    {
      "epoch": 0.2479707898815309,
      "grad_norm": 3.6924219131469727,
      "learning_rate": 7.522619964973731e-05,
      "loss": 0.9386,
      "num_input_tokens_seen": 56949816,
      "step": 3540
    },
    {
      "epoch": 0.24804083812726016,
      "grad_norm": 6.476202487945557,
      "learning_rate": 7.52192014010508e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 56966064,
      "step": 3541
    },
    {
      "epoch": 0.2481108863729894,
      "grad_norm": 4.052863121032715,
      "learning_rate": 7.521220315236429e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 56982448,
      "step": 3542
    },
    {
      "epoch": 0.24818093461871865,
      "grad_norm": 6.569397926330566,
      "learning_rate": 7.520520490367776e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 56998832,
      "step": 3543
    },
    {
      "epoch": 0.24825098286444788,
      "grad_norm": 4.026796817779541,
      "learning_rate": 7.519820665499125e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 57014744,
      "step": 3544
    },
    {
      "epoch": 0.24832103111017714,
      "grad_norm": 3.705080032348633,
      "learning_rate": 7.519120840630472e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 57031128,
      "step": 3545
    },
    {
      "epoch": 0.24839107935590637,
      "grad_norm": 4.828441143035889,
      "learning_rate": 7.518421015761821e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 57047512,
      "step": 3546
    },
    {
      "epoch": 0.24846112760163563,
      "grad_norm": 5.6117777824401855,
      "learning_rate": 7.51772119089317e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 57063840,
      "step": 3547
    },
    {
      "epoch": 0.24853117584736487,
      "grad_norm": 3.955193281173706,
      "learning_rate": 7.517021366024519e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 57079936,
      "step": 3548
    },
    {
      "epoch": 0.24860122409309413,
      "grad_norm": 3.8878116607666016,
      "learning_rate": 7.516321541155868e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 57096320,
      "step": 3549
    },
    {
      "epoch": 0.24867127233882336,
      "grad_norm": 6.119873046875,
      "learning_rate": 7.515621716287217e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 57111632,
      "step": 3550
    },
    {
      "epoch": 0.24874132058455262,
      "grad_norm": 3.757984161376953,
      "learning_rate": 7.514921891418564e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 57128016,
      "step": 3551
    },
    {
      "epoch": 0.24881136883028185,
      "grad_norm": 4.173069953918457,
      "learning_rate": 7.514222066549912e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 57144400,
      "step": 3552
    },
    {
      "epoch": 0.2488814170760111,
      "grad_norm": 3.6985576152801514,
      "learning_rate": 7.51352224168126e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 57160784,
      "step": 3553
    },
    {
      "epoch": 0.24895146532174034,
      "grad_norm": 4.464880466461182,
      "learning_rate": 7.51282241681261e-05,
      "loss": 1.2571,
      "num_input_tokens_seen": 57177168,
      "step": 3554
    },
    {
      "epoch": 0.2490215135674696,
      "grad_norm": 3.699873447418213,
      "learning_rate": 7.512122591943958e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 57193208,
      "step": 3555
    },
    {
      "epoch": 0.24909156181319883,
      "grad_norm": 5.011424541473389,
      "learning_rate": 7.511422767075307e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 57209592,
      "step": 3556
    },
    {
      "epoch": 0.2491616100589281,
      "grad_norm": 4.302685260772705,
      "learning_rate": 7.510722942206656e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 57225976,
      "step": 3557
    },
    {
      "epoch": 0.24923165830465732,
      "grad_norm": 3.991840362548828,
      "learning_rate": 7.510023117338004e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 57242168,
      "step": 3558
    },
    {
      "epoch": 0.24930170655038658,
      "grad_norm": 3.9910435676574707,
      "learning_rate": 7.509323292469352e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 57258552,
      "step": 3559
    },
    {
      "epoch": 0.2493717547961158,
      "grad_norm": 4.170960426330566,
      "learning_rate": 7.5086234676007e-05,
      "loss": 1.0068,
      "num_input_tokens_seen": 57274936,
      "step": 3560
    },
    {
      "epoch": 0.24944180304184507,
      "grad_norm": 4.317671298980713,
      "learning_rate": 7.50792364273205e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 57291320,
      "step": 3561
    },
    {
      "epoch": 0.2495118512875743,
      "grad_norm": 3.871293783187866,
      "learning_rate": 7.507223817863398e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 57307056,
      "step": 3562
    },
    {
      "epoch": 0.24958189953330356,
      "grad_norm": 4.014804840087891,
      "learning_rate": 7.506523992994747e-05,
      "loss": 1.2893,
      "num_input_tokens_seen": 57322376,
      "step": 3563
    },
    {
      "epoch": 0.2496519477790328,
      "grad_norm": 5.13847017288208,
      "learning_rate": 7.505824168126095e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 57338760,
      "step": 3564
    },
    {
      "epoch": 0.24972199602476206,
      "grad_norm": 3.7801826000213623,
      "learning_rate": 7.505124343257443e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 57355144,
      "step": 3565
    },
    {
      "epoch": 0.24979204427049131,
      "grad_norm": 3.662065029144287,
      "learning_rate": 7.504424518388792e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 57371528,
      "step": 3566
    },
    {
      "epoch": 0.24986209251622055,
      "grad_norm": 4.548840522766113,
      "learning_rate": 7.50372469352014e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 57387912,
      "step": 3567
    },
    {
      "epoch": 0.2499321407619498,
      "grad_norm": 4.732056140899658,
      "learning_rate": 7.50302486865149e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 57403120,
      "step": 3568
    },
    {
      "epoch": 0.25000218900767907,
      "grad_norm": 3.7986674308776855,
      "learning_rate": 7.502325043782837e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 57418800,
      "step": 3569
    },
    {
      "epoch": 0.25007223725340827,
      "grad_norm": 6.760079860687256,
      "learning_rate": 7.501625218914186e-05,
      "loss": 0.9101,
      "num_input_tokens_seen": 57432608,
      "step": 3570
    },
    {
      "epoch": 0.25014228549913753,
      "grad_norm": 4.0666985511779785,
      "learning_rate": 7.500925394045535e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 57448296,
      "step": 3571
    },
    {
      "epoch": 0.2502123337448668,
      "grad_norm": 3.7505650520324707,
      "learning_rate": 7.500225569176882e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 57464680,
      "step": 3572
    },
    {
      "epoch": 0.25028238199059605,
      "grad_norm": 5.1084675788879395,
      "learning_rate": 7.499525744308231e-05,
      "loss": 1.317,
      "num_input_tokens_seen": 57481032,
      "step": 3573
    },
    {
      "epoch": 0.25035243023632525,
      "grad_norm": 6.083080768585205,
      "learning_rate": 7.49882591943958e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 57497416,
      "step": 3574
    },
    {
      "epoch": 0.2504224784820545,
      "grad_norm": 4.197649955749512,
      "learning_rate": 7.498126094570929e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 57513800,
      "step": 3575
    },
    {
      "epoch": 0.25049252672778377,
      "grad_norm": 4.637972831726074,
      "learning_rate": 7.497426269702278e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 57529832,
      "step": 3576
    },
    {
      "epoch": 0.25056257497351303,
      "grad_norm": 4.096358776092529,
      "learning_rate": 7.496726444833626e-05,
      "loss": 1.1909,
      "num_input_tokens_seen": 57545432,
      "step": 3577
    },
    {
      "epoch": 0.25063262321924223,
      "grad_norm": 3.9253315925598145,
      "learning_rate": 7.496026619964974e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 57561816,
      "step": 3578
    },
    {
      "epoch": 0.2507026714649715,
      "grad_norm": 5.603836536407471,
      "learning_rate": 7.495326795096321e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 57577336,
      "step": 3579
    },
    {
      "epoch": 0.25077271971070075,
      "grad_norm": 4.588653564453125,
      "learning_rate": 7.49462697022767e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 57593720,
      "step": 3580
    },
    {
      "epoch": 0.25084276795643,
      "grad_norm": 3.989229917526245,
      "learning_rate": 7.49392714535902e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 57609656,
      "step": 3581
    },
    {
      "epoch": 0.2509128162021592,
      "grad_norm": 4.728183269500732,
      "learning_rate": 7.493227320490368e-05,
      "loss": 1.2626,
      "num_input_tokens_seen": 57626040,
      "step": 3582
    },
    {
      "epoch": 0.2509828644478885,
      "grad_norm": 4.269988059997559,
      "learning_rate": 7.492527495621717e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 57641280,
      "step": 3583
    },
    {
      "epoch": 0.25105291269361774,
      "grad_norm": 6.506377696990967,
      "learning_rate": 7.491827670753066e-05,
      "loss": 0.9327,
      "num_input_tokens_seen": 57657664,
      "step": 3584
    },
    {
      "epoch": 0.251122960939347,
      "grad_norm": 6.415282726287842,
      "learning_rate": 7.491127845884413e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 57672704,
      "step": 3585
    },
    {
      "epoch": 0.2511930091850762,
      "grad_norm": 3.969257116317749,
      "learning_rate": 7.490428021015761e-05,
      "loss": 1.1255,
      "num_input_tokens_seen": 57687504,
      "step": 3586
    },
    {
      "epoch": 0.25126305743080546,
      "grad_norm": 3.493469476699829,
      "learning_rate": 7.489728196147111e-05,
      "loss": 0.95,
      "num_input_tokens_seen": 57703512,
      "step": 3587
    },
    {
      "epoch": 0.2513331056765347,
      "grad_norm": 5.777353763580322,
      "learning_rate": 7.48902837127846e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 57719344,
      "step": 3588
    },
    {
      "epoch": 0.251403153922264,
      "grad_norm": 3.6840991973876953,
      "learning_rate": 7.488328546409807e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 57734848,
      "step": 3589
    },
    {
      "epoch": 0.2514732021679932,
      "grad_norm": 6.526551246643066,
      "learning_rate": 7.487628721541156e-05,
      "loss": 1.1651,
      "num_input_tokens_seen": 57751232,
      "step": 3590
    },
    {
      "epoch": 0.25154325041372244,
      "grad_norm": 3.7879719734191895,
      "learning_rate": 7.486928896672505e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 57767616,
      "step": 3591
    },
    {
      "epoch": 0.2516132986594517,
      "grad_norm": 6.891875267028809,
      "learning_rate": 7.486229071803853e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 57783592,
      "step": 3592
    },
    {
      "epoch": 0.25168334690518096,
      "grad_norm": 4.700318336486816,
      "learning_rate": 7.485529246935201e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 57799976,
      "step": 3593
    },
    {
      "epoch": 0.25175339515091016,
      "grad_norm": 6.47390604019165,
      "learning_rate": 7.48482942206655e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 57816360,
      "step": 3594
    },
    {
      "epoch": 0.2518234433966394,
      "grad_norm": 5.045449733734131,
      "learning_rate": 7.484129597197899e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 57832016,
      "step": 3595
    },
    {
      "epoch": 0.2518934916423687,
      "grad_norm": 4.258456230163574,
      "learning_rate": 7.483429772329247e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 57848400,
      "step": 3596
    },
    {
      "epoch": 0.25196353988809794,
      "grad_norm": 3.948582649230957,
      "learning_rate": 7.482729947460596e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 57864784,
      "step": 3597
    },
    {
      "epoch": 0.25203358813382715,
      "grad_norm": 4.017141342163086,
      "learning_rate": 7.482030122591944e-05,
      "loss": 0.8644,
      "num_input_tokens_seen": 57879696,
      "step": 3598
    },
    {
      "epoch": 0.2521036363795564,
      "grad_norm": 3.7428297996520996,
      "learning_rate": 7.481330297723292e-05,
      "loss": 0.9318,
      "num_input_tokens_seen": 57896080,
      "step": 3599
    },
    {
      "epoch": 0.25217368462528567,
      "grad_norm": 4.883368968963623,
      "learning_rate": 7.480630472854641e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 57911976,
      "step": 3600
    },
    {
      "epoch": 0.25217368462528567,
      "eval_loss": 1.136000633239746,
      "eval_runtime": 0.2016,
      "eval_samples_per_second": 4.959,
      "eval_steps_per_second": 4.959,
      "num_input_tokens_seen": 57911976,
      "step": 3600
    },
    {
      "epoch": 0.2522437328710149,
      "grad_norm": 4.399716377258301,
      "learning_rate": 7.479930647985991e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 57927440,
      "step": 3601
    },
    {
      "epoch": 0.2523137811167442,
      "grad_norm": 6.019199371337891,
      "learning_rate": 7.479230823117338e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 57943824,
      "step": 3602
    },
    {
      "epoch": 0.2523838293624734,
      "grad_norm": 4.42507266998291,
      "learning_rate": 7.478530998248687e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 57960208,
      "step": 3603
    },
    {
      "epoch": 0.25245387760820265,
      "grad_norm": 4.0232062339782715,
      "learning_rate": 7.477831173380036e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 57976560,
      "step": 3604
    },
    {
      "epoch": 0.2525239258539319,
      "grad_norm": 3.6392862796783447,
      "learning_rate": 7.477131348511384e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 57992944,
      "step": 3605
    },
    {
      "epoch": 0.25259397409966117,
      "grad_norm": 3.849912643432617,
      "learning_rate": 7.476431523642731e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 58009328,
      "step": 3606
    },
    {
      "epoch": 0.25266402234539037,
      "grad_norm": 3.5331156253814697,
      "learning_rate": 7.475731698774081e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 58025152,
      "step": 3607
    },
    {
      "epoch": 0.25273407059111963,
      "grad_norm": 4.343970775604248,
      "learning_rate": 7.47503187390543e-05,
      "loss": 1.0231,
      "num_input_tokens_seen": 58041536,
      "step": 3608
    },
    {
      "epoch": 0.2528041188368489,
      "grad_norm": 3.6736862659454346,
      "learning_rate": 7.474332049036778e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 58057920,
      "step": 3609
    },
    {
      "epoch": 0.25287416708257815,
      "grad_norm": 6.599121570587158,
      "learning_rate": 7.473632224168127e-05,
      "loss": 1.2235,
      "num_input_tokens_seen": 58073784,
      "step": 3610
    },
    {
      "epoch": 0.25294421532830735,
      "grad_norm": 4.2448930740356445,
      "learning_rate": 7.472932399299475e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 58088776,
      "step": 3611
    },
    {
      "epoch": 0.2530142635740366,
      "grad_norm": 3.416584014892578,
      "learning_rate": 7.472232574430823e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 58105160,
      "step": 3612
    },
    {
      "epoch": 0.2530843118197659,
      "grad_norm": 3.9348700046539307,
      "learning_rate": 7.471532749562172e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 58121528,
      "step": 3613
    },
    {
      "epoch": 0.25315436006549513,
      "grad_norm": 6.208236217498779,
      "learning_rate": 7.470832924693521e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 58137912,
      "step": 3614
    },
    {
      "epoch": 0.25322440831122434,
      "grad_norm": 3.9069888591766357,
      "learning_rate": 7.47013309982487e-05,
      "loss": 0.9958,
      "num_input_tokens_seen": 58154056,
      "step": 3615
    },
    {
      "epoch": 0.2532944565569536,
      "grad_norm": 4.482925891876221,
      "learning_rate": 7.469433274956217e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 58168904,
      "step": 3616
    },
    {
      "epoch": 0.25336450480268286,
      "grad_norm": 4.082488536834717,
      "learning_rate": 7.468733450087566e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 58185288,
      "step": 3617
    },
    {
      "epoch": 0.2534345530484121,
      "grad_norm": 5.994426250457764,
      "learning_rate": 7.468033625218915e-05,
      "loss": 1.1286,
      "num_input_tokens_seen": 58201600,
      "step": 3618
    },
    {
      "epoch": 0.2535046012941413,
      "grad_norm": 3.966487169265747,
      "learning_rate": 7.467333800350262e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 58217752,
      "step": 3619
    },
    {
      "epoch": 0.2535746495398706,
      "grad_norm": 4.3370537757873535,
      "learning_rate": 7.466633975481611e-05,
      "loss": 0.9495,
      "num_input_tokens_seen": 58233672,
      "step": 3620
    },
    {
      "epoch": 0.25364469778559984,
      "grad_norm": 4.638936519622803,
      "learning_rate": 7.465934150612961e-05,
      "loss": 1.1593,
      "num_input_tokens_seen": 58249904,
      "step": 3621
    },
    {
      "epoch": 0.2537147460313291,
      "grad_norm": 3.42993426322937,
      "learning_rate": 7.465234325744309e-05,
      "loss": 0.9112,
      "num_input_tokens_seen": 58265272,
      "step": 3622
    },
    {
      "epoch": 0.2537847942770583,
      "grad_norm": 4.637670516967773,
      "learning_rate": 7.464534500875656e-05,
      "loss": 1.1578,
      "num_input_tokens_seen": 58281656,
      "step": 3623
    },
    {
      "epoch": 0.25385484252278756,
      "grad_norm": 4.470972061157227,
      "learning_rate": 7.463834676007005e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 58297696,
      "step": 3624
    },
    {
      "epoch": 0.2539248907685168,
      "grad_norm": 4.158536434173584,
      "learning_rate": 7.463134851138354e-05,
      "loss": 1.2625,
      "num_input_tokens_seen": 58313960,
      "step": 3625
    },
    {
      "epoch": 0.2539949390142461,
      "grad_norm": 5.2940850257873535,
      "learning_rate": 7.462435026269702e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 58329928,
      "step": 3626
    },
    {
      "epoch": 0.2540649872599753,
      "grad_norm": 4.270470142364502,
      "learning_rate": 7.461735201401052e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 58345544,
      "step": 3627
    },
    {
      "epoch": 0.25413503550570454,
      "grad_norm": 4.488008975982666,
      "learning_rate": 7.4610353765324e-05,
      "loss": 1.2652,
      "num_input_tokens_seen": 58361736,
      "step": 3628
    },
    {
      "epoch": 0.2542050837514338,
      "grad_norm": 3.9760642051696777,
      "learning_rate": 7.460335551663748e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 58377888,
      "step": 3629
    },
    {
      "epoch": 0.25427513199716306,
      "grad_norm": 4.022678852081299,
      "learning_rate": 7.459635726795097e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 58393744,
      "step": 3630
    },
    {
      "epoch": 0.25434518024289227,
      "grad_norm": 6.345690727233887,
      "learning_rate": 7.458935901926446e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 58410064,
      "step": 3631
    },
    {
      "epoch": 0.2544152284886215,
      "grad_norm": 4.0159101486206055,
      "learning_rate": 7.458236077057793e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 58426352,
      "step": 3632
    },
    {
      "epoch": 0.2544852767343508,
      "grad_norm": 4.125208854675293,
      "learning_rate": 7.457536252189142e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 58441936,
      "step": 3633
    },
    {
      "epoch": 0.25455532498008004,
      "grad_norm": 4.429535865783691,
      "learning_rate": 7.456836427320491e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 58457136,
      "step": 3634
    },
    {
      "epoch": 0.2546253732258093,
      "grad_norm": 3.655606269836426,
      "learning_rate": 7.45613660245184e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 58473520,
      "step": 3635
    },
    {
      "epoch": 0.2546954214715385,
      "grad_norm": 3.688188314437866,
      "learning_rate": 7.455436777583187e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 58489904,
      "step": 3636
    },
    {
      "epoch": 0.25476546971726777,
      "grad_norm": 3.95440411567688,
      "learning_rate": 7.454736952714536e-05,
      "loss": 1.2586,
      "num_input_tokens_seen": 58506032,
      "step": 3637
    },
    {
      "epoch": 0.254835517962997,
      "grad_norm": 3.950641632080078,
      "learning_rate": 7.454037127845885e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 58521464,
      "step": 3638
    },
    {
      "epoch": 0.2549055662087263,
      "grad_norm": 4.9607038497924805,
      "learning_rate": 7.453337302977233e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 58537848,
      "step": 3639
    },
    {
      "epoch": 0.2549756144544555,
      "grad_norm": 3.4168713092803955,
      "learning_rate": 7.452637478108582e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 58554232,
      "step": 3640
    },
    {
      "epoch": 0.25504566270018475,
      "grad_norm": 6.897549152374268,
      "learning_rate": 7.451937653239932e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 58570616,
      "step": 3641
    },
    {
      "epoch": 0.255115710945914,
      "grad_norm": 4.009060859680176,
      "learning_rate": 7.451237828371279e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 58587000,
      "step": 3642
    },
    {
      "epoch": 0.25518575919164327,
      "grad_norm": 4.245255470275879,
      "learning_rate": 7.450538003502627e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 58602768,
      "step": 3643
    },
    {
      "epoch": 0.2552558074373725,
      "grad_norm": 3.7547385692596436,
      "learning_rate": 7.449838178633976e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 58619024,
      "step": 3644
    },
    {
      "epoch": 0.25532585568310173,
      "grad_norm": 5.7543745040893555,
      "learning_rate": 7.449138353765324e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 58635408,
      "step": 3645
    },
    {
      "epoch": 0.255395903928831,
      "grad_norm": 3.8786420822143555,
      "learning_rate": 7.448438528896672e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 58651392,
      "step": 3646
    },
    {
      "epoch": 0.25546595217456025,
      "grad_norm": 4.290858745574951,
      "learning_rate": 7.447738704028022e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 58667712,
      "step": 3647
    },
    {
      "epoch": 0.25553600042028946,
      "grad_norm": 3.8005576133728027,
      "learning_rate": 7.447038879159371e-05,
      "loss": 1.1709,
      "num_input_tokens_seen": 58683512,
      "step": 3648
    },
    {
      "epoch": 0.2556060486660187,
      "grad_norm": 3.574735403060913,
      "learning_rate": 7.446339054290719e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 58699296,
      "step": 3649
    },
    {
      "epoch": 0.255676096911748,
      "grad_norm": 4.487549304962158,
      "learning_rate": 7.445639229422066e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 58715680,
      "step": 3650
    },
    {
      "epoch": 0.25574614515747723,
      "grad_norm": 3.80549955368042,
      "learning_rate": 7.444939404553415e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 58732064,
      "step": 3651
    },
    {
      "epoch": 0.25581619340320644,
      "grad_norm": 6.745276927947998,
      "learning_rate": 7.444239579684764e-05,
      "loss": 0.9649,
      "num_input_tokens_seen": 58748416,
      "step": 3652
    },
    {
      "epoch": 0.2558862416489357,
      "grad_norm": 5.366410732269287,
      "learning_rate": 7.443539754816113e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 58764800,
      "step": 3653
    },
    {
      "epoch": 0.25595628989466496,
      "grad_norm": 4.889951705932617,
      "learning_rate": 7.442839929947462e-05,
      "loss": 1.0447,
      "num_input_tokens_seen": 58779776,
      "step": 3654
    },
    {
      "epoch": 0.2560263381403942,
      "grad_norm": 3.776078462600708,
      "learning_rate": 7.44214010507881e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 58796160,
      "step": 3655
    },
    {
      "epoch": 0.2560963863861234,
      "grad_norm": 4.999850749969482,
      "learning_rate": 7.441440280210158e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 58812544,
      "step": 3656
    },
    {
      "epoch": 0.2561664346318527,
      "grad_norm": 4.111214637756348,
      "learning_rate": 7.440740455341507e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 58828696,
      "step": 3657
    },
    {
      "epoch": 0.25623648287758194,
      "grad_norm": 4.49043083190918,
      "learning_rate": 7.440040630472856e-05,
      "loss": 0.8889,
      "num_input_tokens_seen": 58845080,
      "step": 3658
    },
    {
      "epoch": 0.2563065311233112,
      "grad_norm": 4.440788745880127,
      "learning_rate": 7.439340805604203e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 58861464,
      "step": 3659
    },
    {
      "epoch": 0.2563765793690404,
      "grad_norm": 5.642586708068848,
      "learning_rate": 7.438640980735552e-05,
      "loss": 1.3676,
      "num_input_tokens_seen": 58877624,
      "step": 3660
    },
    {
      "epoch": 0.25644662761476966,
      "grad_norm": 3.8768467903137207,
      "learning_rate": 7.437941155866901e-05,
      "loss": 0.9737,
      "num_input_tokens_seen": 58894008,
      "step": 3661
    },
    {
      "epoch": 0.2565166758604989,
      "grad_norm": 3.9855473041534424,
      "learning_rate": 7.43724133099825e-05,
      "loss": 1.0987,
      "num_input_tokens_seen": 58909600,
      "step": 3662
    },
    {
      "epoch": 0.2565867241062282,
      "grad_norm": 3.6692938804626465,
      "learning_rate": 7.436541506129597e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 58925776,
      "step": 3663
    },
    {
      "epoch": 0.2566567723519574,
      "grad_norm": 3.87776517868042,
      "learning_rate": 7.435841681260946e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 58941048,
      "step": 3664
    },
    {
      "epoch": 0.25672682059768664,
      "grad_norm": 3.5173263549804688,
      "learning_rate": 7.435141856392295e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 58957432,
      "step": 3665
    },
    {
      "epoch": 0.2567968688434159,
      "grad_norm": 4.312611103057861,
      "learning_rate": 7.434442031523642e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 58973816,
      "step": 3666
    },
    {
      "epoch": 0.25686691708914516,
      "grad_norm": 3.7889907360076904,
      "learning_rate": 7.433742206654991e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 58989472,
      "step": 3667
    },
    {
      "epoch": 0.25693696533487437,
      "grad_norm": 4.997755527496338,
      "learning_rate": 7.433042381786341e-05,
      "loss": 1.2147,
      "num_input_tokens_seen": 59005856,
      "step": 3668
    },
    {
      "epoch": 0.2570070135806036,
      "grad_norm": 5.839511871337891,
      "learning_rate": 7.432342556917689e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 59022176,
      "step": 3669
    },
    {
      "epoch": 0.2570770618263329,
      "grad_norm": 4.185897350311279,
      "learning_rate": 7.431642732049036e-05,
      "loss": 1.0769,
      "num_input_tokens_seen": 59038296,
      "step": 3670
    },
    {
      "epoch": 0.25714711007206215,
      "grad_norm": 3.6666383743286133,
      "learning_rate": 7.430942907180385e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 59054680,
      "step": 3671
    },
    {
      "epoch": 0.2572171583177914,
      "grad_norm": 3.8587453365325928,
      "learning_rate": 7.430243082311734e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 59070912,
      "step": 3672
    },
    {
      "epoch": 0.2572872065635206,
      "grad_norm": 3.6518352031707764,
      "learning_rate": 7.429543257443083e-05,
      "loss": 1.042,
      "num_input_tokens_seen": 59087296,
      "step": 3673
    },
    {
      "epoch": 0.25735725480924987,
      "grad_norm": 4.629798412322998,
      "learning_rate": 7.428843432574432e-05,
      "loss": 1.2649,
      "num_input_tokens_seen": 59103632,
      "step": 3674
    },
    {
      "epoch": 0.25742730305497913,
      "grad_norm": 6.353034496307373,
      "learning_rate": 7.428143607705781e-05,
      "loss": 1.3823,
      "num_input_tokens_seen": 59120016,
      "step": 3675
    },
    {
      "epoch": 0.2574973513007084,
      "grad_norm": 6.1848273277282715,
      "learning_rate": 7.427443782837128e-05,
      "loss": 1.2275,
      "num_input_tokens_seen": 59136232,
      "step": 3676
    },
    {
      "epoch": 0.2575673995464376,
      "grad_norm": 3.6022186279296875,
      "learning_rate": 7.426743957968476e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 59152616,
      "step": 3677
    },
    {
      "epoch": 0.25763744779216685,
      "grad_norm": 3.6495468616485596,
      "learning_rate": 7.426044133099825e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 59167792,
      "step": 3678
    },
    {
      "epoch": 0.2577074960378961,
      "grad_norm": 4.675189018249512,
      "learning_rate": 7.425344308231174e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 59184176,
      "step": 3679
    },
    {
      "epoch": 0.25777754428362537,
      "grad_norm": 3.657700538635254,
      "learning_rate": 7.424644483362522e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 59199632,
      "step": 3680
    },
    {
      "epoch": 0.2578475925293546,
      "grad_norm": 3.9934394359588623,
      "learning_rate": 7.423944658493871e-05,
      "loss": 1.0598,
      "num_input_tokens_seen": 59215720,
      "step": 3681
    },
    {
      "epoch": 0.25791764077508383,
      "grad_norm": 3.777191400527954,
      "learning_rate": 7.42324483362522e-05,
      "loss": 1.2255,
      "num_input_tokens_seen": 59231248,
      "step": 3682
    },
    {
      "epoch": 0.2579876890208131,
      "grad_norm": 3.9812276363372803,
      "learning_rate": 7.422545008756568e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 59247280,
      "step": 3683
    },
    {
      "epoch": 0.25805773726654235,
      "grad_norm": 3.631455183029175,
      "learning_rate": 7.421845183887916e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 59263664,
      "step": 3684
    },
    {
      "epoch": 0.25812778551227156,
      "grad_norm": 3.803898334503174,
      "learning_rate": 7.421145359019265e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 59279880,
      "step": 3685
    },
    {
      "epoch": 0.2581978337580008,
      "grad_norm": 3.649956703186035,
      "learning_rate": 7.420445534150613e-05,
      "loss": 0.951,
      "num_input_tokens_seen": 59296216,
      "step": 3686
    },
    {
      "epoch": 0.2582678820037301,
      "grad_norm": 4.010924339294434,
      "learning_rate": 7.419745709281962e-05,
      "loss": 1.1987,
      "num_input_tokens_seen": 59312448,
      "step": 3687
    },
    {
      "epoch": 0.25833793024945934,
      "grad_norm": 4.2410759925842285,
      "learning_rate": 7.41904588441331e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 59328456,
      "step": 3688
    },
    {
      "epoch": 0.25840797849518854,
      "grad_norm": 3.9170684814453125,
      "learning_rate": 7.41834605954466e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 59344840,
      "step": 3689
    },
    {
      "epoch": 0.2584780267409178,
      "grad_norm": 3.935624837875366,
      "learning_rate": 7.417646234676007e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 59360040,
      "step": 3690
    },
    {
      "epoch": 0.25854807498664706,
      "grad_norm": 4.3747782707214355,
      "learning_rate": 7.416946409807356e-05,
      "loss": 1.2462,
      "num_input_tokens_seen": 59375896,
      "step": 3691
    },
    {
      "epoch": 0.2586181232323763,
      "grad_norm": 7.553433418273926,
      "learning_rate": 7.416246584938705e-05,
      "loss": 1.4753,
      "num_input_tokens_seen": 59391144,
      "step": 3692
    },
    {
      "epoch": 0.2586881714781055,
      "grad_norm": 3.4443981647491455,
      "learning_rate": 7.415546760070053e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 59407528,
      "step": 3693
    },
    {
      "epoch": 0.2587582197238348,
      "grad_norm": 4.02165412902832,
      "learning_rate": 7.414846935201401e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 59422824,
      "step": 3694
    },
    {
      "epoch": 0.25882826796956404,
      "grad_norm": 3.8880200386047363,
      "learning_rate": 7.414147110332751e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 59439208,
      "step": 3695
    },
    {
      "epoch": 0.2588983162152933,
      "grad_norm": 5.463441371917725,
      "learning_rate": 7.413447285464099e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 59455592,
      "step": 3696
    },
    {
      "epoch": 0.2589683644610225,
      "grad_norm": 7.555225372314453,
      "learning_rate": 7.412747460595446e-05,
      "loss": 1.2278,
      "num_input_tokens_seen": 59471976,
      "step": 3697
    },
    {
      "epoch": 0.25903841270675176,
      "grad_norm": 5.7154436111450195,
      "learning_rate": 7.412047635726795e-05,
      "loss": 1.151,
      "num_input_tokens_seen": 59488360,
      "step": 3698
    },
    {
      "epoch": 0.259108460952481,
      "grad_norm": 5.09559965133667,
      "learning_rate": 7.411347810858144e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 59504536,
      "step": 3699
    },
    {
      "epoch": 0.2591785091982103,
      "grad_norm": 4.7749738693237305,
      "learning_rate": 7.410647985989493e-05,
      "loss": 1.2971,
      "num_input_tokens_seen": 59520488,
      "step": 3700
    },
    {
      "epoch": 0.2592485574439395,
      "grad_norm": 4.323631763458252,
      "learning_rate": 7.409948161120842e-05,
      "loss": 1.1687,
      "num_input_tokens_seen": 59535384,
      "step": 3701
    },
    {
      "epoch": 0.25931860568966875,
      "grad_norm": 3.511822462081909,
      "learning_rate": 7.40924833625219e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 59550888,
      "step": 3702
    },
    {
      "epoch": 0.259388653935398,
      "grad_norm": 4.039402008056641,
      "learning_rate": 7.408548511383538e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 59567184,
      "step": 3703
    },
    {
      "epoch": 0.25945870218112727,
      "grad_norm": 3.6692605018615723,
      "learning_rate": 7.407848686514885e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 59583568,
      "step": 3704
    },
    {
      "epoch": 0.2595287504268565,
      "grad_norm": 4.414707660675049,
      "learning_rate": 7.407148861646234e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 59599088,
      "step": 3705
    },
    {
      "epoch": 0.25959879867258573,
      "grad_norm": 4.073670387268066,
      "learning_rate": 7.406449036777583e-05,
      "loss": 1.2958,
      "num_input_tokens_seen": 59615432,
      "step": 3706
    },
    {
      "epoch": 0.259668846918315,
      "grad_norm": 4.436419486999512,
      "learning_rate": 7.405749211908932e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 59631816,
      "step": 3707
    },
    {
      "epoch": 0.25973889516404425,
      "grad_norm": 5.866218090057373,
      "learning_rate": 7.405049387040281e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 59648200,
      "step": 3708
    },
    {
      "epoch": 0.2598089434097735,
      "grad_norm": 4.133188247680664,
      "learning_rate": 7.40434956217163e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 59664584,
      "step": 3709
    },
    {
      "epoch": 0.2598789916555027,
      "grad_norm": 4.1976213455200195,
      "learning_rate": 7.403649737302977e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 59680288,
      "step": 3710
    },
    {
      "epoch": 0.25994903990123197,
      "grad_norm": 3.990983009338379,
      "learning_rate": 7.402949912434326e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 59696408,
      "step": 3711
    },
    {
      "epoch": 0.26001908814696123,
      "grad_norm": 4.427793025970459,
      "learning_rate": 7.402250087565675e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 59712792,
      "step": 3712
    },
    {
      "epoch": 0.2600891363926905,
      "grad_norm": 5.360867023468018,
      "learning_rate": 7.401550262697024e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 59728968,
      "step": 3713
    },
    {
      "epoch": 0.2601591846384197,
      "grad_norm": 3.8442916870117188,
      "learning_rate": 7.400850437828371e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 59745352,
      "step": 3714
    },
    {
      "epoch": 0.26022923288414895,
      "grad_norm": 3.7610833644866943,
      "learning_rate": 7.40015061295972e-05,
      "loss": 0.9969,
      "num_input_tokens_seen": 59761736,
      "step": 3715
    },
    {
      "epoch": 0.2602992811298782,
      "grad_norm": 3.9050705432891846,
      "learning_rate": 7.399450788091069e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 59778000,
      "step": 3716
    },
    {
      "epoch": 0.2603693293756075,
      "grad_norm": 4.293839454650879,
      "learning_rate": 7.398750963222417e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 59794216,
      "step": 3717
    },
    {
      "epoch": 0.2604393776213367,
      "grad_norm": 3.7403993606567383,
      "learning_rate": 7.398051138353765e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 59810600,
      "step": 3718
    },
    {
      "epoch": 0.26050942586706594,
      "grad_norm": 5.266970157623291,
      "learning_rate": 7.397351313485114e-05,
      "loss": 0.8695,
      "num_input_tokens_seen": 59826984,
      "step": 3719
    },
    {
      "epoch": 0.2605794741127952,
      "grad_norm": 4.385645866394043,
      "learning_rate": 7.396651488616463e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 59843368,
      "step": 3720
    },
    {
      "epoch": 0.26064952235852445,
      "grad_norm": 4.349147796630859,
      "learning_rate": 7.39595166374781e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 59859136,
      "step": 3721
    },
    {
      "epoch": 0.26071957060425366,
      "grad_norm": 4.69277286529541,
      "learning_rate": 7.395251838879161e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 59875024,
      "step": 3722
    },
    {
      "epoch": 0.2607896188499829,
      "grad_norm": 3.602949857711792,
      "learning_rate": 7.394552014010508e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 59891408,
      "step": 3723
    },
    {
      "epoch": 0.2608596670957122,
      "grad_norm": 4.137026786804199,
      "learning_rate": 7.393852189141856e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 59906360,
      "step": 3724
    },
    {
      "epoch": 0.26092971534144144,
      "grad_norm": 4.558672904968262,
      "learning_rate": 7.393152364273205e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 59922744,
      "step": 3725
    },
    {
      "epoch": 0.26099976358717064,
      "grad_norm": 3.977217197418213,
      "learning_rate": 7.392452539404554e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 59938448,
      "step": 3726
    },
    {
      "epoch": 0.2610698118328999,
      "grad_norm": 6.573578834533691,
      "learning_rate": 7.391752714535902e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 59954832,
      "step": 3727
    },
    {
      "epoch": 0.26113986007862916,
      "grad_norm": 4.253365516662598,
      "learning_rate": 7.391052889667251e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 59971216,
      "step": 3728
    },
    {
      "epoch": 0.2612099083243584,
      "grad_norm": 4.279355525970459,
      "learning_rate": 7.3903530647986e-05,
      "loss": 1.0456,
      "num_input_tokens_seen": 59987384,
      "step": 3729
    },
    {
      "epoch": 0.2612799565700876,
      "grad_norm": 5.5035505294799805,
      "learning_rate": 7.389653239929948e-05,
      "loss": 1.236,
      "num_input_tokens_seen": 60003720,
      "step": 3730
    },
    {
      "epoch": 0.2613500048158169,
      "grad_norm": 5.064812660217285,
      "learning_rate": 7.388953415061295e-05,
      "loss": 0.8739,
      "num_input_tokens_seen": 60020104,
      "step": 3731
    },
    {
      "epoch": 0.26142005306154614,
      "grad_norm": 4.716748237609863,
      "learning_rate": 7.388253590192644e-05,
      "loss": 1.2417,
      "num_input_tokens_seen": 60036488,
      "step": 3732
    },
    {
      "epoch": 0.2614901013072754,
      "grad_norm": 4.0947489738464355,
      "learning_rate": 7.387553765323994e-05,
      "loss": 1.1332,
      "num_input_tokens_seen": 60052384,
      "step": 3733
    },
    {
      "epoch": 0.2615601495530046,
      "grad_norm": 3.757126808166504,
      "learning_rate": 7.386853940455342e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 60068624,
      "step": 3734
    },
    {
      "epoch": 0.26163019779873387,
      "grad_norm": 7.364987850189209,
      "learning_rate": 7.38615411558669e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 60084248,
      "step": 3735
    },
    {
      "epoch": 0.2617002460444631,
      "grad_norm": 4.630516052246094,
      "learning_rate": 7.38545429071804e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 60100632,
      "step": 3736
    },
    {
      "epoch": 0.2617702942901924,
      "grad_norm": 5.3436760902404785,
      "learning_rate": 7.384754465849387e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 60116672,
      "step": 3737
    },
    {
      "epoch": 0.2618403425359216,
      "grad_norm": 3.843344211578369,
      "learning_rate": 7.384054640980736e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 60133056,
      "step": 3738
    },
    {
      "epoch": 0.26191039078165085,
      "grad_norm": 4.561652183532715,
      "learning_rate": 7.383354816112085e-05,
      "loss": 1.2304,
      "num_input_tokens_seen": 60149440,
      "step": 3739
    },
    {
      "epoch": 0.2619804390273801,
      "grad_norm": 3.951719045639038,
      "learning_rate": 7.382654991243434e-05,
      "loss": 0.8449,
      "num_input_tokens_seen": 60165824,
      "step": 3740
    },
    {
      "epoch": 0.26205048727310937,
      "grad_norm": 3.702449321746826,
      "learning_rate": 7.381955166374781e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 60181496,
      "step": 3741
    },
    {
      "epoch": 0.2621205355188386,
      "grad_norm": 5.43525505065918,
      "learning_rate": 7.38125534150613e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 60197040,
      "step": 3742
    },
    {
      "epoch": 0.26219058376456783,
      "grad_norm": 3.9709503650665283,
      "learning_rate": 7.380555516637479e-05,
      "loss": 1.1172,
      "num_input_tokens_seen": 60213424,
      "step": 3743
    },
    {
      "epoch": 0.2622606320102971,
      "grad_norm": 3.7183797359466553,
      "learning_rate": 7.379855691768826e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 60229696,
      "step": 3744
    },
    {
      "epoch": 0.26233068025602635,
      "grad_norm": 3.933479070663452,
      "learning_rate": 7.379155866900175e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 60246080,
      "step": 3745
    },
    {
      "epoch": 0.2624007285017556,
      "grad_norm": 4.837695598602295,
      "learning_rate": 7.378456042031524e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 60262464,
      "step": 3746
    },
    {
      "epoch": 0.2624707767474848,
      "grad_norm": 4.791194438934326,
      "learning_rate": 7.377756217162873e-05,
      "loss": 1.2467,
      "num_input_tokens_seen": 60278600,
      "step": 3747
    },
    {
      "epoch": 0.2625408249932141,
      "grad_norm": 4.53259801864624,
      "learning_rate": 7.37705639229422e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 60293856,
      "step": 3748
    },
    {
      "epoch": 0.26261087323894333,
      "grad_norm": 3.87522554397583,
      "learning_rate": 7.37635656742557e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 60309888,
      "step": 3749
    },
    {
      "epoch": 0.2626809214846726,
      "grad_norm": 4.46868896484375,
      "learning_rate": 7.375656742556918e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 60325784,
      "step": 3750
    },
    {
      "epoch": 0.2627509697304018,
      "grad_norm": 3.938703775405884,
      "learning_rate": 7.374956917688266e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 60340696,
      "step": 3751
    },
    {
      "epoch": 0.26282101797613105,
      "grad_norm": 3.5147759914398193,
      "learning_rate": 7.374257092819614e-05,
      "loss": 1.0984,
      "num_input_tokens_seen": 60357080,
      "step": 3752
    },
    {
      "epoch": 0.2628910662218603,
      "grad_norm": 4.008304119110107,
      "learning_rate": 7.373557267950965e-05,
      "loss": 1.046,
      "num_input_tokens_seen": 60373464,
      "step": 3753
    },
    {
      "epoch": 0.2629611144675896,
      "grad_norm": 3.9318859577178955,
      "learning_rate": 7.372857443082312e-05,
      "loss": 1.001,
      "num_input_tokens_seen": 60389848,
      "step": 3754
    },
    {
      "epoch": 0.2630311627133188,
      "grad_norm": 4.046808242797852,
      "learning_rate": 7.372157618213661e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 60406232,
      "step": 3755
    },
    {
      "epoch": 0.26310121095904804,
      "grad_norm": 5.451204299926758,
      "learning_rate": 7.37145779334501e-05,
      "loss": 0.9567,
      "num_input_tokens_seen": 60422544,
      "step": 3756
    },
    {
      "epoch": 0.2631712592047773,
      "grad_norm": 4.395990371704102,
      "learning_rate": 7.370757968476357e-05,
      "loss": 0.9173,
      "num_input_tokens_seen": 60438312,
      "step": 3757
    },
    {
      "epoch": 0.26324130745050656,
      "grad_norm": 5.997600078582764,
      "learning_rate": 7.370058143607705e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 60454696,
      "step": 3758
    },
    {
      "epoch": 0.26331135569623576,
      "grad_norm": 5.588560104370117,
      "learning_rate": 7.369358318739055e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 60470232,
      "step": 3759
    },
    {
      "epoch": 0.263381403941965,
      "grad_norm": 3.2995078563690186,
      "learning_rate": 7.368658493870404e-05,
      "loss": 0.9814,
      "num_input_tokens_seen": 60486224,
      "step": 3760
    },
    {
      "epoch": 0.2634514521876943,
      "grad_norm": 4.141932964324951,
      "learning_rate": 7.367958669001751e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 60502608,
      "step": 3761
    },
    {
      "epoch": 0.26352150043342354,
      "grad_norm": 5.010983943939209,
      "learning_rate": 7.3672588441331e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 60518672,
      "step": 3762
    },
    {
      "epoch": 0.26359154867915274,
      "grad_norm": 3.555612802505493,
      "learning_rate": 7.366559019264449e-05,
      "loss": 1.1037,
      "num_input_tokens_seen": 60534408,
      "step": 3763
    },
    {
      "epoch": 0.263661596924882,
      "grad_norm": 4.006901264190674,
      "learning_rate": 7.365859194395797e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 60550760,
      "step": 3764
    },
    {
      "epoch": 0.26373164517061126,
      "grad_norm": 5.055272579193115,
      "learning_rate": 7.365159369527146e-05,
      "loss": 0.9645,
      "num_input_tokens_seen": 60567144,
      "step": 3765
    },
    {
      "epoch": 0.2638016934163405,
      "grad_norm": 3.860630989074707,
      "learning_rate": 7.364459544658494e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 60583528,
      "step": 3766
    },
    {
      "epoch": 0.2638717416620697,
      "grad_norm": 4.644535541534424,
      "learning_rate": 7.363759719789843e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 60599912,
      "step": 3767
    },
    {
      "epoch": 0.263941789907799,
      "grad_norm": 3.7196872234344482,
      "learning_rate": 7.363059894921191e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 60616296,
      "step": 3768
    },
    {
      "epoch": 0.26401183815352824,
      "grad_norm": 4.477166175842285,
      "learning_rate": 7.36236007005254e-05,
      "loss": 1.2221,
      "num_input_tokens_seen": 60631760,
      "step": 3769
    },
    {
      "epoch": 0.2640818863992575,
      "grad_norm": 4.906933784484863,
      "learning_rate": 7.361660245183889e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 60648144,
      "step": 3770
    },
    {
      "epoch": 0.2641519346449867,
      "grad_norm": 3.784450054168701,
      "learning_rate": 7.360960420315236e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 60664528,
      "step": 3771
    },
    {
      "epoch": 0.26422198289071597,
      "grad_norm": 4.5654191970825195,
      "learning_rate": 7.360260595446585e-05,
      "loss": 0.9199,
      "num_input_tokens_seen": 60680912,
      "step": 3772
    },
    {
      "epoch": 0.2642920311364452,
      "grad_norm": 3.965175151824951,
      "learning_rate": 7.359560770577934e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 60697296,
      "step": 3773
    },
    {
      "epoch": 0.2643620793821745,
      "grad_norm": 5.112542152404785,
      "learning_rate": 7.358860945709283e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 60713328,
      "step": 3774
    },
    {
      "epoch": 0.26443212762790375,
      "grad_norm": 3.8610634803771973,
      "learning_rate": 7.35816112084063e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 60729712,
      "step": 3775
    },
    {
      "epoch": 0.26450217587363295,
      "grad_norm": 3.794217348098755,
      "learning_rate": 7.35746129597198e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 60745824,
      "step": 3776
    },
    {
      "epoch": 0.2645722241193622,
      "grad_norm": 3.7547152042388916,
      "learning_rate": 7.356761471103328e-05,
      "loss": 1.0774,
      "num_input_tokens_seen": 60762024,
      "step": 3777
    },
    {
      "epoch": 0.26464227236509147,
      "grad_norm": 3.492917537689209,
      "learning_rate": 7.356061646234675e-05,
      "loss": 1.0505,
      "num_input_tokens_seen": 60778096,
      "step": 3778
    },
    {
      "epoch": 0.26471232061082073,
      "grad_norm": 3.856019973754883,
      "learning_rate": 7.355361821366026e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 60794480,
      "step": 3779
    },
    {
      "epoch": 0.26478236885654993,
      "grad_norm": 3.68072509765625,
      "learning_rate": 7.354661996497374e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 60810584,
      "step": 3780
    },
    {
      "epoch": 0.2648524171022792,
      "grad_norm": 4.4739909172058105,
      "learning_rate": 7.353962171628722e-05,
      "loss": 1.2492,
      "num_input_tokens_seen": 60826240,
      "step": 3781
    },
    {
      "epoch": 0.26492246534800845,
      "grad_norm": 5.2342610359191895,
      "learning_rate": 7.353262346760071e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 60842216,
      "step": 3782
    },
    {
      "epoch": 0.2649925135937377,
      "grad_norm": 4.408970355987549,
      "learning_rate": 7.35256252189142e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 60857336,
      "step": 3783
    },
    {
      "epoch": 0.2650625618394669,
      "grad_norm": 3.8172199726104736,
      "learning_rate": 7.351862697022767e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 60873720,
      "step": 3784
    },
    {
      "epoch": 0.2651326100851962,
      "grad_norm": 4.250039100646973,
      "learning_rate": 7.351162872154116e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 60890104,
      "step": 3785
    },
    {
      "epoch": 0.26520265833092543,
      "grad_norm": 4.257120609283447,
      "learning_rate": 7.350463047285465e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 60906488,
      "step": 3786
    },
    {
      "epoch": 0.2652727065766547,
      "grad_norm": 4.205286026000977,
      "learning_rate": 7.349763222416814e-05,
      "loss": 1.2149,
      "num_input_tokens_seen": 60922872,
      "step": 3787
    },
    {
      "epoch": 0.2653427548223839,
      "grad_norm": 4.304909706115723,
      "learning_rate": 7.349063397548161e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 60939256,
      "step": 3788
    },
    {
      "epoch": 0.26541280306811316,
      "grad_norm": 4.793664455413818,
      "learning_rate": 7.34836357267951e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 60955440,
      "step": 3789
    },
    {
      "epoch": 0.2654828513138424,
      "grad_norm": 4.383579730987549,
      "learning_rate": 7.347663747810859e-05,
      "loss": 1.1924,
      "num_input_tokens_seen": 60971824,
      "step": 3790
    },
    {
      "epoch": 0.2655528995595717,
      "grad_norm": 3.9962210655212402,
      "learning_rate": 7.346963922942206e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 60987168,
      "step": 3791
    },
    {
      "epoch": 0.2656229478053009,
      "grad_norm": 4.356331825256348,
      "learning_rate": 7.346264098073555e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 61002840,
      "step": 3792
    },
    {
      "epoch": 0.26569299605103014,
      "grad_norm": 5.836807727813721,
      "learning_rate": 7.345564273204904e-05,
      "loss": 1.205,
      "num_input_tokens_seen": 61019224,
      "step": 3793
    },
    {
      "epoch": 0.2657630442967594,
      "grad_norm": 4.778296947479248,
      "learning_rate": 7.344864448336253e-05,
      "loss": 1.0227,
      "num_input_tokens_seen": 61034712,
      "step": 3794
    },
    {
      "epoch": 0.26583309254248866,
      "grad_norm": 6.723006248474121,
      "learning_rate": 7.3441646234676e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 61050328,
      "step": 3795
    },
    {
      "epoch": 0.26590314078821786,
      "grad_norm": 3.773984670639038,
      "learning_rate": 7.34346479859895e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 61066048,
      "step": 3796
    },
    {
      "epoch": 0.2659731890339471,
      "grad_norm": 3.915708065032959,
      "learning_rate": 7.342764973730298e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 61082136,
      "step": 3797
    },
    {
      "epoch": 0.2660432372796764,
      "grad_norm": 6.568943977355957,
      "learning_rate": 7.342065148861646e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 61097216,
      "step": 3798
    },
    {
      "epoch": 0.26611328552540564,
      "grad_norm": 5.0017499923706055,
      "learning_rate": 7.341365323992995e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 61112344,
      "step": 3799
    },
    {
      "epoch": 0.26618333377113484,
      "grad_norm": 4.1988935470581055,
      "learning_rate": 7.340665499124345e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 61128728,
      "step": 3800
    },
    {
      "epoch": 0.26618333377113484,
      "eval_loss": 1.1352765560150146,
      "eval_runtime": 0.2173,
      "eval_samples_per_second": 4.603,
      "eval_steps_per_second": 4.603,
      "num_input_tokens_seen": 61128728,
      "step": 3800
    },
    {
      "epoch": 0.2662533820168641,
      "grad_norm": 3.991041660308838,
      "learning_rate": 7.339965674255692e-05,
      "loss": 1.1468,
      "num_input_tokens_seen": 61145112,
      "step": 3801
    },
    {
      "epoch": 0.26632343026259336,
      "grad_norm": 4.921470642089844,
      "learning_rate": 7.33926584938704e-05,
      "loss": 1.1756,
      "num_input_tokens_seen": 61160952,
      "step": 3802
    },
    {
      "epoch": 0.2663934785083226,
      "grad_norm": 3.835486888885498,
      "learning_rate": 7.33856602451839e-05,
      "loss": 0.782,
      "num_input_tokens_seen": 61177024,
      "step": 3803
    },
    {
      "epoch": 0.2664635267540518,
      "grad_norm": 4.419501304626465,
      "learning_rate": 7.337866199649738e-05,
      "loss": 1.0029,
      "num_input_tokens_seen": 61193408,
      "step": 3804
    },
    {
      "epoch": 0.2665335749997811,
      "grad_norm": 4.003963947296143,
      "learning_rate": 7.337166374781086e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 61209792,
      "step": 3805
    },
    {
      "epoch": 0.26660362324551035,
      "grad_norm": 4.115198612213135,
      "learning_rate": 7.336466549912435e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 61226176,
      "step": 3806
    },
    {
      "epoch": 0.2666736714912396,
      "grad_norm": 3.663464307785034,
      "learning_rate": 7.335766725043784e-05,
      "loss": 1.1447,
      "num_input_tokens_seen": 61242560,
      "step": 3807
    },
    {
      "epoch": 0.2667437197369688,
      "grad_norm": 3.7513012886047363,
      "learning_rate": 7.335066900175132e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 61258944,
      "step": 3808
    },
    {
      "epoch": 0.26681376798269807,
      "grad_norm": 4.693987846374512,
      "learning_rate": 7.33436707530648e-05,
      "loss": 1.2823,
      "num_input_tokens_seen": 61275048,
      "step": 3809
    },
    {
      "epoch": 0.26688381622842733,
      "grad_norm": 6.161116600036621,
      "learning_rate": 7.333667250437829e-05,
      "loss": 1.1606,
      "num_input_tokens_seen": 61291368,
      "step": 3810
    },
    {
      "epoch": 0.2669538644741566,
      "grad_norm": 5.942180633544922,
      "learning_rate": 7.332967425569177e-05,
      "loss": 1.2382,
      "num_input_tokens_seen": 61307680,
      "step": 3811
    },
    {
      "epoch": 0.26702391271988585,
      "grad_norm": 4.940249443054199,
      "learning_rate": 7.332267600700526e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 61324064,
      "step": 3812
    },
    {
      "epoch": 0.26709396096561505,
      "grad_norm": 5.384439468383789,
      "learning_rate": 7.331567775831875e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 61340416,
      "step": 3813
    },
    {
      "epoch": 0.2671640092113443,
      "grad_norm": 5.4137959480285645,
      "learning_rate": 7.330867950963223e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 61356800,
      "step": 3814
    },
    {
      "epoch": 0.26723405745707357,
      "grad_norm": 5.492247581481934,
      "learning_rate": 7.330168126094571e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 61371736,
      "step": 3815
    },
    {
      "epoch": 0.26730410570280283,
      "grad_norm": 5.316330909729004,
      "learning_rate": 7.32946830122592e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 61388120,
      "step": 3816
    },
    {
      "epoch": 0.26737415394853203,
      "grad_norm": 3.976797103881836,
      "learning_rate": 7.328768476357269e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 61403672,
      "step": 3817
    },
    {
      "epoch": 0.2674442021942613,
      "grad_norm": 7.333898544311523,
      "learning_rate": 7.328068651488616e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 61420056,
      "step": 3818
    },
    {
      "epoch": 0.26751425043999055,
      "grad_norm": 3.795746088027954,
      "learning_rate": 7.327368826619965e-05,
      "loss": 1.0545,
      "num_input_tokens_seen": 61436440,
      "step": 3819
    },
    {
      "epoch": 0.2675842986857198,
      "grad_norm": 6.624248027801514,
      "learning_rate": 7.326669001751315e-05,
      "loss": 1.0736,
      "num_input_tokens_seen": 61452824,
      "step": 3820
    },
    {
      "epoch": 0.267654346931449,
      "grad_norm": 4.991429805755615,
      "learning_rate": 7.325969176882663e-05,
      "loss": 1.0681,
      "num_input_tokens_seen": 61469208,
      "step": 3821
    },
    {
      "epoch": 0.2677243951771783,
      "grad_norm": 3.8505215644836426,
      "learning_rate": 7.32526935201401e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 61485592,
      "step": 3822
    },
    {
      "epoch": 0.26779444342290754,
      "grad_norm": 3.7079288959503174,
      "learning_rate": 7.324569527145359e-05,
      "loss": 1.049,
      "num_input_tokens_seen": 61501976,
      "step": 3823
    },
    {
      "epoch": 0.2678644916686368,
      "grad_norm": 3.8987131118774414,
      "learning_rate": 7.323869702276708e-05,
      "loss": 1.0152,
      "num_input_tokens_seen": 61518360,
      "step": 3824
    },
    {
      "epoch": 0.267934539914366,
      "grad_norm": 4.0447516441345215,
      "learning_rate": 7.323169877408055e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 61534744,
      "step": 3825
    },
    {
      "epoch": 0.26800458816009526,
      "grad_norm": 4.089504241943359,
      "learning_rate": 7.322470052539406e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 61551128,
      "step": 3826
    },
    {
      "epoch": 0.2680746364058245,
      "grad_norm": 3.864943265914917,
      "learning_rate": 7.321770227670754e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 61566872,
      "step": 3827
    },
    {
      "epoch": 0.2681446846515538,
      "grad_norm": 4.649239540100098,
      "learning_rate": 7.321070402802102e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 61582704,
      "step": 3828
    },
    {
      "epoch": 0.268214732897283,
      "grad_norm": 7.537643909454346,
      "learning_rate": 7.32037057793345e-05,
      "loss": 1.289,
      "num_input_tokens_seen": 61599088,
      "step": 3829
    },
    {
      "epoch": 0.26828478114301224,
      "grad_norm": 3.312519073486328,
      "learning_rate": 7.3196707530648e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 61615472,
      "step": 3830
    },
    {
      "epoch": 0.2683548293887415,
      "grad_norm": 7.833526134490967,
      "learning_rate": 7.318970928196147e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 61631288,
      "step": 3831
    },
    {
      "epoch": 0.26842487763447076,
      "grad_norm": 3.9574341773986816,
      "learning_rate": 7.318271103327496e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 61646400,
      "step": 3832
    },
    {
      "epoch": 0.26849492588019996,
      "grad_norm": 3.8763623237609863,
      "learning_rate": 7.317571278458845e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 61662784,
      "step": 3833
    },
    {
      "epoch": 0.2685649741259292,
      "grad_norm": 4.006046295166016,
      "learning_rate": 7.316871453590194e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 61678296,
      "step": 3834
    },
    {
      "epoch": 0.2686350223716585,
      "grad_norm": 4.0256500244140625,
      "learning_rate": 7.316171628721541e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 61694680,
      "step": 3835
    },
    {
      "epoch": 0.26870507061738774,
      "grad_norm": 4.045619964599609,
      "learning_rate": 7.31547180385289e-05,
      "loss": 1.0445,
      "num_input_tokens_seen": 61711064,
      "step": 3836
    },
    {
      "epoch": 0.26877511886311695,
      "grad_norm": 4.189207553863525,
      "learning_rate": 7.314771978984239e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 61727448,
      "step": 3837
    },
    {
      "epoch": 0.2688451671088462,
      "grad_norm": 6.098819255828857,
      "learning_rate": 7.314072154115587e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 61743600,
      "step": 3838
    },
    {
      "epoch": 0.26891521535457547,
      "grad_norm": 3.832962989807129,
      "learning_rate": 7.313372329246935e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 61759984,
      "step": 3839
    },
    {
      "epoch": 0.2689852636003047,
      "grad_norm": 4.448224067687988,
      "learning_rate": 7.312672504378284e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 61776368,
      "step": 3840
    },
    {
      "epoch": 0.26905531184603393,
      "grad_norm": 4.621326446533203,
      "learning_rate": 7.311972679509633e-05,
      "loss": 0.9866,
      "num_input_tokens_seen": 61791992,
      "step": 3841
    },
    {
      "epoch": 0.2691253600917632,
      "grad_norm": 4.979477882385254,
      "learning_rate": 7.31127285464098e-05,
      "loss": 1.1592,
      "num_input_tokens_seen": 61807912,
      "step": 3842
    },
    {
      "epoch": 0.26919540833749245,
      "grad_norm": 4.678060054779053,
      "learning_rate": 7.31057302977233e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 61824296,
      "step": 3843
    },
    {
      "epoch": 0.2692654565832217,
      "grad_norm": 5.379042625427246,
      "learning_rate": 7.309873204903678e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 61840680,
      "step": 3844
    },
    {
      "epoch": 0.26933550482895097,
      "grad_norm": 5.836205005645752,
      "learning_rate": 7.309173380035026e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 61856296,
      "step": 3845
    },
    {
      "epoch": 0.26940555307468017,
      "grad_norm": 4.040728569030762,
      "learning_rate": 7.308473555166376e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 61872680,
      "step": 3846
    },
    {
      "epoch": 0.26947560132040943,
      "grad_norm": 5.207007884979248,
      "learning_rate": 7.307773730297725e-05,
      "loss": 0.9293,
      "num_input_tokens_seen": 61889064,
      "step": 3847
    },
    {
      "epoch": 0.2695456495661387,
      "grad_norm": 4.996053695678711,
      "learning_rate": 7.307073905429072e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 61905448,
      "step": 3848
    },
    {
      "epoch": 0.26961569781186795,
      "grad_norm": 3.9249801635742188,
      "learning_rate": 7.30637408056042e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 61921832,
      "step": 3849
    },
    {
      "epoch": 0.26968574605759715,
      "grad_norm": 4.512659072875977,
      "learning_rate": 7.305674255691769e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 61937928,
      "step": 3850
    },
    {
      "epoch": 0.2697557943033264,
      "grad_norm": 3.8067586421966553,
      "learning_rate": 7.304974430823118e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 61953992,
      "step": 3851
    },
    {
      "epoch": 0.26982584254905567,
      "grad_norm": 3.5481879711151123,
      "learning_rate": 7.304274605954466e-05,
      "loss": 0.9524,
      "num_input_tokens_seen": 61969856,
      "step": 3852
    },
    {
      "epoch": 0.26989589079478493,
      "grad_norm": 5.14021635055542,
      "learning_rate": 7.303574781085815e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 61985448,
      "step": 3853
    },
    {
      "epoch": 0.26996593904051414,
      "grad_norm": 4.729730606079102,
      "learning_rate": 7.302874956217164e-05,
      "loss": 0.955,
      "num_input_tokens_seen": 62001832,
      "step": 3854
    },
    {
      "epoch": 0.2700359872862434,
      "grad_norm": 4.081509113311768,
      "learning_rate": 7.302175131348512e-05,
      "loss": 1.3099,
      "num_input_tokens_seen": 62018216,
      "step": 3855
    },
    {
      "epoch": 0.27010603553197265,
      "grad_norm": 3.9220404624938965,
      "learning_rate": 7.301475306479859e-05,
      "loss": 1.256,
      "num_input_tokens_seen": 62034600,
      "step": 3856
    },
    {
      "epoch": 0.2701760837777019,
      "grad_norm": 3.9707326889038086,
      "learning_rate": 7.30077548161121e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 62050984,
      "step": 3857
    },
    {
      "epoch": 0.2702461320234311,
      "grad_norm": 3.985651731491089,
      "learning_rate": 7.300075656742557e-05,
      "loss": 1.0869,
      "num_input_tokens_seen": 62066496,
      "step": 3858
    },
    {
      "epoch": 0.2703161802691604,
      "grad_norm": 4.900750160217285,
      "learning_rate": 7.299375831873906e-05,
      "loss": 1.2112,
      "num_input_tokens_seen": 62082880,
      "step": 3859
    },
    {
      "epoch": 0.27038622851488964,
      "grad_norm": 3.7562901973724365,
      "learning_rate": 7.298676007005255e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 62099264,
      "step": 3860
    },
    {
      "epoch": 0.2704562767606189,
      "grad_norm": 4.3399271965026855,
      "learning_rate": 7.297976182136604e-05,
      "loss": 1.2113,
      "num_input_tokens_seen": 62115648,
      "step": 3861
    },
    {
      "epoch": 0.2705263250063481,
      "grad_norm": 3.792924642562866,
      "learning_rate": 7.297276357267951e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 62132032,
      "step": 3862
    },
    {
      "epoch": 0.27059637325207736,
      "grad_norm": 4.10078763961792,
      "learning_rate": 7.2965765323993e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 62148416,
      "step": 3863
    },
    {
      "epoch": 0.2706664214978066,
      "grad_norm": 3.6712818145751953,
      "learning_rate": 7.295876707530649e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 62164080,
      "step": 3864
    },
    {
      "epoch": 0.2707364697435359,
      "grad_norm": 4.216330051422119,
      "learning_rate": 7.295176882661996e-05,
      "loss": 0.9988,
      "num_input_tokens_seen": 62179952,
      "step": 3865
    },
    {
      "epoch": 0.2708065179892651,
      "grad_norm": 3.803950548171997,
      "learning_rate": 7.294477057793345e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 62196336,
      "step": 3866
    },
    {
      "epoch": 0.27087656623499434,
      "grad_norm": 4.4687676429748535,
      "learning_rate": 7.293777232924694e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 62212072,
      "step": 3867
    },
    {
      "epoch": 0.2709466144807236,
      "grad_norm": 3.8923938274383545,
      "learning_rate": 7.293077408056043e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 62227384,
      "step": 3868
    },
    {
      "epoch": 0.27101666272645286,
      "grad_norm": 3.7378618717193604,
      "learning_rate": 7.29237758318739e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 62243768,
      "step": 3869
    },
    {
      "epoch": 0.27108671097218207,
      "grad_norm": 4.39946985244751,
      "learning_rate": 7.291677758318739e-05,
      "loss": 1.2908,
      "num_input_tokens_seen": 62259760,
      "step": 3870
    },
    {
      "epoch": 0.2711567592179113,
      "grad_norm": 4.526809215545654,
      "learning_rate": 7.290977933450088e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 62275880,
      "step": 3871
    },
    {
      "epoch": 0.2712268074636406,
      "grad_norm": 5.780641078948975,
      "learning_rate": 7.290278108581437e-05,
      "loss": 1.3366,
      "num_input_tokens_seen": 62291992,
      "step": 3872
    },
    {
      "epoch": 0.27129685570936984,
      "grad_norm": 3.932300329208374,
      "learning_rate": 7.289578283712786e-05,
      "loss": 0.9404,
      "num_input_tokens_seen": 62308168,
      "step": 3873
    },
    {
      "epoch": 0.27136690395509905,
      "grad_norm": 6.381493091583252,
      "learning_rate": 7.288878458844135e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 62324552,
      "step": 3874
    },
    {
      "epoch": 0.2714369522008283,
      "grad_norm": 6.920464515686035,
      "learning_rate": 7.288178633975482e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 62340712,
      "step": 3875
    },
    {
      "epoch": 0.27150700044655757,
      "grad_norm": 4.327527046203613,
      "learning_rate": 7.28747880910683e-05,
      "loss": 1.2133,
      "num_input_tokens_seen": 62355904,
      "step": 3876
    },
    {
      "epoch": 0.2715770486922868,
      "grad_norm": 6.8873610496521,
      "learning_rate": 7.286778984238178e-05,
      "loss": 1.1857,
      "num_input_tokens_seen": 62372288,
      "step": 3877
    },
    {
      "epoch": 0.2716470969380161,
      "grad_norm": 4.397764205932617,
      "learning_rate": 7.286079159369527e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 62388672,
      "step": 3878
    },
    {
      "epoch": 0.2717171451837453,
      "grad_norm": 4.200334072113037,
      "learning_rate": 7.285379334500876e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 62403728,
      "step": 3879
    },
    {
      "epoch": 0.27178719342947455,
      "grad_norm": 3.8102898597717285,
      "learning_rate": 7.284679509632225e-05,
      "loss": 1.2455,
      "num_input_tokens_seen": 62419712,
      "step": 3880
    },
    {
      "epoch": 0.2718572416752038,
      "grad_norm": 5.665886878967285,
      "learning_rate": 7.283979684763574e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 62435648,
      "step": 3881
    },
    {
      "epoch": 0.27192728992093307,
      "grad_norm": 5.59833288192749,
      "learning_rate": 7.283279859894921e-05,
      "loss": 1.1289,
      "num_input_tokens_seen": 62451760,
      "step": 3882
    },
    {
      "epoch": 0.2719973381666623,
      "grad_norm": 4.3096699714660645,
      "learning_rate": 7.282580035026269e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 62468144,
      "step": 3883
    },
    {
      "epoch": 0.27206738641239153,
      "grad_norm": 3.584202766418457,
      "learning_rate": 7.281880210157619e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 62484528,
      "step": 3884
    },
    {
      "epoch": 0.2721374346581208,
      "grad_norm": 5.078696250915527,
      "learning_rate": 7.281180385288967e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 62500912,
      "step": 3885
    },
    {
      "epoch": 0.27220748290385005,
      "grad_norm": 3.4883761405944824,
      "learning_rate": 7.280480560420315e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 62517296,
      "step": 3886
    },
    {
      "epoch": 0.27227753114957925,
      "grad_norm": 3.938286066055298,
      "learning_rate": 7.279780735551664e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 62532896,
      "step": 3887
    },
    {
      "epoch": 0.2723475793953085,
      "grad_norm": 3.7150652408599854,
      "learning_rate": 7.279080910683013e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 62549072,
      "step": 3888
    },
    {
      "epoch": 0.2724176276410378,
      "grad_norm": 5.31076717376709,
      "learning_rate": 7.278381085814361e-05,
      "loss": 0.9943,
      "num_input_tokens_seen": 62564384,
      "step": 3889
    },
    {
      "epoch": 0.27248767588676703,
      "grad_norm": 4.8600053787231445,
      "learning_rate": 7.27768126094571e-05,
      "loss": 1.1767,
      "num_input_tokens_seen": 62580768,
      "step": 3890
    },
    {
      "epoch": 0.27255772413249624,
      "grad_norm": 3.5890231132507324,
      "learning_rate": 7.276981436077058e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 62596928,
      "step": 3891
    },
    {
      "epoch": 0.2726277723782255,
      "grad_norm": 4.171263217926025,
      "learning_rate": 7.276281611208407e-05,
      "loss": 1.0013,
      "num_input_tokens_seen": 62613312,
      "step": 3892
    },
    {
      "epoch": 0.27269782062395476,
      "grad_norm": 5.907830715179443,
      "learning_rate": 7.275581786339755e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 62627840,
      "step": 3893
    },
    {
      "epoch": 0.272767868869684,
      "grad_norm": 3.912140369415283,
      "learning_rate": 7.274881961471104e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 62643760,
      "step": 3894
    },
    {
      "epoch": 0.2728379171154132,
      "grad_norm": 3.9871180057525635,
      "learning_rate": 7.274182136602453e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 62660144,
      "step": 3895
    },
    {
      "epoch": 0.2729079653611425,
      "grad_norm": 3.8014907836914062,
      "learning_rate": 7.2734823117338e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 62676200,
      "step": 3896
    },
    {
      "epoch": 0.27297801360687174,
      "grad_norm": 3.7584786415100098,
      "learning_rate": 7.272782486865149e-05,
      "loss": 1.0366,
      "num_input_tokens_seen": 62692584,
      "step": 3897
    },
    {
      "epoch": 0.273048061852601,
      "grad_norm": 3.573341131210327,
      "learning_rate": 7.272082661996498e-05,
      "loss": 0.8726,
      "num_input_tokens_seen": 62708968,
      "step": 3898
    },
    {
      "epoch": 0.2731181100983302,
      "grad_norm": 4.013971328735352,
      "learning_rate": 7.271382837127847e-05,
      "loss": 0.991,
      "num_input_tokens_seen": 62725352,
      "step": 3899
    },
    {
      "epoch": 0.27318815834405946,
      "grad_norm": 4.3081488609313965,
      "learning_rate": 7.270683012259195e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 62741736,
      "step": 3900
    },
    {
      "epoch": 0.2732582065897887,
      "grad_norm": 3.857982635498047,
      "learning_rate": 7.269983187390544e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 62757624,
      "step": 3901
    },
    {
      "epoch": 0.273328254835518,
      "grad_norm": 3.5167486667633057,
      "learning_rate": 7.269283362521892e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 62774008,
      "step": 3902
    },
    {
      "epoch": 0.2733983030812472,
      "grad_norm": 4.025612831115723,
      "learning_rate": 7.26858353765324e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 62789560,
      "step": 3903
    },
    {
      "epoch": 0.27346835132697644,
      "grad_norm": 3.6391422748565674,
      "learning_rate": 7.267883712784588e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 62805824,
      "step": 3904
    },
    {
      "epoch": 0.2735383995727057,
      "grad_norm": 4.352347373962402,
      "learning_rate": 7.267183887915937e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 62821368,
      "step": 3905
    },
    {
      "epoch": 0.27360844781843496,
      "grad_norm": 3.782601833343506,
      "learning_rate": 7.266484063047286e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 62837024,
      "step": 3906
    },
    {
      "epoch": 0.27367849606416417,
      "grad_norm": 3.860903263092041,
      "learning_rate": 7.265784238178635e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 62853408,
      "step": 3907
    },
    {
      "epoch": 0.2737485443098934,
      "grad_norm": 6.185113430023193,
      "learning_rate": 7.265084413309984e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 62869792,
      "step": 3908
    },
    {
      "epoch": 0.2738185925556227,
      "grad_norm": 6.02334451675415,
      "learning_rate": 7.264384588441331e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 62886088,
      "step": 3909
    },
    {
      "epoch": 0.27388864080135195,
      "grad_norm": 4.019417762756348,
      "learning_rate": 7.263684763572679e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 62902472,
      "step": 3910
    },
    {
      "epoch": 0.27395868904708115,
      "grad_norm": 4.0645527839660645,
      "learning_rate": 7.262984938704029e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 62918552,
      "step": 3911
    },
    {
      "epoch": 0.2740287372928104,
      "grad_norm": 3.978803873062134,
      "learning_rate": 7.262285113835378e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 62934272,
      "step": 3912
    },
    {
      "epoch": 0.27409878553853967,
      "grad_norm": 4.659839630126953,
      "learning_rate": 7.261585288966725e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 62950656,
      "step": 3913
    },
    {
      "epoch": 0.27416883378426893,
      "grad_norm": 4.378306865692139,
      "learning_rate": 7.260885464098074e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 62966120,
      "step": 3914
    },
    {
      "epoch": 0.2742388820299982,
      "grad_norm": 3.723999261856079,
      "learning_rate": 7.260185639229423e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 62982504,
      "step": 3915
    },
    {
      "epoch": 0.2743089302757274,
      "grad_norm": 4.133684158325195,
      "learning_rate": 7.25948581436077e-05,
      "loss": 0.9707,
      "num_input_tokens_seen": 62998888,
      "step": 3916
    },
    {
      "epoch": 0.27437897852145665,
      "grad_norm": 3.8377842903137207,
      "learning_rate": 7.258785989492119e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 63015272,
      "step": 3917
    },
    {
      "epoch": 0.2744490267671859,
      "grad_norm": 3.546846389770508,
      "learning_rate": 7.258086164623468e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 63031656,
      "step": 3918
    },
    {
      "epoch": 0.27451907501291517,
      "grad_norm": 3.8629097938537598,
      "learning_rate": 7.257386339754817e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 63047208,
      "step": 3919
    },
    {
      "epoch": 0.2745891232586444,
      "grad_norm": 3.780395984649658,
      "learning_rate": 7.256686514886165e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 63063592,
      "step": 3920
    },
    {
      "epoch": 0.27465917150437363,
      "grad_norm": 3.5188148021698,
      "learning_rate": 7.255986690017513e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 63079976,
      "step": 3921
    },
    {
      "epoch": 0.2747292197501029,
      "grad_norm": 4.295319080352783,
      "learning_rate": 7.255286865148862e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 63096360,
      "step": 3922
    },
    {
      "epoch": 0.27479926799583215,
      "grad_norm": 6.307181358337402,
      "learning_rate": 7.25458704028021e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 63112744,
      "step": 3923
    },
    {
      "epoch": 0.27486931624156136,
      "grad_norm": 4.0670342445373535,
      "learning_rate": 7.253887215411559e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 63129000,
      "step": 3924
    },
    {
      "epoch": 0.2749393644872906,
      "grad_norm": 4.441539287567139,
      "learning_rate": 7.253187390542907e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 63145304,
      "step": 3925
    },
    {
      "epoch": 0.2750094127330199,
      "grad_norm": 6.151254653930664,
      "learning_rate": 7.252487565674256e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 63161688,
      "step": 3926
    },
    {
      "epoch": 0.27507946097874914,
      "grad_norm": 5.355491638183594,
      "learning_rate": 7.251787740805605e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 63176128,
      "step": 3927
    },
    {
      "epoch": 0.27514950922447834,
      "grad_norm": 3.4603800773620605,
      "learning_rate": 7.251087915936954e-05,
      "loss": 0.7811,
      "num_input_tokens_seen": 63192512,
      "step": 3928
    },
    {
      "epoch": 0.2752195574702076,
      "grad_norm": 5.412753105163574,
      "learning_rate": 7.250388091068302e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 63208896,
      "step": 3929
    },
    {
      "epoch": 0.27528960571593686,
      "grad_norm": 3.928074598312378,
      "learning_rate": 7.249688266199649e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 63224296,
      "step": 3930
    },
    {
      "epoch": 0.2753596539616661,
      "grad_norm": 4.239214897155762,
      "learning_rate": 7.248988441330998e-05,
      "loss": 0.9697,
      "num_input_tokens_seen": 63239312,
      "step": 3931
    },
    {
      "epoch": 0.2754297022073953,
      "grad_norm": 3.8074252605438232,
      "learning_rate": 7.248288616462348e-05,
      "loss": 1.0834,
      "num_input_tokens_seen": 63255664,
      "step": 3932
    },
    {
      "epoch": 0.2754997504531246,
      "grad_norm": 3.721026659011841,
      "learning_rate": 7.247588791593696e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 63272048,
      "step": 3933
    },
    {
      "epoch": 0.27556979869885384,
      "grad_norm": 4.076726913452148,
      "learning_rate": 7.246888966725044e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 63288432,
      "step": 3934
    },
    {
      "epoch": 0.2756398469445831,
      "grad_norm": 4.238835334777832,
      "learning_rate": 7.246189141856393e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 63304168,
      "step": 3935
    },
    {
      "epoch": 0.2757098951903123,
      "grad_norm": 4.4860148429870605,
      "learning_rate": 7.245489316987741e-05,
      "loss": 1.1763,
      "num_input_tokens_seen": 63320552,
      "step": 3936
    },
    {
      "epoch": 0.27577994343604156,
      "grad_norm": 6.002726078033447,
      "learning_rate": 7.244789492119088e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 63336792,
      "step": 3937
    },
    {
      "epoch": 0.2758499916817708,
      "grad_norm": 3.799751043319702,
      "learning_rate": 7.244089667250439e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 63353176,
      "step": 3938
    },
    {
      "epoch": 0.2759200399275001,
      "grad_norm": 4.905911445617676,
      "learning_rate": 7.243389842381787e-05,
      "loss": 0.8847,
      "num_input_tokens_seen": 63369560,
      "step": 3939
    },
    {
      "epoch": 0.2759900881732293,
      "grad_norm": 5.141537666320801,
      "learning_rate": 7.242690017513135e-05,
      "loss": 1.109,
      "num_input_tokens_seen": 63385944,
      "step": 3940
    },
    {
      "epoch": 0.27606013641895855,
      "grad_norm": 5.276777267456055,
      "learning_rate": 7.241990192644484e-05,
      "loss": 0.9881,
      "num_input_tokens_seen": 63401672,
      "step": 3941
    },
    {
      "epoch": 0.2761301846646878,
      "grad_norm": 5.267075538635254,
      "learning_rate": 7.241290367775833e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 63417792,
      "step": 3942
    },
    {
      "epoch": 0.27620023291041707,
      "grad_norm": 4.065691947937012,
      "learning_rate": 7.24059054290718e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 63434176,
      "step": 3943
    },
    {
      "epoch": 0.27627028115614627,
      "grad_norm": 7.921762466430664,
      "learning_rate": 7.239890718038529e-05,
      "loss": 1.3552,
      "num_input_tokens_seen": 63450032,
      "step": 3944
    },
    {
      "epoch": 0.27634032940187553,
      "grad_norm": 3.55094313621521,
      "learning_rate": 7.239190893169878e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 63466416,
      "step": 3945
    },
    {
      "epoch": 0.2764103776476048,
      "grad_norm": 5.732813358306885,
      "learning_rate": 7.238491068301227e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 63482296,
      "step": 3946
    },
    {
      "epoch": 0.27648042589333405,
      "grad_norm": 3.9143989086151123,
      "learning_rate": 7.237791243432574e-05,
      "loss": 0.9218,
      "num_input_tokens_seen": 63498472,
      "step": 3947
    },
    {
      "epoch": 0.2765504741390633,
      "grad_norm": 4.123042106628418,
      "learning_rate": 7.237091418563923e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 63513856,
      "step": 3948
    },
    {
      "epoch": 0.2766205223847925,
      "grad_norm": 3.7550277709960938,
      "learning_rate": 7.236391593695272e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 63529432,
      "step": 3949
    },
    {
      "epoch": 0.27669057063052177,
      "grad_norm": 3.841831922531128,
      "learning_rate": 7.23569176882662e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 63545816,
      "step": 3950
    },
    {
      "epoch": 0.27676061887625103,
      "grad_norm": 4.626603126525879,
      "learning_rate": 7.234991943957968e-05,
      "loss": 1.3412,
      "num_input_tokens_seen": 63561960,
      "step": 3951
    },
    {
      "epoch": 0.2768306671219803,
      "grad_norm": 3.874140977859497,
      "learning_rate": 7.234292119089319e-05,
      "loss": 1.0549,
      "num_input_tokens_seen": 63578344,
      "step": 3952
    },
    {
      "epoch": 0.2769007153677095,
      "grad_norm": 3.6525163650512695,
      "learning_rate": 7.233592294220666e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 63594520,
      "step": 3953
    },
    {
      "epoch": 0.27697076361343875,
      "grad_norm": 5.065535068511963,
      "learning_rate": 7.232892469352015e-05,
      "loss": 1.1913,
      "num_input_tokens_seen": 63610904,
      "step": 3954
    },
    {
      "epoch": 0.277040811859168,
      "grad_norm": 7.97597599029541,
      "learning_rate": 7.232192644483364e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 63625896,
      "step": 3955
    },
    {
      "epoch": 0.27711086010489727,
      "grad_norm": 5.0254645347595215,
      "learning_rate": 7.231492819614711e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 63642280,
      "step": 3956
    },
    {
      "epoch": 0.2771809083506265,
      "grad_norm": 4.171605587005615,
      "learning_rate": 7.230792994746059e-05,
      "loss": 1.3166,
      "num_input_tokens_seen": 63658400,
      "step": 3957
    },
    {
      "epoch": 0.27725095659635574,
      "grad_norm": 4.036003589630127,
      "learning_rate": 7.230093169877409e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 63674784,
      "step": 3958
    },
    {
      "epoch": 0.277321004842085,
      "grad_norm": 4.664374828338623,
      "learning_rate": 7.229393345008758e-05,
      "loss": 1.3189,
      "num_input_tokens_seen": 63691168,
      "step": 3959
    },
    {
      "epoch": 0.27739105308781425,
      "grad_norm": 3.7217307090759277,
      "learning_rate": 7.228693520140105e-05,
      "loss": 1.2532,
      "num_input_tokens_seen": 63707552,
      "step": 3960
    },
    {
      "epoch": 0.27746110133354346,
      "grad_norm": 3.622593879699707,
      "learning_rate": 7.227993695271454e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 63723936,
      "step": 3961
    },
    {
      "epoch": 0.2775311495792727,
      "grad_norm": 4.154850006103516,
      "learning_rate": 7.227293870402803e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 63740320,
      "step": 3962
    },
    {
      "epoch": 0.277601197825002,
      "grad_norm": 4.157016754150391,
      "learning_rate": 7.22659404553415e-05,
      "loss": 0.8815,
      "num_input_tokens_seen": 63756456,
      "step": 3963
    },
    {
      "epoch": 0.27767124607073124,
      "grad_norm": 4.652394771575928,
      "learning_rate": 7.2258942206655e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 63772840,
      "step": 3964
    },
    {
      "epoch": 0.27774129431646044,
      "grad_norm": 7.87667989730835,
      "learning_rate": 7.225194395796848e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 63788800,
      "step": 3965
    },
    {
      "epoch": 0.2778113425621897,
      "grad_norm": 4.333608627319336,
      "learning_rate": 7.224494570928197e-05,
      "loss": 1.3465,
      "num_input_tokens_seen": 63805088,
      "step": 3966
    },
    {
      "epoch": 0.27788139080791896,
      "grad_norm": 7.2095184326171875,
      "learning_rate": 7.223794746059545e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 63821472,
      "step": 3967
    },
    {
      "epoch": 0.2779514390536482,
      "grad_norm": 3.9144251346588135,
      "learning_rate": 7.223094921190893e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 63837048,
      "step": 3968
    },
    {
      "epoch": 0.2780214872993774,
      "grad_norm": 4.380809783935547,
      "learning_rate": 7.222395096322242e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 63852872,
      "step": 3969
    },
    {
      "epoch": 0.2780915355451067,
      "grad_norm": 3.637685537338257,
      "learning_rate": 7.22169527145359e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 63868864,
      "step": 3970
    },
    {
      "epoch": 0.27816158379083594,
      "grad_norm": 4.742129802703857,
      "learning_rate": 7.220995446584939e-05,
      "loss": 1.2344,
      "num_input_tokens_seen": 63885248,
      "step": 3971
    },
    {
      "epoch": 0.2782316320365652,
      "grad_norm": 4.7221269607543945,
      "learning_rate": 7.220295621716289e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 63901632,
      "step": 3972
    },
    {
      "epoch": 0.2783016802822944,
      "grad_norm": 3.6607449054718018,
      "learning_rate": 7.219595796847636e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 63917688,
      "step": 3973
    },
    {
      "epoch": 0.27837172852802367,
      "grad_norm": 4.264851093292236,
      "learning_rate": 7.218895971978984e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 63934072,
      "step": 3974
    },
    {
      "epoch": 0.2784417767737529,
      "grad_norm": 5.0043511390686035,
      "learning_rate": 7.218196147110333e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 63950200,
      "step": 3975
    },
    {
      "epoch": 0.2785118250194822,
      "grad_norm": 4.323488235473633,
      "learning_rate": 7.217496322241682e-05,
      "loss": 1.1791,
      "num_input_tokens_seen": 63966584,
      "step": 3976
    },
    {
      "epoch": 0.2785818732652114,
      "grad_norm": 6.721888065338135,
      "learning_rate": 7.216796497373029e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 63982440,
      "step": 3977
    },
    {
      "epoch": 0.27865192151094065,
      "grad_norm": 6.3528289794921875,
      "learning_rate": 7.21609667250438e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 63998824,
      "step": 3978
    },
    {
      "epoch": 0.2787219697566699,
      "grad_norm": 5.293467044830322,
      "learning_rate": 7.215396847635728e-05,
      "loss": 1.2791,
      "num_input_tokens_seen": 64014984,
      "step": 3979
    },
    {
      "epoch": 0.27879201800239917,
      "grad_norm": 3.8228442668914795,
      "learning_rate": 7.214697022767076e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 64031080,
      "step": 3980
    },
    {
      "epoch": 0.27886206624812837,
      "grad_norm": 3.8407061100006104,
      "learning_rate": 7.213997197898425e-05,
      "loss": 1.1772,
      "num_input_tokens_seen": 64046416,
      "step": 3981
    },
    {
      "epoch": 0.27893211449385763,
      "grad_norm": 3.9471728801727295,
      "learning_rate": 7.213297373029773e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 64062784,
      "step": 3982
    },
    {
      "epoch": 0.2790021627395869,
      "grad_norm": 4.1796722412109375,
      "learning_rate": 7.212597548161121e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 64077504,
      "step": 3983
    },
    {
      "epoch": 0.27907221098531615,
      "grad_norm": 3.78998064994812,
      "learning_rate": 7.21189772329247e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 64093888,
      "step": 3984
    },
    {
      "epoch": 0.2791422592310454,
      "grad_norm": 3.383371591567993,
      "learning_rate": 7.211197898423819e-05,
      "loss": 0.8832,
      "num_input_tokens_seen": 64110272,
      "step": 3985
    },
    {
      "epoch": 0.2792123074767746,
      "grad_norm": 3.6502346992492676,
      "learning_rate": 7.210498073555168e-05,
      "loss": 1.0114,
      "num_input_tokens_seen": 64125464,
      "step": 3986
    },
    {
      "epoch": 0.27928235572250387,
      "grad_norm": 3.9421629905700684,
      "learning_rate": 7.209798248686515e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 64141848,
      "step": 3987
    },
    {
      "epoch": 0.27935240396823313,
      "grad_norm": 4.40875244140625,
      "learning_rate": 7.209098423817864e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 64158232,
      "step": 3988
    },
    {
      "epoch": 0.2794224522139624,
      "grad_norm": 5.909340858459473,
      "learning_rate": 7.208398598949213e-05,
      "loss": 1.121,
      "num_input_tokens_seen": 64174616,
      "step": 3989
    },
    {
      "epoch": 0.2794925004596916,
      "grad_norm": 4.548187732696533,
      "learning_rate": 7.20769877408056e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 64191000,
      "step": 3990
    },
    {
      "epoch": 0.27956254870542085,
      "grad_norm": 4.1479926109313965,
      "learning_rate": 7.206998949211909e-05,
      "loss": 1.1205,
      "num_input_tokens_seen": 64207384,
      "step": 3991
    },
    {
      "epoch": 0.2796325969511501,
      "grad_norm": 4.229100227355957,
      "learning_rate": 7.206299124343259e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 64223304,
      "step": 3992
    },
    {
      "epoch": 0.2797026451968794,
      "grad_norm": 7.431615352630615,
      "learning_rate": 7.205599299474607e-05,
      "loss": 1.2793,
      "num_input_tokens_seen": 64239688,
      "step": 3993
    },
    {
      "epoch": 0.2797726934426086,
      "grad_norm": 4.346622943878174,
      "learning_rate": 7.204899474605954e-05,
      "loss": 1.2385,
      "num_input_tokens_seen": 64255752,
      "step": 3994
    },
    {
      "epoch": 0.27984274168833784,
      "grad_norm": 3.771306276321411,
      "learning_rate": 7.204199649737303e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 64271760,
      "step": 3995
    },
    {
      "epoch": 0.2799127899340671,
      "grad_norm": 4.411479473114014,
      "learning_rate": 7.203499824868652e-05,
      "loss": 1.3369,
      "num_input_tokens_seen": 64288144,
      "step": 3996
    },
    {
      "epoch": 0.27998283817979636,
      "grad_norm": 4.591271877288818,
      "learning_rate": 7.2028e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 64304528,
      "step": 3997
    },
    {
      "epoch": 0.28005288642552556,
      "grad_norm": 3.88271427154541,
      "learning_rate": 7.20210017513135e-05,
      "loss": 0.9181,
      "num_input_tokens_seen": 64320912,
      "step": 3998
    },
    {
      "epoch": 0.2801229346712548,
      "grad_norm": 10.80846118927002,
      "learning_rate": 7.201400350262699e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 64337296,
      "step": 3999
    },
    {
      "epoch": 0.2801929829169841,
      "grad_norm": 3.7112953662872314,
      "learning_rate": 7.200700525394046e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 64353680,
      "step": 4000
    },
    {
      "epoch": 0.2801929829169841,
      "eval_loss": 1.1334750652313232,
      "eval_runtime": 0.1958,
      "eval_samples_per_second": 5.106,
      "eval_steps_per_second": 5.106,
      "num_input_tokens_seen": 64353680,
      "step": 4000
    },
    {
      "epoch": 0.28026303116271334,
      "grad_norm": 7.529544830322266,
      "learning_rate": 7.200000700525394e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 64368408,
      "step": 4001
    },
    {
      "epoch": 0.28033307940844254,
      "grad_norm": 3.761939764022827,
      "learning_rate": 7.199300875656742e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 64384792,
      "step": 4002
    },
    {
      "epoch": 0.2804031276541718,
      "grad_norm": 4.091811656951904,
      "learning_rate": 7.198601050788091e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 64400520,
      "step": 4003
    },
    {
      "epoch": 0.28047317589990106,
      "grad_norm": 5.5972795486450195,
      "learning_rate": 7.19790122591944e-05,
      "loss": 1.0957,
      "num_input_tokens_seen": 64416904,
      "step": 4004
    },
    {
      "epoch": 0.2805432241456303,
      "grad_norm": 3.4631423950195312,
      "learning_rate": 7.197201401050789e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 64432168,
      "step": 4005
    },
    {
      "epoch": 0.2806132723913595,
      "grad_norm": 6.3156938552856445,
      "learning_rate": 7.196501576182138e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 64447752,
      "step": 4006
    },
    {
      "epoch": 0.2806833206370888,
      "grad_norm": 10.07819652557373,
      "learning_rate": 7.195801751313485e-05,
      "loss": 0.99,
      "num_input_tokens_seen": 64464136,
      "step": 4007
    },
    {
      "epoch": 0.28075336888281804,
      "grad_norm": 4.695057392120361,
      "learning_rate": 7.195101926444834e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 64480520,
      "step": 4008
    },
    {
      "epoch": 0.2808234171285473,
      "grad_norm": 4.74672269821167,
      "learning_rate": 7.194402101576183e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 64496904,
      "step": 4009
    },
    {
      "epoch": 0.2808934653742765,
      "grad_norm": 3.834928512573242,
      "learning_rate": 7.19370227670753e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 64513288,
      "step": 4010
    },
    {
      "epoch": 0.28096351362000577,
      "grad_norm": 4.1937103271484375,
      "learning_rate": 7.19300245183888e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 64528992,
      "step": 4011
    },
    {
      "epoch": 0.281033561865735,
      "grad_norm": 4.1531243324279785,
      "learning_rate": 7.192302626970228e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 64544776,
      "step": 4012
    },
    {
      "epoch": 0.2811036101114643,
      "grad_norm": 5.006285190582275,
      "learning_rate": 7.191602802101577e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 64560944,
      "step": 4013
    },
    {
      "epoch": 0.2811736583571935,
      "grad_norm": 3.433964252471924,
      "learning_rate": 7.190902977232925e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 64577328,
      "step": 4014
    },
    {
      "epoch": 0.28124370660292275,
      "grad_norm": 6.165640354156494,
      "learning_rate": 7.190203152364274e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 64593672,
      "step": 4015
    },
    {
      "epoch": 0.281313754848652,
      "grad_norm": 6.037381649017334,
      "learning_rate": 7.189503327495622e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 64610056,
      "step": 4016
    },
    {
      "epoch": 0.28138380309438127,
      "grad_norm": 4.2639923095703125,
      "learning_rate": 7.18880350262697e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 64625864,
      "step": 4017
    },
    {
      "epoch": 0.28145385134011053,
      "grad_norm": 3.8862967491149902,
      "learning_rate": 7.188103677758319e-05,
      "loss": 0.9448,
      "num_input_tokens_seen": 64642248,
      "step": 4018
    },
    {
      "epoch": 0.28152389958583973,
      "grad_norm": 3.9584991931915283,
      "learning_rate": 7.187403852889669e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 64658632,
      "step": 4019
    },
    {
      "epoch": 0.281593947831569,
      "grad_norm": 6.037077903747559,
      "learning_rate": 7.186704028021017e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 64675016,
      "step": 4020
    },
    {
      "epoch": 0.28166399607729825,
      "grad_norm": 3.750059127807617,
      "learning_rate": 7.186004203152364e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 64691400,
      "step": 4021
    },
    {
      "epoch": 0.2817340443230275,
      "grad_norm": 4.364743709564209,
      "learning_rate": 7.185304378283713e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 64706512,
      "step": 4022
    },
    {
      "epoch": 0.2818040925687567,
      "grad_norm": 3.463717460632324,
      "learning_rate": 7.184604553415062e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 64722688,
      "step": 4023
    },
    {
      "epoch": 0.281874140814486,
      "grad_norm": 3.939438819885254,
      "learning_rate": 7.18390472854641e-05,
      "loss": 1.3356,
      "num_input_tokens_seen": 64738856,
      "step": 4024
    },
    {
      "epoch": 0.28194418906021523,
      "grad_norm": 3.813849687576294,
      "learning_rate": 7.18320490367776e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 64755240,
      "step": 4025
    },
    {
      "epoch": 0.2820142373059445,
      "grad_norm": 3.5874619483947754,
      "learning_rate": 7.182505078809108e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 64771184,
      "step": 4026
    },
    {
      "epoch": 0.2820842855516737,
      "grad_norm": 4.544376850128174,
      "learning_rate": 7.181805253940456e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 64787568,
      "step": 4027
    },
    {
      "epoch": 0.28215433379740296,
      "grad_norm": 3.6816799640655518,
      "learning_rate": 7.181105429071803e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 64803064,
      "step": 4028
    },
    {
      "epoch": 0.2822243820431322,
      "grad_norm": 7.1433939933776855,
      "learning_rate": 7.180405604203152e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 64818736,
      "step": 4029
    },
    {
      "epoch": 0.2822944302888615,
      "grad_norm": 4.308315753936768,
      "learning_rate": 7.179705779334501e-05,
      "loss": 1.1992,
      "num_input_tokens_seen": 64834848,
      "step": 4030
    },
    {
      "epoch": 0.2823644785345907,
      "grad_norm": 4.985830783843994,
      "learning_rate": 7.17900595446585e-05,
      "loss": 1.1996,
      "num_input_tokens_seen": 64851224,
      "step": 4031
    },
    {
      "epoch": 0.28243452678031994,
      "grad_norm": 4.884370803833008,
      "learning_rate": 7.178306129597199e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 64867608,
      "step": 4032
    },
    {
      "epoch": 0.2825045750260492,
      "grad_norm": 4.335781097412109,
      "learning_rate": 7.177606304728548e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 64883840,
      "step": 4033
    },
    {
      "epoch": 0.28257462327177846,
      "grad_norm": 3.729811191558838,
      "learning_rate": 7.176906479859895e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 64899872,
      "step": 4034
    },
    {
      "epoch": 0.28264467151750766,
      "grad_norm": 3.7386136054992676,
      "learning_rate": 7.176206654991244e-05,
      "loss": 0.7835,
      "num_input_tokens_seen": 64916256,
      "step": 4035
    },
    {
      "epoch": 0.2827147197632369,
      "grad_norm": 3.8022067546844482,
      "learning_rate": 7.175506830122593e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 64932640,
      "step": 4036
    },
    {
      "epoch": 0.2827847680089662,
      "grad_norm": 4.713296890258789,
      "learning_rate": 7.17480700525394e-05,
      "loss": 1.2877,
      "num_input_tokens_seen": 64948520,
      "step": 4037
    },
    {
      "epoch": 0.28285481625469544,
      "grad_norm": 3.682568073272705,
      "learning_rate": 7.174107180385289e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 64964904,
      "step": 4038
    },
    {
      "epoch": 0.28292486450042464,
      "grad_norm": 4.533677101135254,
      "learning_rate": 7.173407355516638e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 64981288,
      "step": 4039
    },
    {
      "epoch": 0.2829949127461539,
      "grad_norm": 4.343021392822266,
      "learning_rate": 7.172707530647987e-05,
      "loss": 1.2843,
      "num_input_tokens_seen": 64997640,
      "step": 4040
    },
    {
      "epoch": 0.28306496099188316,
      "grad_norm": 4.942739009857178,
      "learning_rate": 7.172007705779334e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 65012456,
      "step": 4041
    },
    {
      "epoch": 0.2831350092376124,
      "grad_norm": 6.1112213134765625,
      "learning_rate": 7.171307880910683e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 65028840,
      "step": 4042
    },
    {
      "epoch": 0.2832050574833416,
      "grad_norm": 4.650609016418457,
      "learning_rate": 7.170608056042032e-05,
      "loss": 0.8585,
      "num_input_tokens_seen": 65044608,
      "step": 4043
    },
    {
      "epoch": 0.2832751057290709,
      "grad_norm": 5.383882522583008,
      "learning_rate": 7.169908231173381e-05,
      "loss": 1.3442,
      "num_input_tokens_seen": 65060992,
      "step": 4044
    },
    {
      "epoch": 0.28334515397480015,
      "grad_norm": 3.569399118423462,
      "learning_rate": 7.169208406304729e-05,
      "loss": 1.0645,
      "num_input_tokens_seen": 65077096,
      "step": 4045
    },
    {
      "epoch": 0.2834152022205294,
      "grad_norm": 5.199350833892822,
      "learning_rate": 7.168508581436079e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 65093480,
      "step": 4046
    },
    {
      "epoch": 0.2834852504662586,
      "grad_norm": 4.172554969787598,
      "learning_rate": 7.167808756567426e-05,
      "loss": 1.1721,
      "num_input_tokens_seen": 65109864,
      "step": 4047
    },
    {
      "epoch": 0.28355529871198787,
      "grad_norm": 3.822197437286377,
      "learning_rate": 7.167108931698774e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 65126248,
      "step": 4048
    },
    {
      "epoch": 0.28362534695771713,
      "grad_norm": 3.8899435997009277,
      "learning_rate": 7.166409106830123e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 65141984,
      "step": 4049
    },
    {
      "epoch": 0.2836953952034464,
      "grad_norm": 4.559451580047607,
      "learning_rate": 7.165709281961471e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 65157984,
      "step": 4050
    },
    {
      "epoch": 0.28376544344917565,
      "grad_norm": 5.253831386566162,
      "learning_rate": 7.16500945709282e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 65174040,
      "step": 4051
    },
    {
      "epoch": 0.28383549169490485,
      "grad_norm": 3.827268123626709,
      "learning_rate": 7.164309632224169e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 65190424,
      "step": 4052
    },
    {
      "epoch": 0.2839055399406341,
      "grad_norm": 4.432236194610596,
      "learning_rate": 7.163609807355518e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 65206808,
      "step": 4053
    },
    {
      "epoch": 0.28397558818636337,
      "grad_norm": 5.008002281188965,
      "learning_rate": 7.162909982486866e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 65222744,
      "step": 4054
    },
    {
      "epoch": 0.28404563643209263,
      "grad_norm": 3.748089551925659,
      "learning_rate": 7.162210157618213e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 65238592,
      "step": 4055
    },
    {
      "epoch": 0.28411568467782183,
      "grad_norm": 4.073843002319336,
      "learning_rate": 7.161510332749562e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 65254464,
      "step": 4056
    },
    {
      "epoch": 0.2841857329235511,
      "grad_norm": 4.0271100997924805,
      "learning_rate": 7.160810507880911e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 65269744,
      "step": 4057
    },
    {
      "epoch": 0.28425578116928035,
      "grad_norm": 4.266842365264893,
      "learning_rate": 7.16011068301226e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 65286128,
      "step": 4058
    },
    {
      "epoch": 0.2843258294150096,
      "grad_norm": 3.4473531246185303,
      "learning_rate": 7.159410858143608e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 65301864,
      "step": 4059
    },
    {
      "epoch": 0.2843958776607388,
      "grad_norm": 3.717029333114624,
      "learning_rate": 7.158711033274957e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 65317880,
      "step": 4060
    },
    {
      "epoch": 0.2844659259064681,
      "grad_norm": 4.008082866668701,
      "learning_rate": 7.158011208406305e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 65334096,
      "step": 4061
    },
    {
      "epoch": 0.28453597415219734,
      "grad_norm": 5.350658893585205,
      "learning_rate": 7.157311383537654e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 65348288,
      "step": 4062
    },
    {
      "epoch": 0.2846060223979266,
      "grad_norm": 8.911882400512695,
      "learning_rate": 7.156611558669003e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 65364672,
      "step": 4063
    },
    {
      "epoch": 0.2846760706436558,
      "grad_norm": 4.207833766937256,
      "learning_rate": 7.155911733800351e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 65380600,
      "step": 4064
    },
    {
      "epoch": 0.28474611888938506,
      "grad_norm": 3.492713689804077,
      "learning_rate": 7.155211908931699e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 65396920,
      "step": 4065
    },
    {
      "epoch": 0.2848161671351143,
      "grad_norm": 3.866763114929199,
      "learning_rate": 7.154512084063048e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 65413136,
      "step": 4066
    },
    {
      "epoch": 0.2848862153808436,
      "grad_norm": 4.352143287658691,
      "learning_rate": 7.153812259194397e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 65428368,
      "step": 4067
    },
    {
      "epoch": 0.2849562636265728,
      "grad_norm": 5.335500717163086,
      "learning_rate": 7.153112434325744e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 65444752,
      "step": 4068
    },
    {
      "epoch": 0.28502631187230204,
      "grad_norm": 3.7467970848083496,
      "learning_rate": 7.152412609457093e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 65461136,
      "step": 4069
    },
    {
      "epoch": 0.2850963601180313,
      "grad_norm": 3.410472869873047,
      "learning_rate": 7.151712784588442e-05,
      "loss": 0.8464,
      "num_input_tokens_seen": 65477520,
      "step": 4070
    },
    {
      "epoch": 0.28516640836376056,
      "grad_norm": 6.551929950714111,
      "learning_rate": 7.151012959719791e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 65493904,
      "step": 4071
    },
    {
      "epoch": 0.28523645660948976,
      "grad_norm": 3.4140212535858154,
      "learning_rate": 7.150313134851138e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 65510288,
      "step": 4072
    },
    {
      "epoch": 0.285306504855219,
      "grad_norm": 4.227553367614746,
      "learning_rate": 7.149613309982488e-05,
      "loss": 1.0793,
      "num_input_tokens_seen": 65526672,
      "step": 4073
    },
    {
      "epoch": 0.2853765531009483,
      "grad_norm": 4.202794551849365,
      "learning_rate": 7.148913485113836e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 65542456,
      "step": 4074
    },
    {
      "epoch": 0.28544660134667754,
      "grad_norm": 5.172013759613037,
      "learning_rate": 7.148213660245183e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 65558384,
      "step": 4075
    },
    {
      "epoch": 0.28551664959240675,
      "grad_norm": 3.716113567352295,
      "learning_rate": 7.147513835376532e-05,
      "loss": 0.8515,
      "num_input_tokens_seen": 65574768,
      "step": 4076
    },
    {
      "epoch": 0.285586697838136,
      "grad_norm": 8.10258674621582,
      "learning_rate": 7.146814010507881e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 65590632,
      "step": 4077
    },
    {
      "epoch": 0.28565674608386527,
      "grad_norm": 3.649273157119751,
      "learning_rate": 7.14611418563923e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 65607016,
      "step": 4078
    },
    {
      "epoch": 0.2857267943295945,
      "grad_norm": 4.202502250671387,
      "learning_rate": 7.145414360770579e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 65622856,
      "step": 4079
    },
    {
      "epoch": 0.28579684257532373,
      "grad_norm": 4.027415752410889,
      "learning_rate": 7.144714535901928e-05,
      "loss": 1.26,
      "num_input_tokens_seen": 65639240,
      "step": 4080
    },
    {
      "epoch": 0.285866890821053,
      "grad_norm": 4.549161434173584,
      "learning_rate": 7.144014711033275e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 65655624,
      "step": 4081
    },
    {
      "epoch": 0.28593693906678225,
      "grad_norm": 4.43501615524292,
      "learning_rate": 7.143314886164623e-05,
      "loss": 1.0735,
      "num_input_tokens_seen": 65671016,
      "step": 4082
    },
    {
      "epoch": 0.2860069873125115,
      "grad_norm": 3.739610433578491,
      "learning_rate": 7.142615061295972e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 65687072,
      "step": 4083
    },
    {
      "epoch": 0.2860770355582407,
      "grad_norm": 3.725759506225586,
      "learning_rate": 7.14191523642732e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 65703456,
      "step": 4084
    },
    {
      "epoch": 0.28614708380396997,
      "grad_norm": 3.706056594848633,
      "learning_rate": 7.14121541155867e-05,
      "loss": 1.0643,
      "num_input_tokens_seen": 65719552,
      "step": 4085
    },
    {
      "epoch": 0.28621713204969923,
      "grad_norm": 4.971164703369141,
      "learning_rate": 7.140515586690018e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 65735936,
      "step": 4086
    },
    {
      "epoch": 0.2862871802954285,
      "grad_norm": 7.377131938934326,
      "learning_rate": 7.139815761821367e-05,
      "loss": 0.8867,
      "num_input_tokens_seen": 65752320,
      "step": 4087
    },
    {
      "epoch": 0.28635722854115775,
      "grad_norm": 4.293169975280762,
      "learning_rate": 7.139115936952715e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 65768704,
      "step": 4088
    },
    {
      "epoch": 0.28642727678688695,
      "grad_norm": 3.4757955074310303,
      "learning_rate": 7.138416112084063e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 65785088,
      "step": 4089
    },
    {
      "epoch": 0.2864973250326162,
      "grad_norm": 4.5705695152282715,
      "learning_rate": 7.137716287215412e-05,
      "loss": 1.209,
      "num_input_tokens_seen": 65801472,
      "step": 4090
    },
    {
      "epoch": 0.28656737327834547,
      "grad_norm": 5.240487575531006,
      "learning_rate": 7.137016462346761e-05,
      "loss": 0.9684,
      "num_input_tokens_seen": 65817856,
      "step": 4091
    },
    {
      "epoch": 0.28663742152407473,
      "grad_norm": 3.7815425395965576,
      "learning_rate": 7.136316637478109e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 65833872,
      "step": 4092
    },
    {
      "epoch": 0.28670746976980394,
      "grad_norm": 5.411090850830078,
      "learning_rate": 7.135616812609457e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 65849064,
      "step": 4093
    },
    {
      "epoch": 0.2867775180155332,
      "grad_norm": 4.07004451751709,
      "learning_rate": 7.134916987740806e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 65865448,
      "step": 4094
    },
    {
      "epoch": 0.28684756626126245,
      "grad_norm": 3.636051893234253,
      "learning_rate": 7.134217162872154e-05,
      "loss": 0.9363,
      "num_input_tokens_seen": 65881320,
      "step": 4095
    },
    {
      "epoch": 0.2869176145069917,
      "grad_norm": 4.265620708465576,
      "learning_rate": 7.133517338003503e-05,
      "loss": 1.2098,
      "num_input_tokens_seen": 65896832,
      "step": 4096
    },
    {
      "epoch": 0.2869876627527209,
      "grad_norm": 4.145105838775635,
      "learning_rate": 7.132817513134852e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 65912960,
      "step": 4097
    },
    {
      "epoch": 0.2870577109984502,
      "grad_norm": 3.6198408603668213,
      "learning_rate": 7.1321176882662e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 65929344,
      "step": 4098
    },
    {
      "epoch": 0.28712775924417944,
      "grad_norm": 4.000823497772217,
      "learning_rate": 7.131417863397548e-05,
      "loss": 1.2109,
      "num_input_tokens_seen": 65945480,
      "step": 4099
    },
    {
      "epoch": 0.2871978074899087,
      "grad_norm": 4.2647271156311035,
      "learning_rate": 7.130718038528898e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 65961672,
      "step": 4100
    },
    {
      "epoch": 0.2872678557356379,
      "grad_norm": 4.704364776611328,
      "learning_rate": 7.130018213660246e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 65976848,
      "step": 4101
    },
    {
      "epoch": 0.28733790398136716,
      "grad_norm": 3.8795642852783203,
      "learning_rate": 7.129318388791593e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 65993120,
      "step": 4102
    },
    {
      "epoch": 0.2874079522270964,
      "grad_norm": 4.356956958770752,
      "learning_rate": 7.128618563922942e-05,
      "loss": 1.4218,
      "num_input_tokens_seen": 66008448,
      "step": 4103
    },
    {
      "epoch": 0.2874780004728257,
      "grad_norm": 3.5145177841186523,
      "learning_rate": 7.127918739054291e-05,
      "loss": 0.9055,
      "num_input_tokens_seen": 66024712,
      "step": 4104
    },
    {
      "epoch": 0.2875480487185549,
      "grad_norm": 3.7384872436523438,
      "learning_rate": 7.12721891418564e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 66041096,
      "step": 4105
    },
    {
      "epoch": 0.28761809696428414,
      "grad_norm": 3.9706084728240967,
      "learning_rate": 7.126519089316989e-05,
      "loss": 1.1538,
      "num_input_tokens_seen": 66056880,
      "step": 4106
    },
    {
      "epoch": 0.2876881452100134,
      "grad_norm": 3.692093849182129,
      "learning_rate": 7.125819264448337e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 66073264,
      "step": 4107
    },
    {
      "epoch": 0.28775819345574266,
      "grad_norm": 4.967808246612549,
      "learning_rate": 7.125119439579685e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 66089648,
      "step": 4108
    },
    {
      "epoch": 0.28782824170147187,
      "grad_norm": 3.8627805709838867,
      "learning_rate": 7.124419614711032e-05,
      "loss": 1.1056,
      "num_input_tokens_seen": 66105992,
      "step": 4109
    },
    {
      "epoch": 0.2878982899472011,
      "grad_norm": 3.7407474517822266,
      "learning_rate": 7.123719789842381e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 66122040,
      "step": 4110
    },
    {
      "epoch": 0.2879683381929304,
      "grad_norm": 4.028223514556885,
      "learning_rate": 7.123019964973732e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 66138056,
      "step": 4111
    },
    {
      "epoch": 0.28803838643865964,
      "grad_norm": 4.248149394989014,
      "learning_rate": 7.122320140105079e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 66154384,
      "step": 4112
    },
    {
      "epoch": 0.28810843468438885,
      "grad_norm": 3.49904465675354,
      "learning_rate": 7.121620315236428e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 66170016,
      "step": 4113
    },
    {
      "epoch": 0.2881784829301181,
      "grad_norm": 5.039339542388916,
      "learning_rate": 7.120920490367777e-05,
      "loss": 0.8658,
      "num_input_tokens_seen": 66185744,
      "step": 4114
    },
    {
      "epoch": 0.28824853117584737,
      "grad_norm": 3.800870656967163,
      "learning_rate": 7.120220665499124e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 66202128,
      "step": 4115
    },
    {
      "epoch": 0.2883185794215766,
      "grad_norm": 4.8073530197143555,
      "learning_rate": 7.119520840630473e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 66217840,
      "step": 4116
    },
    {
      "epoch": 0.28838862766730583,
      "grad_norm": 3.495415210723877,
      "learning_rate": 7.118821015761822e-05,
      "loss": 0.8693,
      "num_input_tokens_seen": 66234224,
      "step": 4117
    },
    {
      "epoch": 0.2884586759130351,
      "grad_norm": 4.46912956237793,
      "learning_rate": 7.118121190893171e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 66249968,
      "step": 4118
    },
    {
      "epoch": 0.28852872415876435,
      "grad_norm": 4.553129196166992,
      "learning_rate": 7.117421366024518e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 66265304,
      "step": 4119
    },
    {
      "epoch": 0.2885987724044936,
      "grad_norm": 3.713836193084717,
      "learning_rate": 7.116721541155867e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 66281680,
      "step": 4120
    },
    {
      "epoch": 0.28866882065022287,
      "grad_norm": 3.9745819568634033,
      "learning_rate": 7.116021716287216e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 66298064,
      "step": 4121
    },
    {
      "epoch": 0.28873886889595207,
      "grad_norm": 6.237453937530518,
      "learning_rate": 7.115321891418564e-05,
      "loss": 1.3598,
      "num_input_tokens_seen": 66314448,
      "step": 4122
    },
    {
      "epoch": 0.28880891714168133,
      "grad_norm": 3.7947497367858887,
      "learning_rate": 7.114622066549912e-05,
      "loss": 0.9342,
      "num_input_tokens_seen": 66330832,
      "step": 4123
    },
    {
      "epoch": 0.2888789653874106,
      "grad_norm": 5.574815273284912,
      "learning_rate": 7.113922241681261e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 66347216,
      "step": 4124
    },
    {
      "epoch": 0.28894901363313985,
      "grad_norm": 3.538344144821167,
      "learning_rate": 7.11322241681261e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 66363352,
      "step": 4125
    },
    {
      "epoch": 0.28901906187886905,
      "grad_norm": 3.792769193649292,
      "learning_rate": 7.112522591943958e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 66379736,
      "step": 4126
    },
    {
      "epoch": 0.2890891101245983,
      "grad_norm": 4.527935981750488,
      "learning_rate": 7.111822767075308e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 66396120,
      "step": 4127
    },
    {
      "epoch": 0.2891591583703276,
      "grad_norm": 3.753326416015625,
      "learning_rate": 7.111122942206655e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 66412424,
      "step": 4128
    },
    {
      "epoch": 0.28922920661605683,
      "grad_norm": 4.310519218444824,
      "learning_rate": 7.110423117338003e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 66428176,
      "step": 4129
    },
    {
      "epoch": 0.28929925486178604,
      "grad_norm": 3.9848945140838623,
      "learning_rate": 7.109723292469352e-05,
      "loss": 1.2687,
      "num_input_tokens_seen": 66444560,
      "step": 4130
    },
    {
      "epoch": 0.2893693031075153,
      "grad_norm": 4.654316425323486,
      "learning_rate": 7.109023467600702e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 66460944,
      "step": 4131
    },
    {
      "epoch": 0.28943935135324456,
      "grad_norm": 4.566670894622803,
      "learning_rate": 7.10832364273205e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 66475928,
      "step": 4132
    },
    {
      "epoch": 0.2895093995989738,
      "grad_norm": 4.4292988777160645,
      "learning_rate": 7.107623817863398e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 66491904,
      "step": 4133
    },
    {
      "epoch": 0.289579447844703,
      "grad_norm": 6.520173072814941,
      "learning_rate": 7.106923992994747e-05,
      "loss": 0.9938,
      "num_input_tokens_seen": 66507256,
      "step": 4134
    },
    {
      "epoch": 0.2896494960904323,
      "grad_norm": 3.8424220085144043,
      "learning_rate": 7.106224168126095e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 66522736,
      "step": 4135
    },
    {
      "epoch": 0.28971954433616154,
      "grad_norm": 4.742796897888184,
      "learning_rate": 7.105524343257442e-05,
      "loss": 1.0296,
      "num_input_tokens_seen": 66538480,
      "step": 4136
    },
    {
      "epoch": 0.2897895925818908,
      "grad_norm": 3.552365779876709,
      "learning_rate": 7.104824518388792e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 66554576,
      "step": 4137
    },
    {
      "epoch": 0.28985964082762,
      "grad_norm": 6.649835109710693,
      "learning_rate": 7.104124693520141e-05,
      "loss": 0.9729,
      "num_input_tokens_seen": 66570000,
      "step": 4138
    },
    {
      "epoch": 0.28992968907334926,
      "grad_norm": 3.9890356063842773,
      "learning_rate": 7.103424868651489e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 66585640,
      "step": 4139
    },
    {
      "epoch": 0.2899997373190785,
      "grad_norm": 3.80637526512146,
      "learning_rate": 7.102725043782838e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 66601696,
      "step": 4140
    },
    {
      "epoch": 0.2900697855648078,
      "grad_norm": 4.089916706085205,
      "learning_rate": 7.102025218914186e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 66618080,
      "step": 4141
    },
    {
      "epoch": 0.290139833810537,
      "grad_norm": 3.2609710693359375,
      "learning_rate": 7.101325394045534e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 66634216,
      "step": 4142
    },
    {
      "epoch": 0.29020988205626624,
      "grad_norm": 4.3664093017578125,
      "learning_rate": 7.100625569176883e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 66650600,
      "step": 4143
    },
    {
      "epoch": 0.2902799303019955,
      "grad_norm": 4.460801124572754,
      "learning_rate": 7.099925744308232e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 66666592,
      "step": 4144
    },
    {
      "epoch": 0.29034997854772476,
      "grad_norm": 4.474677562713623,
      "learning_rate": 7.09922591943958e-05,
      "loss": 1.0016,
      "num_input_tokens_seen": 66681544,
      "step": 4145
    },
    {
      "epoch": 0.29042002679345397,
      "grad_norm": 3.6482129096984863,
      "learning_rate": 7.098526094570928e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 66697928,
      "step": 4146
    },
    {
      "epoch": 0.2904900750391832,
      "grad_norm": 3.483290195465088,
      "learning_rate": 7.097826269702277e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 66714312,
      "step": 4147
    },
    {
      "epoch": 0.2905601232849125,
      "grad_norm": 4.703539848327637,
      "learning_rate": 7.097126444833626e-05,
      "loss": 0.9718,
      "num_input_tokens_seen": 66729632,
      "step": 4148
    },
    {
      "epoch": 0.29063017153064175,
      "grad_norm": 3.8614907264709473,
      "learning_rate": 7.096426619964973e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 66746016,
      "step": 4149
    },
    {
      "epoch": 0.29070021977637095,
      "grad_norm": 3.612683057785034,
      "learning_rate": 7.095726795096322e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 66762400,
      "step": 4150
    },
    {
      "epoch": 0.2907702680221002,
      "grad_norm": 3.980149984359741,
      "learning_rate": 7.095026970227672e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 66778392,
      "step": 4151
    },
    {
      "epoch": 0.29084031626782947,
      "grad_norm": 3.857588052749634,
      "learning_rate": 7.09432714535902e-05,
      "loss": 1.0506,
      "num_input_tokens_seen": 66794200,
      "step": 4152
    },
    {
      "epoch": 0.29091036451355873,
      "grad_norm": 5.106949806213379,
      "learning_rate": 7.093627320490367e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 66810584,
      "step": 4153
    },
    {
      "epoch": 0.29098041275928793,
      "grad_norm": 4.338438987731934,
      "learning_rate": 7.092927495621718e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 66826208,
      "step": 4154
    },
    {
      "epoch": 0.2910504610050172,
      "grad_norm": 3.962877035140991,
      "learning_rate": 7.092227670753065e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 66842592,
      "step": 4155
    },
    {
      "epoch": 0.29112050925074645,
      "grad_norm": 3.8490965366363525,
      "learning_rate": 7.091527845884413e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 66858832,
      "step": 4156
    },
    {
      "epoch": 0.2911905574964757,
      "grad_norm": 4.559625148773193,
      "learning_rate": 7.090828021015763e-05,
      "loss": 1.3951,
      "num_input_tokens_seen": 66875216,
      "step": 4157
    },
    {
      "epoch": 0.29126060574220497,
      "grad_norm": 8.37543773651123,
      "learning_rate": 7.090128196147112e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 66891600,
      "step": 4158
    },
    {
      "epoch": 0.2913306539879342,
      "grad_norm": 4.128559112548828,
      "learning_rate": 7.089428371278459e-05,
      "loss": 0.8789,
      "num_input_tokens_seen": 66907984,
      "step": 4159
    },
    {
      "epoch": 0.29140070223366343,
      "grad_norm": 4.81403112411499,
      "learning_rate": 7.088728546409808e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 66923240,
      "step": 4160
    },
    {
      "epoch": 0.2914707504793927,
      "grad_norm": 4.534300804138184,
      "learning_rate": 7.088028721541157e-05,
      "loss": 0.8906,
      "num_input_tokens_seen": 66939624,
      "step": 4161
    },
    {
      "epoch": 0.29154079872512195,
      "grad_norm": 4.46708869934082,
      "learning_rate": 7.087328896672504e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 66955968,
      "step": 4162
    },
    {
      "epoch": 0.29161084697085116,
      "grad_norm": 4.142822265625,
      "learning_rate": 7.086629071803853e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 66971680,
      "step": 4163
    },
    {
      "epoch": 0.2916808952165804,
      "grad_norm": 3.686167001724243,
      "learning_rate": 7.085929246935202e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 66987952,
      "step": 4164
    },
    {
      "epoch": 0.2917509434623097,
      "grad_norm": 8.076430320739746,
      "learning_rate": 7.085229422066551e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 67004336,
      "step": 4165
    },
    {
      "epoch": 0.29182099170803893,
      "grad_norm": 8.69857120513916,
      "learning_rate": 7.084529597197898e-05,
      "loss": 1.2295,
      "num_input_tokens_seen": 67020216,
      "step": 4166
    },
    {
      "epoch": 0.29189103995376814,
      "grad_norm": 3.7867684364318848,
      "learning_rate": 7.083829772329247e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 67035600,
      "step": 4167
    },
    {
      "epoch": 0.2919610881994974,
      "grad_norm": 5.560591697692871,
      "learning_rate": 7.083129947460596e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 67051680,
      "step": 4168
    },
    {
      "epoch": 0.29203113644522666,
      "grad_norm": 3.857120990753174,
      "learning_rate": 7.082430122591944e-05,
      "loss": 1.1991,
      "num_input_tokens_seen": 67068064,
      "step": 4169
    },
    {
      "epoch": 0.2921011846909559,
      "grad_norm": 4.343360900878906,
      "learning_rate": 7.081730297723293e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 67084448,
      "step": 4170
    },
    {
      "epoch": 0.2921712329366851,
      "grad_norm": 4.198531150817871,
      "learning_rate": 7.081030472854643e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 67100832,
      "step": 4171
    },
    {
      "epoch": 0.2922412811824144,
      "grad_norm": 3.539684772491455,
      "learning_rate": 7.08033064798599e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 67117216,
      "step": 4172
    },
    {
      "epoch": 0.29231132942814364,
      "grad_norm": 4.2374444007873535,
      "learning_rate": 7.079630823117338e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 67133600,
      "step": 4173
    },
    {
      "epoch": 0.2923813776738729,
      "grad_norm": 4.106996059417725,
      "learning_rate": 7.078930998248687e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 67149984,
      "step": 4174
    },
    {
      "epoch": 0.2924514259196021,
      "grad_norm": 3.7100484371185303,
      "learning_rate": 7.078231173380035e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 67166168,
      "step": 4175
    },
    {
      "epoch": 0.29252147416533136,
      "grad_norm": 5.189118385314941,
      "learning_rate": 7.077531348511383e-05,
      "loss": 0.9642,
      "num_input_tokens_seen": 67181472,
      "step": 4176
    },
    {
      "epoch": 0.2925915224110606,
      "grad_norm": 4.540155410766602,
      "learning_rate": 7.076831523642733e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 67197856,
      "step": 4177
    },
    {
      "epoch": 0.2926615706567899,
      "grad_norm": 4.748345375061035,
      "learning_rate": 7.076131698774082e-05,
      "loss": 0.8845,
      "num_input_tokens_seen": 67214240,
      "step": 4178
    },
    {
      "epoch": 0.2927316189025191,
      "grad_norm": 4.252089023590088,
      "learning_rate": 7.07543187390543e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 67230312,
      "step": 4179
    },
    {
      "epoch": 0.29280166714824835,
      "grad_norm": 4.273370742797852,
      "learning_rate": 7.074732049036777e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 67246152,
      "step": 4180
    },
    {
      "epoch": 0.2928717153939776,
      "grad_norm": 3.9271481037139893,
      "learning_rate": 7.074032224168127e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 67261688,
      "step": 4181
    },
    {
      "epoch": 0.29294176363970686,
      "grad_norm": 3.875622034072876,
      "learning_rate": 7.073332399299475e-05,
      "loss": 1.2345,
      "num_input_tokens_seen": 67278072,
      "step": 4182
    },
    {
      "epoch": 0.29301181188543607,
      "grad_norm": 3.8089005947113037,
      "learning_rate": 7.072632574430824e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 67293760,
      "step": 4183
    },
    {
      "epoch": 0.29308186013116533,
      "grad_norm": 4.402803421020508,
      "learning_rate": 7.071932749562172e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 67310144,
      "step": 4184
    },
    {
      "epoch": 0.2931519083768946,
      "grad_norm": 4.4534783363342285,
      "learning_rate": 7.071232924693521e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 67326528,
      "step": 4185
    },
    {
      "epoch": 0.29322195662262385,
      "grad_norm": 4.247747898101807,
      "learning_rate": 7.070533099824869e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 67342080,
      "step": 4186
    },
    {
      "epoch": 0.29329200486835305,
      "grad_norm": 5.280468463897705,
      "learning_rate": 7.069833274956218e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 67357168,
      "step": 4187
    },
    {
      "epoch": 0.2933620531140823,
      "grad_norm": 5.14320707321167,
      "learning_rate": 7.069133450087567e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 67373552,
      "step": 4188
    },
    {
      "epoch": 0.29343210135981157,
      "grad_norm": 4.131645679473877,
      "learning_rate": 7.068433625218914e-05,
      "loss": 1.3795,
      "num_input_tokens_seen": 67389936,
      "step": 4189
    },
    {
      "epoch": 0.29350214960554083,
      "grad_norm": 4.727990627288818,
      "learning_rate": 7.067733800350263e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 67406320,
      "step": 4190
    },
    {
      "epoch": 0.2935721978512701,
      "grad_norm": 5.857666969299316,
      "learning_rate": 7.067033975481612e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 67422680,
      "step": 4191
    },
    {
      "epoch": 0.2936422460969993,
      "grad_norm": 4.185948371887207,
      "learning_rate": 7.06633415061296e-05,
      "loss": 1.2738,
      "num_input_tokens_seen": 67439064,
      "step": 4192
    },
    {
      "epoch": 0.29371229434272855,
      "grad_norm": 3.749274969100952,
      "learning_rate": 7.065634325744308e-05,
      "loss": 1.0327,
      "num_input_tokens_seen": 67454680,
      "step": 4193
    },
    {
      "epoch": 0.2937823425884578,
      "grad_norm": 4.332368850708008,
      "learning_rate": 7.064934500875657e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 67470800,
      "step": 4194
    },
    {
      "epoch": 0.29385239083418707,
      "grad_norm": 5.514054775238037,
      "learning_rate": 7.064234676007006e-05,
      "loss": 1.2602,
      "num_input_tokens_seen": 67487184,
      "step": 4195
    },
    {
      "epoch": 0.2939224390799163,
      "grad_norm": 4.534146785736084,
      "learning_rate": 7.063534851138353e-05,
      "loss": 1.2929,
      "num_input_tokens_seen": 67503504,
      "step": 4196
    },
    {
      "epoch": 0.29399248732564554,
      "grad_norm": 4.86776876449585,
      "learning_rate": 7.062835026269702e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 67519056,
      "step": 4197
    },
    {
      "epoch": 0.2940625355713748,
      "grad_norm": 3.8528504371643066,
      "learning_rate": 7.062135201401052e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 67535440,
      "step": 4198
    },
    {
      "epoch": 0.29413258381710405,
      "grad_norm": 4.244069576263428,
      "learning_rate": 7.0614353765324e-05,
      "loss": 1.1733,
      "num_input_tokens_seen": 67551264,
      "step": 4199
    },
    {
      "epoch": 0.29420263206283326,
      "grad_norm": 3.5963211059570312,
      "learning_rate": 7.060735551663747e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 67567648,
      "step": 4200
    },
    {
      "epoch": 0.29420263206283326,
      "eval_loss": 1.1331984996795654,
      "eval_runtime": 0.203,
      "eval_samples_per_second": 4.927,
      "eval_steps_per_second": 4.927,
      "num_input_tokens_seen": 67567648,
      "step": 4200
    },
    {
      "epoch": 0.2942726803085625,
      "grad_norm": 4.51765775680542,
      "learning_rate": 7.060035726795096e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 67583792,
      "step": 4201
    },
    {
      "epoch": 0.2943427285542918,
      "grad_norm": 4.541067123413086,
      "learning_rate": 7.059335901926445e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 67599856,
      "step": 4202
    },
    {
      "epoch": 0.29441277680002104,
      "grad_norm": 4.095570087432861,
      "learning_rate": 7.058636077057794e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 67616240,
      "step": 4203
    },
    {
      "epoch": 0.29448282504575024,
      "grad_norm": 4.616795539855957,
      "learning_rate": 7.057936252189143e-05,
      "loss": 1.2549,
      "num_input_tokens_seen": 67632496,
      "step": 4204
    },
    {
      "epoch": 0.2945528732914795,
      "grad_norm": 3.8619420528411865,
      "learning_rate": 7.057236427320492e-05,
      "loss": 0.9626,
      "num_input_tokens_seen": 67648880,
      "step": 4205
    },
    {
      "epoch": 0.29462292153720876,
      "grad_norm": 4.194519996643066,
      "learning_rate": 7.056536602451839e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 67665264,
      "step": 4206
    },
    {
      "epoch": 0.294692969782938,
      "grad_norm": 4.835122585296631,
      "learning_rate": 7.055836777583187e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 67681648,
      "step": 4207
    },
    {
      "epoch": 0.2947630180286672,
      "grad_norm": 4.2085280418396,
      "learning_rate": 7.055136952714537e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 67697960,
      "step": 4208
    },
    {
      "epoch": 0.2948330662743965,
      "grad_norm": 4.439855575561523,
      "learning_rate": 7.054437127845884e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 67714344,
      "step": 4209
    },
    {
      "epoch": 0.29490311452012574,
      "grad_norm": 5.427484035491943,
      "learning_rate": 7.053737302977233e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 67730728,
      "step": 4210
    },
    {
      "epoch": 0.294973162765855,
      "grad_norm": 3.6627275943756104,
      "learning_rate": 7.053037478108582e-05,
      "loss": 0.8507,
      "num_input_tokens_seen": 67746704,
      "step": 4211
    },
    {
      "epoch": 0.2950432110115842,
      "grad_norm": 4.450380325317383,
      "learning_rate": 7.052337653239931e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 67762320,
      "step": 4212
    },
    {
      "epoch": 0.29511325925731346,
      "grad_norm": 3.6644749641418457,
      "learning_rate": 7.051637828371279e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 67778704,
      "step": 4213
    },
    {
      "epoch": 0.2951833075030427,
      "grad_norm": 4.331392288208008,
      "learning_rate": 7.050938003502627e-05,
      "loss": 1.0854,
      "num_input_tokens_seen": 67795088,
      "step": 4214
    },
    {
      "epoch": 0.295253355748772,
      "grad_norm": 4.157777786254883,
      "learning_rate": 7.050238178633976e-05,
      "loss": 1.2039,
      "num_input_tokens_seen": 67811472,
      "step": 4215
    },
    {
      "epoch": 0.2953234039945012,
      "grad_norm": 3.858069896697998,
      "learning_rate": 7.049538353765324e-05,
      "loss": 1.1751,
      "num_input_tokens_seen": 67827488,
      "step": 4216
    },
    {
      "epoch": 0.29539345224023045,
      "grad_norm": 4.279262542724609,
      "learning_rate": 7.048838528896673e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 67843872,
      "step": 4217
    },
    {
      "epoch": 0.2954635004859597,
      "grad_norm": 4.539918422698975,
      "learning_rate": 7.048138704028021e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 67860256,
      "step": 4218
    },
    {
      "epoch": 0.29553354873168897,
      "grad_norm": 3.738811492919922,
      "learning_rate": 7.04743887915937e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 67876224,
      "step": 4219
    },
    {
      "epoch": 0.29560359697741817,
      "grad_norm": 4.634495258331299,
      "learning_rate": 7.046739054290718e-05,
      "loss": 0.9273,
      "num_input_tokens_seen": 67892040,
      "step": 4220
    },
    {
      "epoch": 0.29567364522314743,
      "grad_norm": 5.988262176513672,
      "learning_rate": 7.046039229422067e-05,
      "loss": 0.956,
      "num_input_tokens_seen": 67908424,
      "step": 4221
    },
    {
      "epoch": 0.2957436934688767,
      "grad_norm": 7.2220258712768555,
      "learning_rate": 7.045339404553416e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 67924808,
      "step": 4222
    },
    {
      "epoch": 0.29581374171460595,
      "grad_norm": 8.866394996643066,
      "learning_rate": 7.044639579684764e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 67941192,
      "step": 4223
    },
    {
      "epoch": 0.2958837899603352,
      "grad_norm": 4.791526794433594,
      "learning_rate": 7.043939754816112e-05,
      "loss": 1.1966,
      "num_input_tokens_seen": 67957576,
      "step": 4224
    },
    {
      "epoch": 0.2959538382060644,
      "grad_norm": 3.8345704078674316,
      "learning_rate": 7.043239929947462e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 67973112,
      "step": 4225
    },
    {
      "epoch": 0.29602388645179367,
      "grad_norm": 5.0572099685668945,
      "learning_rate": 7.04254010507881e-05,
      "loss": 1.3761,
      "num_input_tokens_seen": 67989360,
      "step": 4226
    },
    {
      "epoch": 0.29609393469752293,
      "grad_norm": 4.467088222503662,
      "learning_rate": 7.041840280210157e-05,
      "loss": 0.981,
      "num_input_tokens_seen": 68005744,
      "step": 4227
    },
    {
      "epoch": 0.2961639829432522,
      "grad_norm": 6.415910243988037,
      "learning_rate": 7.041140455341506e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 68021592,
      "step": 4228
    },
    {
      "epoch": 0.2962340311889814,
      "grad_norm": 4.432079315185547,
      "learning_rate": 7.040440630472855e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 68037976,
      "step": 4229
    },
    {
      "epoch": 0.29630407943471065,
      "grad_norm": 4.207062721252441,
      "learning_rate": 7.039740805604204e-05,
      "loss": 1.2702,
      "num_input_tokens_seen": 68054328,
      "step": 4230
    },
    {
      "epoch": 0.2963741276804399,
      "grad_norm": 4.825972557067871,
      "learning_rate": 7.039040980735553e-05,
      "loss": 1.3091,
      "num_input_tokens_seen": 68070416,
      "step": 4231
    },
    {
      "epoch": 0.2964441759261692,
      "grad_norm": 3.917593002319336,
      "learning_rate": 7.038341155866901e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 68086800,
      "step": 4232
    },
    {
      "epoch": 0.2965142241718984,
      "grad_norm": 3.8865675926208496,
      "learning_rate": 7.037641330998249e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 68103184,
      "step": 4233
    },
    {
      "epoch": 0.29658427241762764,
      "grad_norm": 3.8321971893310547,
      "learning_rate": 7.036941506129596e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 68119568,
      "step": 4234
    },
    {
      "epoch": 0.2966543206633569,
      "grad_norm": 5.020960807800293,
      "learning_rate": 7.036241681260947e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 68135416,
      "step": 4235
    },
    {
      "epoch": 0.29672436890908616,
      "grad_norm": 3.653468608856201,
      "learning_rate": 7.035541856392294e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 68151800,
      "step": 4236
    },
    {
      "epoch": 0.29679441715481536,
      "grad_norm": 3.8133575916290283,
      "learning_rate": 7.034842031523643e-05,
      "loss": 1.0867,
      "num_input_tokens_seen": 68168184,
      "step": 4237
    },
    {
      "epoch": 0.2968644654005446,
      "grad_norm": 3.6642141342163086,
      "learning_rate": 7.034142206654992e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 68184080,
      "step": 4238
    },
    {
      "epoch": 0.2969345136462739,
      "grad_norm": 4.362963676452637,
      "learning_rate": 7.033442381786341e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 68199928,
      "step": 4239
    },
    {
      "epoch": 0.29700456189200314,
      "grad_norm": 3.6831562519073486,
      "learning_rate": 7.032742556917688e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 68215952,
      "step": 4240
    },
    {
      "epoch": 0.29707461013773234,
      "grad_norm": 4.906534194946289,
      "learning_rate": 7.032042732049037e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 68232336,
      "step": 4241
    },
    {
      "epoch": 0.2971446583834616,
      "grad_norm": 3.446749687194824,
      "learning_rate": 7.031342907180386e-05,
      "loss": 0.8306,
      "num_input_tokens_seen": 68247832,
      "step": 4242
    },
    {
      "epoch": 0.29721470662919086,
      "grad_norm": 4.729014873504639,
      "learning_rate": 7.030643082311735e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 68264216,
      "step": 4243
    },
    {
      "epoch": 0.2972847548749201,
      "grad_norm": 4.196920871734619,
      "learning_rate": 7.029943257443082e-05,
      "loss": 1.1496,
      "num_input_tokens_seen": 68280600,
      "step": 4244
    },
    {
      "epoch": 0.2973548031206493,
      "grad_norm": 7.193357467651367,
      "learning_rate": 7.029243432574431e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 68296984,
      "step": 4245
    },
    {
      "epoch": 0.2974248513663786,
      "grad_norm": 4.00344181060791,
      "learning_rate": 7.02854360770578e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 68312720,
      "step": 4246
    },
    {
      "epoch": 0.29749489961210784,
      "grad_norm": 4.04103422164917,
      "learning_rate": 7.027843782837128e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 68328608,
      "step": 4247
    },
    {
      "epoch": 0.2975649478578371,
      "grad_norm": 4.010391712188721,
      "learning_rate": 7.027143957968476e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 68343288,
      "step": 4248
    },
    {
      "epoch": 0.2976349961035663,
      "grad_norm": 6.364760398864746,
      "learning_rate": 7.026444133099825e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 68359672,
      "step": 4249
    },
    {
      "epoch": 0.29770504434929557,
      "grad_norm": 5.682034969329834,
      "learning_rate": 7.025744308231174e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 68376056,
      "step": 4250
    },
    {
      "epoch": 0.2977750925950248,
      "grad_norm": 3.6160550117492676,
      "learning_rate": 7.025044483362522e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 68392440,
      "step": 4251
    },
    {
      "epoch": 0.2978451408407541,
      "grad_norm": 4.839343070983887,
      "learning_rate": 7.024344658493872e-05,
      "loss": 0.9924,
      "num_input_tokens_seen": 68408608,
      "step": 4252
    },
    {
      "epoch": 0.2979151890864833,
      "grad_norm": 5.255819320678711,
      "learning_rate": 7.02364483362522e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 68424944,
      "step": 4253
    },
    {
      "epoch": 0.29798523733221255,
      "grad_norm": 3.7549142837524414,
      "learning_rate": 7.022945008756567e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 68441328,
      "step": 4254
    },
    {
      "epoch": 0.2980552855779418,
      "grad_norm": 5.159091472625732,
      "learning_rate": 7.022245183887916e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 68457712,
      "step": 4255
    },
    {
      "epoch": 0.29812533382367107,
      "grad_norm": 3.8031342029571533,
      "learning_rate": 7.021545359019265e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 68474072,
      "step": 4256
    },
    {
      "epoch": 0.29819538206940027,
      "grad_norm": 6.039318084716797,
      "learning_rate": 7.020845534150613e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 68490456,
      "step": 4257
    },
    {
      "epoch": 0.29826543031512953,
      "grad_norm": 3.9376237392425537,
      "learning_rate": 7.020145709281962e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 68506760,
      "step": 4258
    },
    {
      "epoch": 0.2983354785608588,
      "grad_norm": 4.599661827087402,
      "learning_rate": 7.019445884413311e-05,
      "loss": 0.9722,
      "num_input_tokens_seen": 68523144,
      "step": 4259
    },
    {
      "epoch": 0.29840552680658805,
      "grad_norm": 3.743640661239624,
      "learning_rate": 7.018746059544659e-05,
      "loss": 1.157,
      "num_input_tokens_seen": 68539448,
      "step": 4260
    },
    {
      "epoch": 0.2984755750523173,
      "grad_norm": 6.111955642700195,
      "learning_rate": 7.018046234676006e-05,
      "loss": 1.2148,
      "num_input_tokens_seen": 68555832,
      "step": 4261
    },
    {
      "epoch": 0.2985456232980465,
      "grad_norm": 4.297199249267578,
      "learning_rate": 7.017346409807356e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 68572216,
      "step": 4262
    },
    {
      "epoch": 0.2986156715437758,
      "grad_norm": 4.126640319824219,
      "learning_rate": 7.016646584938705e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 68588600,
      "step": 4263
    },
    {
      "epoch": 0.29868571978950503,
      "grad_norm": 3.8142640590667725,
      "learning_rate": 7.015946760070053e-05,
      "loss": 1.1943,
      "num_input_tokens_seen": 68604336,
      "step": 4264
    },
    {
      "epoch": 0.2987557680352343,
      "grad_norm": 3.9500539302825928,
      "learning_rate": 7.015246935201402e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 68620056,
      "step": 4265
    },
    {
      "epoch": 0.2988258162809635,
      "grad_norm": 4.431976318359375,
      "learning_rate": 7.01454711033275e-05,
      "loss": 1.3419,
      "num_input_tokens_seen": 68636328,
      "step": 4266
    },
    {
      "epoch": 0.29889586452669276,
      "grad_norm": 5.619480609893799,
      "learning_rate": 7.013847285464098e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 68651984,
      "step": 4267
    },
    {
      "epoch": 0.298965912772422,
      "grad_norm": 3.8473827838897705,
      "learning_rate": 7.013147460595447e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 68668176,
      "step": 4268
    },
    {
      "epoch": 0.2990359610181513,
      "grad_norm": 5.942142486572266,
      "learning_rate": 7.012447635726796e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 68684560,
      "step": 4269
    },
    {
      "epoch": 0.2991060092638805,
      "grad_norm": 6.194666862487793,
      "learning_rate": 7.011747810858145e-05,
      "loss": 0.9782,
      "num_input_tokens_seen": 68699816,
      "step": 4270
    },
    {
      "epoch": 0.29917605750960974,
      "grad_norm": 4.336294651031494,
      "learning_rate": 7.011047985989492e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 68716200,
      "step": 4271
    },
    {
      "epoch": 0.299246105755339,
      "grad_norm": 4.277907371520996,
      "learning_rate": 7.010348161120841e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 68732584,
      "step": 4272
    },
    {
      "epoch": 0.29931615400106826,
      "grad_norm": 5.045118808746338,
      "learning_rate": 7.00964833625219e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 68748840,
      "step": 4273
    },
    {
      "epoch": 0.29938620224679746,
      "grad_norm": 4.3978400230407715,
      "learning_rate": 7.008948511383537e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 68765224,
      "step": 4274
    },
    {
      "epoch": 0.2994562504925267,
      "grad_norm": 5.052615165710449,
      "learning_rate": 7.008248686514886e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 68780808,
      "step": 4275
    },
    {
      "epoch": 0.299526298738256,
      "grad_norm": 6.902999401092529,
      "learning_rate": 7.007548861646235e-05,
      "loss": 1.0952,
      "num_input_tokens_seen": 68797064,
      "step": 4276
    },
    {
      "epoch": 0.29959634698398524,
      "grad_norm": 5.947190761566162,
      "learning_rate": 7.006849036777584e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 68812904,
      "step": 4277
    },
    {
      "epoch": 0.29966639522971444,
      "grad_norm": 5.443974018096924,
      "learning_rate": 7.006149211908931e-05,
      "loss": 1.146,
      "num_input_tokens_seen": 68828736,
      "step": 4278
    },
    {
      "epoch": 0.2997364434754437,
      "grad_norm": 3.9849112033843994,
      "learning_rate": 7.005449387040282e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 68843928,
      "step": 4279
    },
    {
      "epoch": 0.29980649172117296,
      "grad_norm": 5.787483215332031,
      "learning_rate": 7.004749562171629e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 68860312,
      "step": 4280
    },
    {
      "epoch": 0.2998765399669022,
      "grad_norm": 3.8437387943267822,
      "learning_rate": 7.004049737302977e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 68876696,
      "step": 4281
    },
    {
      "epoch": 0.2999465882126314,
      "grad_norm": 3.94879150390625,
      "learning_rate": 7.003349912434325e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 68893080,
      "step": 4282
    },
    {
      "epoch": 0.3000166364583607,
      "grad_norm": 4.746649265289307,
      "learning_rate": 7.002650087565676e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 68909464,
      "step": 4283
    },
    {
      "epoch": 0.30008668470408995,
      "grad_norm": 4.1024861335754395,
      "learning_rate": 7.001950262697023e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 68925352,
      "step": 4284
    },
    {
      "epoch": 0.3001567329498192,
      "grad_norm": 4.5073699951171875,
      "learning_rate": 7.001250437828372e-05,
      "loss": 1.2092,
      "num_input_tokens_seen": 68941736,
      "step": 4285
    },
    {
      "epoch": 0.3002267811955484,
      "grad_norm": 4.947534561157227,
      "learning_rate": 7.000550612959721e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 68958120,
      "step": 4286
    },
    {
      "epoch": 0.30029682944127767,
      "grad_norm": 3.8399429321289062,
      "learning_rate": 6.999850788091068e-05,
      "loss": 1.1268,
      "num_input_tokens_seen": 68974232,
      "step": 4287
    },
    {
      "epoch": 0.30036687768700693,
      "grad_norm": 3.9180405139923096,
      "learning_rate": 6.999150963222416e-05,
      "loss": 1.1666,
      "num_input_tokens_seen": 68990616,
      "step": 4288
    },
    {
      "epoch": 0.3004369259327362,
      "grad_norm": 3.9542794227600098,
      "learning_rate": 6.998451138353766e-05,
      "loss": 1.1474,
      "num_input_tokens_seen": 69006952,
      "step": 4289
    },
    {
      "epoch": 0.3005069741784654,
      "grad_norm": 3.5275325775146484,
      "learning_rate": 6.997751313485115e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 69023336,
      "step": 4290
    },
    {
      "epoch": 0.30057702242419465,
      "grad_norm": 3.9485349655151367,
      "learning_rate": 6.997051488616462e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 69038392,
      "step": 4291
    },
    {
      "epoch": 0.3006470706699239,
      "grad_norm": 3.4944114685058594,
      "learning_rate": 6.996351663747811e-05,
      "loss": 0.7473,
      "num_input_tokens_seen": 69054160,
      "step": 4292
    },
    {
      "epoch": 0.30071711891565317,
      "grad_norm": 3.387148380279541,
      "learning_rate": 6.99565183887916e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 69070056,
      "step": 4293
    },
    {
      "epoch": 0.30078716716138243,
      "grad_norm": 3.9591586589813232,
      "learning_rate": 6.994952014010508e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 69086240,
      "step": 4294
    },
    {
      "epoch": 0.30085721540711163,
      "grad_norm": 8.32682991027832,
      "learning_rate": 6.994252189141857e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 69102408,
      "step": 4295
    },
    {
      "epoch": 0.3009272636528409,
      "grad_norm": 3.5885214805603027,
      "learning_rate": 6.993552364273205e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 69118376,
      "step": 4296
    },
    {
      "epoch": 0.30099731189857015,
      "grad_norm": 4.784765243530273,
      "learning_rate": 6.992852539404554e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 69133664,
      "step": 4297
    },
    {
      "epoch": 0.3010673601442994,
      "grad_norm": 6.456319808959961,
      "learning_rate": 6.992152714535902e-05,
      "loss": 1.3836,
      "num_input_tokens_seen": 69148224,
      "step": 4298
    },
    {
      "epoch": 0.3011374083900286,
      "grad_norm": 5.820954322814941,
      "learning_rate": 6.99145288966725e-05,
      "loss": 0.9987,
      "num_input_tokens_seen": 69164440,
      "step": 4299
    },
    {
      "epoch": 0.3012074566357579,
      "grad_norm": 6.690483570098877,
      "learning_rate": 6.9907530647986e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 69180824,
      "step": 4300
    },
    {
      "epoch": 0.30127750488148713,
      "grad_norm": 3.8018131256103516,
      "learning_rate": 6.990053239929947e-05,
      "loss": 1.1643,
      "num_input_tokens_seen": 69197016,
      "step": 4301
    },
    {
      "epoch": 0.3013475531272164,
      "grad_norm": 4.574918746948242,
      "learning_rate": 6.989353415061296e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 69213400,
      "step": 4302
    },
    {
      "epoch": 0.3014176013729456,
      "grad_norm": 3.3843026161193848,
      "learning_rate": 6.988653590192646e-05,
      "loss": 0.9762,
      "num_input_tokens_seen": 69229784,
      "step": 4303
    },
    {
      "epoch": 0.30148764961867486,
      "grad_norm": 6.179981708526611,
      "learning_rate": 6.987953765323994e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 69246168,
      "step": 4304
    },
    {
      "epoch": 0.3015576978644041,
      "grad_norm": 4.759994029998779,
      "learning_rate": 6.987253940455341e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 69262512,
      "step": 4305
    },
    {
      "epoch": 0.3016277461101334,
      "grad_norm": 3.719902992248535,
      "learning_rate": 6.986554115586691e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 69278496,
      "step": 4306
    },
    {
      "epoch": 0.3016977943558626,
      "grad_norm": 3.6757240295410156,
      "learning_rate": 6.985854290718039e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 69294880,
      "step": 4307
    },
    {
      "epoch": 0.30176784260159184,
      "grad_norm": 4.316056251525879,
      "learning_rate": 6.985154465849386e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 69311264,
      "step": 4308
    },
    {
      "epoch": 0.3018378908473211,
      "grad_norm": 5.248560428619385,
      "learning_rate": 6.984454640980736e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 69327648,
      "step": 4309
    },
    {
      "epoch": 0.30190793909305036,
      "grad_norm": 3.601381540298462,
      "learning_rate": 6.983754816112085e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 69344032,
      "step": 4310
    },
    {
      "epoch": 0.30197798733877956,
      "grad_norm": 4.555902004241943,
      "learning_rate": 6.983054991243433e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 69360416,
      "step": 4311
    },
    {
      "epoch": 0.3020480355845088,
      "grad_norm": 4.615258693695068,
      "learning_rate": 6.982355166374782e-05,
      "loss": 1.1759,
      "num_input_tokens_seen": 69375728,
      "step": 4312
    },
    {
      "epoch": 0.3021180838302381,
      "grad_norm": 5.953250408172607,
      "learning_rate": 6.98165534150613e-05,
      "loss": 1.1161,
      "num_input_tokens_seen": 69391768,
      "step": 4313
    },
    {
      "epoch": 0.30218813207596734,
      "grad_norm": 4.049426555633545,
      "learning_rate": 6.980955516637478e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 69407328,
      "step": 4314
    },
    {
      "epoch": 0.30225818032169655,
      "grad_norm": 4.012260437011719,
      "learning_rate": 6.980255691768827e-05,
      "loss": 1.338,
      "num_input_tokens_seen": 69423712,
      "step": 4315
    },
    {
      "epoch": 0.3023282285674258,
      "grad_norm": 3.8932242393493652,
      "learning_rate": 6.979555866900176e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 69440096,
      "step": 4316
    },
    {
      "epoch": 0.30239827681315506,
      "grad_norm": 7.58411169052124,
      "learning_rate": 6.978856042031525e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 69456088,
      "step": 4317
    },
    {
      "epoch": 0.3024683250588843,
      "grad_norm": 5.275664806365967,
      "learning_rate": 6.978156217162872e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 69471768,
      "step": 4318
    },
    {
      "epoch": 0.30253837330461353,
      "grad_norm": 3.6384737491607666,
      "learning_rate": 6.977456392294221e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 69488152,
      "step": 4319
    },
    {
      "epoch": 0.3026084215503428,
      "grad_norm": 5.059805870056152,
      "learning_rate": 6.97675656742557e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 69504536,
      "step": 4320
    },
    {
      "epoch": 0.30267846979607205,
      "grad_norm": 5.672605037689209,
      "learning_rate": 6.976056742556917e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 69520920,
      "step": 4321
    },
    {
      "epoch": 0.3027485180418013,
      "grad_norm": 3.5066421031951904,
      "learning_rate": 6.975356917688266e-05,
      "loss": 1.1437,
      "num_input_tokens_seen": 69537304,
      "step": 4322
    },
    {
      "epoch": 0.3028185662875305,
      "grad_norm": 4.403011798858643,
      "learning_rate": 6.974657092819616e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 69553688,
      "step": 4323
    },
    {
      "epoch": 0.30288861453325977,
      "grad_norm": 3.87226939201355,
      "learning_rate": 6.973957267950964e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 69570072,
      "step": 4324
    },
    {
      "epoch": 0.30295866277898903,
      "grad_norm": 4.516434192657471,
      "learning_rate": 6.973257443082311e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 69585056,
      "step": 4325
    },
    {
      "epoch": 0.3030287110247183,
      "grad_norm": 4.07093620300293,
      "learning_rate": 6.97255761821366e-05,
      "loss": 1.1811,
      "num_input_tokens_seen": 69601440,
      "step": 4326
    },
    {
      "epoch": 0.3030987592704475,
      "grad_norm": 3.663632392883301,
      "learning_rate": 6.971857793345009e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 69617824,
      "step": 4327
    },
    {
      "epoch": 0.30316880751617675,
      "grad_norm": 3.791191577911377,
      "learning_rate": 6.971157968476357e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 69634208,
      "step": 4328
    },
    {
      "epoch": 0.303238855761906,
      "grad_norm": 4.766335964202881,
      "learning_rate": 6.970458143607707e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 69650592,
      "step": 4329
    },
    {
      "epoch": 0.30330890400763527,
      "grad_norm": 3.6603240966796875,
      "learning_rate": 6.969758318739056e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 69666976,
      "step": 4330
    },
    {
      "epoch": 0.30337895225336453,
      "grad_norm": 4.231273174285889,
      "learning_rate": 6.969058493870403e-05,
      "loss": 1.1661,
      "num_input_tokens_seen": 69683360,
      "step": 4331
    },
    {
      "epoch": 0.30344900049909374,
      "grad_norm": 3.7526698112487793,
      "learning_rate": 6.968358669001751e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 69699744,
      "step": 4332
    },
    {
      "epoch": 0.303519048744823,
      "grad_norm": 3.8541617393493652,
      "learning_rate": 6.967658844133101e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 69715960,
      "step": 4333
    },
    {
      "epoch": 0.30358909699055225,
      "grad_norm": 3.914926767349243,
      "learning_rate": 6.966959019264448e-05,
      "loss": 0.9525,
      "num_input_tokens_seen": 69732344,
      "step": 4334
    },
    {
      "epoch": 0.3036591452362815,
      "grad_norm": 4.39329719543457,
      "learning_rate": 6.966259194395797e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 69748728,
      "step": 4335
    },
    {
      "epoch": 0.3037291934820107,
      "grad_norm": 3.914006233215332,
      "learning_rate": 6.965559369527146e-05,
      "loss": 1.1397,
      "num_input_tokens_seen": 69765112,
      "step": 4336
    },
    {
      "epoch": 0.30379924172774,
      "grad_norm": 4.536770343780518,
      "learning_rate": 6.964859544658495e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 69781496,
      "step": 4337
    },
    {
      "epoch": 0.30386928997346924,
      "grad_norm": 4.147655010223389,
      "learning_rate": 6.964159719789843e-05,
      "loss": 0.7535,
      "num_input_tokens_seen": 69797880,
      "step": 4338
    },
    {
      "epoch": 0.3039393382191985,
      "grad_norm": 4.224967956542969,
      "learning_rate": 6.963459894921191e-05,
      "loss": 1.1842,
      "num_input_tokens_seen": 69814264,
      "step": 4339
    },
    {
      "epoch": 0.3040093864649277,
      "grad_norm": 4.415369033813477,
      "learning_rate": 6.96276007005254e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 69830648,
      "step": 4340
    },
    {
      "epoch": 0.30407943471065696,
      "grad_norm": 3.5865182876586914,
      "learning_rate": 6.962060245183888e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 69846608,
      "step": 4341
    },
    {
      "epoch": 0.3041494829563862,
      "grad_norm": 6.16670560836792,
      "learning_rate": 6.961360420315237e-05,
      "loss": 1.0684,
      "num_input_tokens_seen": 69862232,
      "step": 4342
    },
    {
      "epoch": 0.3042195312021155,
      "grad_norm": 7.907288074493408,
      "learning_rate": 6.960660595446586e-05,
      "loss": 0.9661,
      "num_input_tokens_seen": 69878616,
      "step": 4343
    },
    {
      "epoch": 0.3042895794478447,
      "grad_norm": 3.7910618782043457,
      "learning_rate": 6.959960770577934e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 69895000,
      "step": 4344
    },
    {
      "epoch": 0.30435962769357394,
      "grad_norm": 3.4832661151885986,
      "learning_rate": 6.959260945709282e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 69911384,
      "step": 4345
    },
    {
      "epoch": 0.3044296759393032,
      "grad_norm": 3.563248872756958,
      "learning_rate": 6.958561120840631e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 69927768,
      "step": 4346
    },
    {
      "epoch": 0.30449972418503246,
      "grad_norm": 4.838014602661133,
      "learning_rate": 6.95786129597198e-05,
      "loss": 1.2449,
      "num_input_tokens_seen": 69944152,
      "step": 4347
    },
    {
      "epoch": 0.30456977243076166,
      "grad_norm": 3.6796975135803223,
      "learning_rate": 6.957161471103327e-05,
      "loss": 0.8156,
      "num_input_tokens_seen": 69959968,
      "step": 4348
    },
    {
      "epoch": 0.3046398206764909,
      "grad_norm": 4.028040885925293,
      "learning_rate": 6.956461646234677e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 69975960,
      "step": 4349
    },
    {
      "epoch": 0.3047098689222202,
      "grad_norm": 4.073189735412598,
      "learning_rate": 6.955761821366026e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 69991656,
      "step": 4350
    },
    {
      "epoch": 0.30477991716794944,
      "grad_norm": 5.757152080535889,
      "learning_rate": 6.955061996497374e-05,
      "loss": 1.2074,
      "num_input_tokens_seen": 70008040,
      "step": 4351
    },
    {
      "epoch": 0.30484996541367865,
      "grad_norm": 5.49181604385376,
      "learning_rate": 6.954362171628721e-05,
      "loss": 1.0235,
      "num_input_tokens_seen": 70024424,
      "step": 4352
    },
    {
      "epoch": 0.3049200136594079,
      "grad_norm": 5.573401927947998,
      "learning_rate": 6.95366234676007e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 70040808,
      "step": 4353
    },
    {
      "epoch": 0.30499006190513717,
      "grad_norm": 3.491823673248291,
      "learning_rate": 6.952962521891419e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 70057192,
      "step": 4354
    },
    {
      "epoch": 0.3050601101508664,
      "grad_norm": 6.05043888092041,
      "learning_rate": 6.952262697022768e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 70073576,
      "step": 4355
    },
    {
      "epoch": 0.30513015839659563,
      "grad_norm": 3.848910331726074,
      "learning_rate": 6.951562872154117e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 70089960,
      "step": 4356
    },
    {
      "epoch": 0.3052002066423249,
      "grad_norm": 4.134339332580566,
      "learning_rate": 6.950863047285465e-05,
      "loss": 1.2447,
      "num_input_tokens_seen": 70106344,
      "step": 4357
    },
    {
      "epoch": 0.30527025488805415,
      "grad_norm": 3.6560862064361572,
      "learning_rate": 6.950163222416813e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 70122056,
      "step": 4358
    },
    {
      "epoch": 0.3053403031337834,
      "grad_norm": 3.813434600830078,
      "learning_rate": 6.94946339754816e-05,
      "loss": 1.0149,
      "num_input_tokens_seen": 70138408,
      "step": 4359
    },
    {
      "epoch": 0.3054103513795126,
      "grad_norm": 5.002225875854492,
      "learning_rate": 6.948763572679511e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 70154792,
      "step": 4360
    },
    {
      "epoch": 0.30548039962524187,
      "grad_norm": 3.8483340740203857,
      "learning_rate": 6.948063747810858e-05,
      "loss": 0.9643,
      "num_input_tokens_seen": 70171176,
      "step": 4361
    },
    {
      "epoch": 0.30555044787097113,
      "grad_norm": 5.18534517288208,
      "learning_rate": 6.947363922942207e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 70187336,
      "step": 4362
    },
    {
      "epoch": 0.3056204961167004,
      "grad_norm": 3.92976713180542,
      "learning_rate": 6.946664098073556e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 70203720,
      "step": 4363
    },
    {
      "epoch": 0.30569054436242965,
      "grad_norm": 3.4534151554107666,
      "learning_rate": 6.945964273204905e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 70220104,
      "step": 4364
    },
    {
      "epoch": 0.30576059260815885,
      "grad_norm": 3.7937867641448975,
      "learning_rate": 6.945264448336252e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 70236488,
      "step": 4365
    },
    {
      "epoch": 0.3058306408538881,
      "grad_norm": 3.9063713550567627,
      "learning_rate": 6.944564623467601e-05,
      "loss": 1.0969,
      "num_input_tokens_seen": 70252872,
      "step": 4366
    },
    {
      "epoch": 0.3059006890996174,
      "grad_norm": 3.9363296031951904,
      "learning_rate": 6.94386479859895e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 70269256,
      "step": 4367
    },
    {
      "epoch": 0.30597073734534663,
      "grad_norm": 4.722838401794434,
      "learning_rate": 6.943164973730297e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 70285640,
      "step": 4368
    },
    {
      "epoch": 0.30604078559107584,
      "grad_norm": 4.053229808807373,
      "learning_rate": 6.942465148861646e-05,
      "loss": 1.2669,
      "num_input_tokens_seen": 70301688,
      "step": 4369
    },
    {
      "epoch": 0.3061108338368051,
      "grad_norm": 3.71604323387146,
      "learning_rate": 6.941765323992995e-05,
      "loss": 1.0619,
      "num_input_tokens_seen": 70318072,
      "step": 4370
    },
    {
      "epoch": 0.30618088208253436,
      "grad_norm": 3.8376901149749756,
      "learning_rate": 6.941065499124344e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 70334456,
      "step": 4371
    },
    {
      "epoch": 0.3062509303282636,
      "grad_norm": 4.157979488372803,
      "learning_rate": 6.940365674255692e-05,
      "loss": 1.2379,
      "num_input_tokens_seen": 70350424,
      "step": 4372
    },
    {
      "epoch": 0.3063209785739928,
      "grad_norm": 4.173924922943115,
      "learning_rate": 6.93966584938704e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 70366808,
      "step": 4373
    },
    {
      "epoch": 0.3063910268197221,
      "grad_norm": 4.114030838012695,
      "learning_rate": 6.938966024518389e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 70383000,
      "step": 4374
    },
    {
      "epoch": 0.30646107506545134,
      "grad_norm": 4.31168794631958,
      "learning_rate": 6.938266199649738e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 70399016,
      "step": 4375
    },
    {
      "epoch": 0.3065311233111806,
      "grad_norm": 6.187852382659912,
      "learning_rate": 6.937566374781087e-05,
      "loss": 0.9274,
      "num_input_tokens_seen": 70413016,
      "step": 4376
    },
    {
      "epoch": 0.3066011715569098,
      "grad_norm": 4.700244903564453,
      "learning_rate": 6.936866549912436e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 70427552,
      "step": 4377
    },
    {
      "epoch": 0.30667121980263906,
      "grad_norm": 4.941024303436279,
      "learning_rate": 6.936166725043783e-05,
      "loss": 1.1653,
      "num_input_tokens_seen": 70443936,
      "step": 4378
    },
    {
      "epoch": 0.3067412680483683,
      "grad_norm": 3.8171792030334473,
      "learning_rate": 6.935466900175131e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 70460320,
      "step": 4379
    },
    {
      "epoch": 0.3068113162940976,
      "grad_norm": 5.006760597229004,
      "learning_rate": 6.93476707530648e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 70476704,
      "step": 4380
    },
    {
      "epoch": 0.3068813645398268,
      "grad_norm": 3.8567628860473633,
      "learning_rate": 6.934067250437829e-05,
      "loss": 1.1478,
      "num_input_tokens_seen": 70493016,
      "step": 4381
    },
    {
      "epoch": 0.30695141278555604,
      "grad_norm": 3.7168126106262207,
      "learning_rate": 6.933367425569177e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 70509400,
      "step": 4382
    },
    {
      "epoch": 0.3070214610312853,
      "grad_norm": 4.72265625,
      "learning_rate": 6.932667600700526e-05,
      "loss": 1.0319,
      "num_input_tokens_seen": 70525592,
      "step": 4383
    },
    {
      "epoch": 0.30709150927701456,
      "grad_norm": 4.502997875213623,
      "learning_rate": 6.931967775831875e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 70541976,
      "step": 4384
    },
    {
      "epoch": 0.30716155752274377,
      "grad_norm": 4.090621471405029,
      "learning_rate": 6.931267950963223e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 70558360,
      "step": 4385
    },
    {
      "epoch": 0.307231605768473,
      "grad_norm": 3.501185655593872,
      "learning_rate": 6.93056812609457e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 70574640,
      "step": 4386
    },
    {
      "epoch": 0.3073016540142023,
      "grad_norm": 3.937352180480957,
      "learning_rate": 6.92986830122592e-05,
      "loss": 1.1003,
      "num_input_tokens_seen": 70591024,
      "step": 4387
    },
    {
      "epoch": 0.30737170225993155,
      "grad_norm": 8.832700729370117,
      "learning_rate": 6.929168476357268e-05,
      "loss": 0.9773,
      "num_input_tokens_seen": 70606720,
      "step": 4388
    },
    {
      "epoch": 0.30744175050566075,
      "grad_norm": 3.8081719875335693,
      "learning_rate": 6.928468651488617e-05,
      "loss": 0.8348,
      "num_input_tokens_seen": 70622864,
      "step": 4389
    },
    {
      "epoch": 0.30751179875139,
      "grad_norm": 3.836366653442383,
      "learning_rate": 6.927768826619966e-05,
      "loss": 1.0858,
      "num_input_tokens_seen": 70639248,
      "step": 4390
    },
    {
      "epoch": 0.30758184699711927,
      "grad_norm": 5.150767803192139,
      "learning_rate": 6.927069001751314e-05,
      "loss": 1.0395,
      "num_input_tokens_seen": 70655128,
      "step": 4391
    },
    {
      "epoch": 0.3076518952428485,
      "grad_norm": 5.0762434005737305,
      "learning_rate": 6.926369176882662e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 70671512,
      "step": 4392
    },
    {
      "epoch": 0.30772194348857773,
      "grad_norm": 3.7713098526000977,
      "learning_rate": 6.925669352014011e-05,
      "loss": 0.9273,
      "num_input_tokens_seen": 70687896,
      "step": 4393
    },
    {
      "epoch": 0.307791991734307,
      "grad_norm": 5.246247291564941,
      "learning_rate": 6.92496952714536e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 70704280,
      "step": 4394
    },
    {
      "epoch": 0.30786203998003625,
      "grad_norm": 3.5723984241485596,
      "learning_rate": 6.924269702276707e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 70720664,
      "step": 4395
    },
    {
      "epoch": 0.3079320882257655,
      "grad_norm": 3.5165982246398926,
      "learning_rate": 6.923569877408056e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 70736968,
      "step": 4396
    },
    {
      "epoch": 0.30800213647149477,
      "grad_norm": 4.140204429626465,
      "learning_rate": 6.922870052539405e-05,
      "loss": 0.9557,
      "num_input_tokens_seen": 70753352,
      "step": 4397
    },
    {
      "epoch": 0.308072184717224,
      "grad_norm": 7.949122428894043,
      "learning_rate": 6.922170227670754e-05,
      "loss": 0.9429,
      "num_input_tokens_seen": 70769720,
      "step": 4398
    },
    {
      "epoch": 0.30814223296295323,
      "grad_norm": 6.45367431640625,
      "learning_rate": 6.921470402802101e-05,
      "loss": 1.2214,
      "num_input_tokens_seen": 70784984,
      "step": 4399
    },
    {
      "epoch": 0.3082122812086825,
      "grad_norm": 4.139477252960205,
      "learning_rate": 6.92077057793345e-05,
      "loss": 1.2376,
      "num_input_tokens_seen": 70800376,
      "step": 4400
    },
    {
      "epoch": 0.3082122812086825,
      "eval_loss": 1.1308872699737549,
      "eval_runtime": 0.2076,
      "eval_samples_per_second": 4.818,
      "eval_steps_per_second": 4.818,
      "num_input_tokens_seen": 70800376,
      "step": 4400
    },
    {
      "epoch": 0.30828232945441175,
      "grad_norm": 4.095129013061523,
      "learning_rate": 6.920070753064799e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 70816560,
      "step": 4401
    },
    {
      "epoch": 0.30835237770014096,
      "grad_norm": 3.6730854511260986,
      "learning_rate": 6.919370928196148e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 70831848,
      "step": 4402
    },
    {
      "epoch": 0.3084224259458702,
      "grad_norm": 4.013517379760742,
      "learning_rate": 6.918671103327497e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 70847408,
      "step": 4403
    },
    {
      "epoch": 0.3084924741915995,
      "grad_norm": 5.617120742797852,
      "learning_rate": 6.917971278458846e-05,
      "loss": 0.9883,
      "num_input_tokens_seen": 70862080,
      "step": 4404
    },
    {
      "epoch": 0.30856252243732873,
      "grad_norm": 3.5201385021209717,
      "learning_rate": 6.917271453590193e-05,
      "loss": 0.9537,
      "num_input_tokens_seen": 70878464,
      "step": 4405
    },
    {
      "epoch": 0.30863257068305794,
      "grad_norm": 5.116230010986328,
      "learning_rate": 6.91657162872154e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 70894848,
      "step": 4406
    },
    {
      "epoch": 0.3087026189287872,
      "grad_norm": 3.4510743618011475,
      "learning_rate": 6.91587180385289e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 70911232,
      "step": 4407
    },
    {
      "epoch": 0.30877266717451646,
      "grad_norm": 4.719654083251953,
      "learning_rate": 6.915171978984238e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 70927616,
      "step": 4408
    },
    {
      "epoch": 0.3088427154202457,
      "grad_norm": 4.52898645401001,
      "learning_rate": 6.914472154115587e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 70944000,
      "step": 4409
    },
    {
      "epoch": 0.3089127636659749,
      "grad_norm": 4.237354755401611,
      "learning_rate": 6.913772329246936e-05,
      "loss": 1.1614,
      "num_input_tokens_seen": 70960384,
      "step": 4410
    },
    {
      "epoch": 0.3089828119117042,
      "grad_norm": 5.489138126373291,
      "learning_rate": 6.913072504378285e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 70976768,
      "step": 4411
    },
    {
      "epoch": 0.30905286015743344,
      "grad_norm": 5.482370853424072,
      "learning_rate": 6.912372679509632e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 70992496,
      "step": 4412
    },
    {
      "epoch": 0.3091229084031627,
      "grad_norm": 3.8174126148223877,
      "learning_rate": 6.91167285464098e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 71008880,
      "step": 4413
    },
    {
      "epoch": 0.3091929566488919,
      "grad_norm": 4.064924716949463,
      "learning_rate": 6.91097302977233e-05,
      "loss": 0.8307,
      "num_input_tokens_seen": 71023912,
      "step": 4414
    },
    {
      "epoch": 0.30926300489462116,
      "grad_norm": 3.955643653869629,
      "learning_rate": 6.910273204903678e-05,
      "loss": 1.2599,
      "num_input_tokens_seen": 71040296,
      "step": 4415
    },
    {
      "epoch": 0.3093330531403504,
      "grad_norm": 3.771191358566284,
      "learning_rate": 6.909573380035026e-05,
      "loss": 1.0682,
      "num_input_tokens_seen": 71056680,
      "step": 4416
    },
    {
      "epoch": 0.3094031013860797,
      "grad_norm": 5.4105963706970215,
      "learning_rate": 6.908873555166375e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 71072640,
      "step": 4417
    },
    {
      "epoch": 0.3094731496318089,
      "grad_norm": 4.549078464508057,
      "learning_rate": 6.908173730297724e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 71087336,
      "step": 4418
    },
    {
      "epoch": 0.30954319787753815,
      "grad_norm": 3.998065233230591,
      "learning_rate": 6.907473905429072e-05,
      "loss": 1.2753,
      "num_input_tokens_seen": 71102952,
      "step": 4419
    },
    {
      "epoch": 0.3096132461232674,
      "grad_norm": 3.834508180618286,
      "learning_rate": 6.90677408056042e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 71119328,
      "step": 4420
    },
    {
      "epoch": 0.30968329436899666,
      "grad_norm": 3.932875156402588,
      "learning_rate": 6.90607425569177e-05,
      "loss": 1.1568,
      "num_input_tokens_seen": 71134968,
      "step": 4421
    },
    {
      "epoch": 0.30975334261472587,
      "grad_norm": 3.712484359741211,
      "learning_rate": 6.905374430823118e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 71150976,
      "step": 4422
    },
    {
      "epoch": 0.30982339086045513,
      "grad_norm": 3.6733663082122803,
      "learning_rate": 6.904674605954466e-05,
      "loss": 0.8884,
      "num_input_tokens_seen": 71167232,
      "step": 4423
    },
    {
      "epoch": 0.3098934391061844,
      "grad_norm": 3.9877066612243652,
      "learning_rate": 6.903974781085815e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 71182704,
      "step": 4424
    },
    {
      "epoch": 0.30996348735191365,
      "grad_norm": 3.908582925796509,
      "learning_rate": 6.903274956217163e-05,
      "loss": 0.9944,
      "num_input_tokens_seen": 71198920,
      "step": 4425
    },
    {
      "epoch": 0.31003353559764285,
      "grad_norm": 4.310460090637207,
      "learning_rate": 6.902575131348511e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 71215256,
      "step": 4426
    },
    {
      "epoch": 0.3101035838433721,
      "grad_norm": 3.8914272785186768,
      "learning_rate": 6.90187530647986e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 71231432,
      "step": 4427
    },
    {
      "epoch": 0.31017363208910137,
      "grad_norm": 5.774794578552246,
      "learning_rate": 6.901175481611209e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 71246944,
      "step": 4428
    },
    {
      "epoch": 0.31024368033483063,
      "grad_norm": 6.370543956756592,
      "learning_rate": 6.900475656742558e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 71263120,
      "step": 4429
    },
    {
      "epoch": 0.31031372858055983,
      "grad_norm": 3.8334455490112305,
      "learning_rate": 6.899775831873906e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 71279040,
      "step": 4430
    },
    {
      "epoch": 0.3103837768262891,
      "grad_norm": 3.624006509780884,
      "learning_rate": 6.899076007005255e-05,
      "loss": 0.9908,
      "num_input_tokens_seen": 71295424,
      "step": 4431
    },
    {
      "epoch": 0.31045382507201835,
      "grad_norm": 3.8340702056884766,
      "learning_rate": 6.898376182136603e-05,
      "loss": 1.1257,
      "num_input_tokens_seen": 71311808,
      "step": 4432
    },
    {
      "epoch": 0.3105238733177476,
      "grad_norm": 4.4179277420043945,
      "learning_rate": 6.89767635726795e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 71327560,
      "step": 4433
    },
    {
      "epoch": 0.31059392156347687,
      "grad_norm": 5.758373260498047,
      "learning_rate": 6.896976532399299e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 71342848,
      "step": 4434
    },
    {
      "epoch": 0.3106639698092061,
      "grad_norm": 3.7063519954681396,
      "learning_rate": 6.896276707530648e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 71359232,
      "step": 4435
    },
    {
      "epoch": 0.31073401805493533,
      "grad_norm": 4.19386625289917,
      "learning_rate": 6.895576882661997e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 71375616,
      "step": 4436
    },
    {
      "epoch": 0.3108040663006646,
      "grad_norm": 4.116868019104004,
      "learning_rate": 6.894877057793346e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 71392000,
      "step": 4437
    },
    {
      "epoch": 0.31087411454639385,
      "grad_norm": 4.810275077819824,
      "learning_rate": 6.894177232924695e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 71407448,
      "step": 4438
    },
    {
      "epoch": 0.31094416279212306,
      "grad_norm": 4.026486873626709,
      "learning_rate": 6.893477408056042e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 71423832,
      "step": 4439
    },
    {
      "epoch": 0.3110142110378523,
      "grad_norm": 4.268560886383057,
      "learning_rate": 6.89277758318739e-05,
      "loss": 1.0746,
      "num_input_tokens_seen": 71440216,
      "step": 4440
    },
    {
      "epoch": 0.3110842592835816,
      "grad_norm": 3.3299612998962402,
      "learning_rate": 6.89207775831874e-05,
      "loss": 0.7408,
      "num_input_tokens_seen": 71456160,
      "step": 4441
    },
    {
      "epoch": 0.31115430752931084,
      "grad_norm": 3.678912401199341,
      "learning_rate": 6.891377933450089e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 71472344,
      "step": 4442
    },
    {
      "epoch": 0.31122435577504004,
      "grad_norm": 3.3206088542938232,
      "learning_rate": 6.890678108581436e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 71488728,
      "step": 4443
    },
    {
      "epoch": 0.3112944040207693,
      "grad_norm": 3.6073081493377686,
      "learning_rate": 6.889978283712785e-05,
      "loss": 1.18,
      "num_input_tokens_seen": 71505112,
      "step": 4444
    },
    {
      "epoch": 0.31136445226649856,
      "grad_norm": 4.998234748840332,
      "learning_rate": 6.889278458844134e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 71521496,
      "step": 4445
    },
    {
      "epoch": 0.3114345005122278,
      "grad_norm": 3.7966136932373047,
      "learning_rate": 6.888578633975481e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 71537880,
      "step": 4446
    },
    {
      "epoch": 0.311504548757957,
      "grad_norm": 3.7041022777557373,
      "learning_rate": 6.88787880910683e-05,
      "loss": 1.0701,
      "num_input_tokens_seen": 71554136,
      "step": 4447
    },
    {
      "epoch": 0.3115745970036863,
      "grad_norm": 4.155350208282471,
      "learning_rate": 6.887178984238179e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 71570520,
      "step": 4448
    },
    {
      "epoch": 0.31164464524941554,
      "grad_norm": 3.608290195465088,
      "learning_rate": 6.886479159369528e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 71586904,
      "step": 4449
    },
    {
      "epoch": 0.3117146934951448,
      "grad_norm": 5.258309841156006,
      "learning_rate": 6.885779334500875e-05,
      "loss": 1.0929,
      "num_input_tokens_seen": 71602280,
      "step": 4450
    },
    {
      "epoch": 0.311784741740874,
      "grad_norm": 4.176782608032227,
      "learning_rate": 6.885079509632224e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 71618248,
      "step": 4451
    },
    {
      "epoch": 0.31185478998660326,
      "grad_norm": 3.219015121459961,
      "learning_rate": 6.884379684763573e-05,
      "loss": 0.8699,
      "num_input_tokens_seen": 71634632,
      "step": 4452
    },
    {
      "epoch": 0.3119248382323325,
      "grad_norm": 3.485370397567749,
      "learning_rate": 6.883679859894921e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 71651016,
      "step": 4453
    },
    {
      "epoch": 0.3119948864780618,
      "grad_norm": 4.25452184677124,
      "learning_rate": 6.88298003502627e-05,
      "loss": 1.1649,
      "num_input_tokens_seen": 71667400,
      "step": 4454
    },
    {
      "epoch": 0.312064934723791,
      "grad_norm": 4.2082133293151855,
      "learning_rate": 6.882280210157618e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 71682656,
      "step": 4455
    },
    {
      "epoch": 0.31213498296952025,
      "grad_norm": 3.366642475128174,
      "learning_rate": 6.881580385288967e-05,
      "loss": 0.8594,
      "num_input_tokens_seen": 71699040,
      "step": 4456
    },
    {
      "epoch": 0.3122050312152495,
      "grad_norm": 3.795114278793335,
      "learning_rate": 6.880880560420316e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 71715424,
      "step": 4457
    },
    {
      "epoch": 0.31227507946097877,
      "grad_norm": 7.3179121017456055,
      "learning_rate": 6.880180735551665e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 71731808,
      "step": 4458
    },
    {
      "epoch": 0.31234512770670797,
      "grad_norm": 5.0151848793029785,
      "learning_rate": 6.879480910683012e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 71746688,
      "step": 4459
    },
    {
      "epoch": 0.31241517595243723,
      "grad_norm": 4.136596202850342,
      "learning_rate": 6.87878108581436e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 71763016,
      "step": 4460
    },
    {
      "epoch": 0.3124852241981665,
      "grad_norm": 3.6476573944091797,
      "learning_rate": 6.878081260945709e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 71779400,
      "step": 4461
    },
    {
      "epoch": 0.31255527244389575,
      "grad_norm": 4.907565593719482,
      "learning_rate": 6.877381436077059e-05,
      "loss": 1.0391,
      "num_input_tokens_seen": 71795784,
      "step": 4462
    },
    {
      "epoch": 0.31262532068962495,
      "grad_norm": 3.82183575630188,
      "learning_rate": 6.876681611208407e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 71812168,
      "step": 4463
    },
    {
      "epoch": 0.3126953689353542,
      "grad_norm": 4.63422966003418,
      "learning_rate": 6.875981786339755e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 71828296,
      "step": 4464
    },
    {
      "epoch": 0.31276541718108347,
      "grad_norm": 4.02967643737793,
      "learning_rate": 6.875281961471104e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 71844096,
      "step": 4465
    },
    {
      "epoch": 0.31283546542681273,
      "grad_norm": 3.477452516555786,
      "learning_rate": 6.874582136602452e-05,
      "loss": 1.0503,
      "num_input_tokens_seen": 71860480,
      "step": 4466
    },
    {
      "epoch": 0.312905513672542,
      "grad_norm": 4.4327168464660645,
      "learning_rate": 6.873882311733799e-05,
      "loss": 1.2305,
      "num_input_tokens_seen": 71876208,
      "step": 4467
    },
    {
      "epoch": 0.3129755619182712,
      "grad_norm": 3.8214218616485596,
      "learning_rate": 6.87318248686515e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 71892592,
      "step": 4468
    },
    {
      "epoch": 0.31304561016400045,
      "grad_norm": 3.4210402965545654,
      "learning_rate": 6.872482661996498e-05,
      "loss": 0.8505,
      "num_input_tokens_seen": 71908976,
      "step": 4469
    },
    {
      "epoch": 0.3131156584097297,
      "grad_norm": 3.567034959793091,
      "learning_rate": 6.871782837127846e-05,
      "loss": 0.7866,
      "num_input_tokens_seen": 71925200,
      "step": 4470
    },
    {
      "epoch": 0.313185706655459,
      "grad_norm": 4.694231986999512,
      "learning_rate": 6.871083012259195e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 71941584,
      "step": 4471
    },
    {
      "epoch": 0.3132557549011882,
      "grad_norm": 5.802227973937988,
      "learning_rate": 6.870383187390544e-05,
      "loss": 1.1923,
      "num_input_tokens_seen": 71957968,
      "step": 4472
    },
    {
      "epoch": 0.31332580314691744,
      "grad_norm": 4.238499641418457,
      "learning_rate": 6.869683362521891e-05,
      "loss": 1.3381,
      "num_input_tokens_seen": 71973376,
      "step": 4473
    },
    {
      "epoch": 0.3133958513926467,
      "grad_norm": 4.2250213623046875,
      "learning_rate": 6.86898353765324e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 71988560,
      "step": 4474
    },
    {
      "epoch": 0.31346589963837596,
      "grad_norm": 4.052889823913574,
      "learning_rate": 6.868283712784589e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 72004232,
      "step": 4475
    },
    {
      "epoch": 0.31353594788410516,
      "grad_norm": 3.694481134414673,
      "learning_rate": 6.867583887915938e-05,
      "loss": 1.187,
      "num_input_tokens_seen": 72020616,
      "step": 4476
    },
    {
      "epoch": 0.3136059961298344,
      "grad_norm": 4.2295074462890625,
      "learning_rate": 6.866884063047285e-05,
      "loss": 1.2454,
      "num_input_tokens_seen": 72036912,
      "step": 4477
    },
    {
      "epoch": 0.3136760443755637,
      "grad_norm": 3.9813766479492188,
      "learning_rate": 6.866184238178634e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 72053296,
      "step": 4478
    },
    {
      "epoch": 0.31374609262129294,
      "grad_norm": 4.473883152008057,
      "learning_rate": 6.865484413309983e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 72069680,
      "step": 4479
    },
    {
      "epoch": 0.31381614086702214,
      "grad_norm": 3.7663521766662598,
      "learning_rate": 6.86478458844133e-05,
      "loss": 0.937,
      "num_input_tokens_seen": 72085840,
      "step": 4480
    },
    {
      "epoch": 0.3138861891127514,
      "grad_norm": 3.9587883949279785,
      "learning_rate": 6.864084763572679e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 72102224,
      "step": 4481
    },
    {
      "epoch": 0.31395623735848066,
      "grad_norm": 3.953232526779175,
      "learning_rate": 6.86338493870403e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 72118608,
      "step": 4482
    },
    {
      "epoch": 0.3140262856042099,
      "grad_norm": 3.917574882507324,
      "learning_rate": 6.862685113835377e-05,
      "loss": 1.1883,
      "num_input_tokens_seen": 72134504,
      "step": 4483
    },
    {
      "epoch": 0.3140963338499391,
      "grad_norm": 3.756253242492676,
      "learning_rate": 6.861985288966726e-05,
      "loss": 1.1057,
      "num_input_tokens_seen": 72150368,
      "step": 4484
    },
    {
      "epoch": 0.3141663820956684,
      "grad_norm": 4.146200656890869,
      "learning_rate": 6.861285464098075e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 72166752,
      "step": 4485
    },
    {
      "epoch": 0.31423643034139764,
      "grad_norm": 3.98949933052063,
      "learning_rate": 6.860585639229422e-05,
      "loss": 1.2088,
      "num_input_tokens_seen": 72182696,
      "step": 4486
    },
    {
      "epoch": 0.3143064785871269,
      "grad_norm": 3.99951434135437,
      "learning_rate": 6.85988581436077e-05,
      "loss": 1.218,
      "num_input_tokens_seen": 72199080,
      "step": 4487
    },
    {
      "epoch": 0.3143765268328561,
      "grad_norm": 4.351415157318115,
      "learning_rate": 6.85918598949212e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 72215176,
      "step": 4488
    },
    {
      "epoch": 0.31444657507858537,
      "grad_norm": 4.563141822814941,
      "learning_rate": 6.858486164623469e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 72231560,
      "step": 4489
    },
    {
      "epoch": 0.3145166233243146,
      "grad_norm": 4.523083686828613,
      "learning_rate": 6.857786339754816e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 72246920,
      "step": 4490
    },
    {
      "epoch": 0.3145866715700439,
      "grad_norm": 4.032657623291016,
      "learning_rate": 6.857086514886165e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 72263304,
      "step": 4491
    },
    {
      "epoch": 0.3146567198157731,
      "grad_norm": 4.755338191986084,
      "learning_rate": 6.856386690017514e-05,
      "loss": 1.0756,
      "num_input_tokens_seen": 72279688,
      "step": 4492
    },
    {
      "epoch": 0.31472676806150235,
      "grad_norm": 4.037180423736572,
      "learning_rate": 6.855686865148862e-05,
      "loss": 1.2973,
      "num_input_tokens_seen": 72296072,
      "step": 4493
    },
    {
      "epoch": 0.3147968163072316,
      "grad_norm": 3.308746099472046,
      "learning_rate": 6.85498704028021e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 72312360,
      "step": 4494
    },
    {
      "epoch": 0.31486686455296087,
      "grad_norm": 4.204549789428711,
      "learning_rate": 6.854287215411559e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 72328744,
      "step": 4495
    },
    {
      "epoch": 0.31493691279869007,
      "grad_norm": 4.142894744873047,
      "learning_rate": 6.853587390542908e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 72344944,
      "step": 4496
    },
    {
      "epoch": 0.31500696104441933,
      "grad_norm": 5.43609094619751,
      "learning_rate": 6.852887565674256e-05,
      "loss": 0.9369,
      "num_input_tokens_seen": 72360672,
      "step": 4497
    },
    {
      "epoch": 0.3150770092901486,
      "grad_norm": 4.20035982131958,
      "learning_rate": 6.852187740805604e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 72376744,
      "step": 4498
    },
    {
      "epoch": 0.31514705753587785,
      "grad_norm": 3.6777737140655518,
      "learning_rate": 6.851487915936953e-05,
      "loss": 1.0489,
      "num_input_tokens_seen": 72393128,
      "step": 4499
    },
    {
      "epoch": 0.31521710578160705,
      "grad_norm": 5.047235488891602,
      "learning_rate": 6.850788091068301e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 72408016,
      "step": 4500
    },
    {
      "epoch": 0.3152871540273363,
      "grad_norm": 4.095731258392334,
      "learning_rate": 6.85008826619965e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 72424400,
      "step": 4501
    },
    {
      "epoch": 0.3153572022730656,
      "grad_norm": 3.6437504291534424,
      "learning_rate": 6.849388441331e-05,
      "loss": 1.1428,
      "num_input_tokens_seen": 72440368,
      "step": 4502
    },
    {
      "epoch": 0.31542725051879483,
      "grad_norm": 5.345888614654541,
      "learning_rate": 6.848688616462347e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 72456752,
      "step": 4503
    },
    {
      "epoch": 0.3154972987645241,
      "grad_norm": 4.471817970275879,
      "learning_rate": 6.847988791593695e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 72472952,
      "step": 4504
    },
    {
      "epoch": 0.3155673470102533,
      "grad_norm": 3.8012888431549072,
      "learning_rate": 6.847288966725044e-05,
      "loss": 1.1961,
      "num_input_tokens_seen": 72489256,
      "step": 4505
    },
    {
      "epoch": 0.31563739525598256,
      "grad_norm": 7.531235218048096,
      "learning_rate": 6.846589141856393e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 72503752,
      "step": 4506
    },
    {
      "epoch": 0.3157074435017118,
      "grad_norm": 4.075259208679199,
      "learning_rate": 6.84588931698774e-05,
      "loss": 1.1834,
      "num_input_tokens_seen": 72520136,
      "step": 4507
    },
    {
      "epoch": 0.3157774917474411,
      "grad_norm": 5.203637599945068,
      "learning_rate": 6.84518949211909e-05,
      "loss": 1.1198,
      "num_input_tokens_seen": 72536520,
      "step": 4508
    },
    {
      "epoch": 0.3158475399931703,
      "grad_norm": 5.733241081237793,
      "learning_rate": 6.844489667250439e-05,
      "loss": 1.1981,
      "num_input_tokens_seen": 72551976,
      "step": 4509
    },
    {
      "epoch": 0.31591758823889954,
      "grad_norm": 4.182814121246338,
      "learning_rate": 6.843789842381787e-05,
      "loss": 1.0879,
      "num_input_tokens_seen": 72568360,
      "step": 4510
    },
    {
      "epoch": 0.3159876364846288,
      "grad_norm": 5.769293785095215,
      "learning_rate": 6.843090017513136e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 72584744,
      "step": 4511
    },
    {
      "epoch": 0.31605768473035806,
      "grad_norm": 5.052547454833984,
      "learning_rate": 6.842390192644484e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 72600608,
      "step": 4512
    },
    {
      "epoch": 0.31612773297608726,
      "grad_norm": 3.7260072231292725,
      "learning_rate": 6.841690367775832e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 72616720,
      "step": 4513
    },
    {
      "epoch": 0.3161977812218165,
      "grad_norm": 4.230448246002197,
      "learning_rate": 6.840990542907181e-05,
      "loss": 1.3362,
      "num_input_tokens_seen": 72632896,
      "step": 4514
    },
    {
      "epoch": 0.3162678294675458,
      "grad_norm": 3.7840049266815186,
      "learning_rate": 6.84029071803853e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 72649280,
      "step": 4515
    },
    {
      "epoch": 0.31633787771327504,
      "grad_norm": 3.6891443729400635,
      "learning_rate": 6.839590893169878e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 72665664,
      "step": 4516
    },
    {
      "epoch": 0.31640792595900424,
      "grad_norm": 5.132042407989502,
      "learning_rate": 6.838891068301226e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 72682048,
      "step": 4517
    },
    {
      "epoch": 0.3164779742047335,
      "grad_norm": 4.329607009887695,
      "learning_rate": 6.838191243432575e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 72698136,
      "step": 4518
    },
    {
      "epoch": 0.31654802245046276,
      "grad_norm": 4.269455432891846,
      "learning_rate": 6.837491418563924e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 72714296,
      "step": 4519
    },
    {
      "epoch": 0.316618070696192,
      "grad_norm": 3.5963287353515625,
      "learning_rate": 6.836791593695271e-05,
      "loss": 0.8834,
      "num_input_tokens_seen": 72730680,
      "step": 4520
    },
    {
      "epoch": 0.3166881189419212,
      "grad_norm": 3.9145658016204834,
      "learning_rate": 6.83609176882662e-05,
      "loss": 1.1385,
      "num_input_tokens_seen": 72746296,
      "step": 4521
    },
    {
      "epoch": 0.3167581671876505,
      "grad_norm": 4.266791820526123,
      "learning_rate": 6.83539194395797e-05,
      "loss": 1.0825,
      "num_input_tokens_seen": 72762680,
      "step": 4522
    },
    {
      "epoch": 0.31682821543337975,
      "grad_norm": 3.850743532180786,
      "learning_rate": 6.834692119089318e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 72778816,
      "step": 4523
    },
    {
      "epoch": 0.316898263679109,
      "grad_norm": 3.8117008209228516,
      "learning_rate": 6.833992294220665e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 72794576,
      "step": 4524
    },
    {
      "epoch": 0.3169683119248382,
      "grad_norm": 4.469017028808594,
      "learning_rate": 6.833292469352014e-05,
      "loss": 1.2533,
      "num_input_tokens_seen": 72810960,
      "step": 4525
    },
    {
      "epoch": 0.31703836017056747,
      "grad_norm": 3.538980007171631,
      "learning_rate": 6.832592644483363e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 72826480,
      "step": 4526
    },
    {
      "epoch": 0.3171084084162967,
      "grad_norm": 3.6429643630981445,
      "learning_rate": 6.83189281961471e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 72842440,
      "step": 4527
    },
    {
      "epoch": 0.317178456662026,
      "grad_norm": 3.876481056213379,
      "learning_rate": 6.831192994746061e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 72858424,
      "step": 4528
    },
    {
      "epoch": 0.3172485049077552,
      "grad_norm": 5.119854927062988,
      "learning_rate": 6.83049316987741e-05,
      "loss": 1.1704,
      "num_input_tokens_seen": 72874808,
      "step": 4529
    },
    {
      "epoch": 0.31731855315348445,
      "grad_norm": 3.908071994781494,
      "learning_rate": 6.829793345008757e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 72891192,
      "step": 4530
    },
    {
      "epoch": 0.3173886013992137,
      "grad_norm": 4.499825954437256,
      "learning_rate": 6.829093520140105e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 72907576,
      "step": 4531
    },
    {
      "epoch": 0.31745864964494297,
      "grad_norm": 4.060844421386719,
      "learning_rate": 6.828393695271453e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 72923960,
      "step": 4532
    },
    {
      "epoch": 0.3175286978906722,
      "grad_norm": 4.47066068649292,
      "learning_rate": 6.827693870402802e-05,
      "loss": 0.859,
      "num_input_tokens_seen": 72939576,
      "step": 4533
    },
    {
      "epoch": 0.31759874613640143,
      "grad_norm": 3.6252682209014893,
      "learning_rate": 6.826994045534151e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 72955136,
      "step": 4534
    },
    {
      "epoch": 0.3176687943821307,
      "grad_norm": 4.25836181640625,
      "learning_rate": 6.8262942206655e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 72971520,
      "step": 4535
    },
    {
      "epoch": 0.31773884262785995,
      "grad_norm": 3.6240739822387695,
      "learning_rate": 6.825594395796849e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 72987440,
      "step": 4536
    },
    {
      "epoch": 0.3178088908735892,
      "grad_norm": 4.470614910125732,
      "learning_rate": 6.824894570928196e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 73003824,
      "step": 4537
    },
    {
      "epoch": 0.3178789391193184,
      "grad_norm": 3.5759263038635254,
      "learning_rate": 6.824194746059545e-05,
      "loss": 0.9902,
      "num_input_tokens_seen": 73020208,
      "step": 4538
    },
    {
      "epoch": 0.3179489873650477,
      "grad_norm": 4.424665451049805,
      "learning_rate": 6.823494921190894e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 73036592,
      "step": 4539
    },
    {
      "epoch": 0.31801903561077693,
      "grad_norm": 3.803205966949463,
      "learning_rate": 6.822795096322242e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 73052976,
      "step": 4540
    },
    {
      "epoch": 0.3180890838565062,
      "grad_norm": 4.25760555267334,
      "learning_rate": 6.82209527145359e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 73069360,
      "step": 4541
    },
    {
      "epoch": 0.3181591321022354,
      "grad_norm": 4.006928443908691,
      "learning_rate": 6.82139544658494e-05,
      "loss": 0.9056,
      "num_input_tokens_seen": 73084624,
      "step": 4542
    },
    {
      "epoch": 0.31822918034796466,
      "grad_norm": 3.56350040435791,
      "learning_rate": 6.820695621716288e-05,
      "loss": 0.8721,
      "num_input_tokens_seen": 73100008,
      "step": 4543
    },
    {
      "epoch": 0.3182992285936939,
      "grad_norm": 3.7276062965393066,
      "learning_rate": 6.819995796847636e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 73116392,
      "step": 4544
    },
    {
      "epoch": 0.3183692768394232,
      "grad_norm": 4.955738544464111,
      "learning_rate": 6.819295971978985e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 73131920,
      "step": 4545
    },
    {
      "epoch": 0.3184393250851524,
      "grad_norm": 3.5275161266326904,
      "learning_rate": 6.818596147110333e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 73148304,
      "step": 4546
    },
    {
      "epoch": 0.31850937333088164,
      "grad_norm": 6.4245924949646,
      "learning_rate": 6.817896322241681e-05,
      "loss": 1.2968,
      "num_input_tokens_seen": 73164688,
      "step": 4547
    },
    {
      "epoch": 0.3185794215766109,
      "grad_norm": 4.1172966957092285,
      "learning_rate": 6.81719649737303e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 73181072,
      "step": 4548
    },
    {
      "epoch": 0.31864946982234016,
      "grad_norm": 3.849090337753296,
      "learning_rate": 6.81649667250438e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 73197456,
      "step": 4549
    },
    {
      "epoch": 0.31871951806806936,
      "grad_norm": 6.241509437561035,
      "learning_rate": 6.815796847635728e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 73213568,
      "step": 4550
    },
    {
      "epoch": 0.3187895663137986,
      "grad_norm": 4.039997577667236,
      "learning_rate": 6.815097022767075e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 73229648,
      "step": 4551
    },
    {
      "epoch": 0.3188596145595279,
      "grad_norm": 3.757549285888672,
      "learning_rate": 6.814397197898424e-05,
      "loss": 1.1547,
      "num_input_tokens_seen": 73245952,
      "step": 4552
    },
    {
      "epoch": 0.31892966280525714,
      "grad_norm": 4.177220821380615,
      "learning_rate": 6.813697373029773e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 73262336,
      "step": 4553
    },
    {
      "epoch": 0.31899971105098635,
      "grad_norm": 3.659167766571045,
      "learning_rate": 6.812997548161122e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 73278304,
      "step": 4554
    },
    {
      "epoch": 0.3190697592967156,
      "grad_norm": 4.289649486541748,
      "learning_rate": 6.81229772329247e-05,
      "loss": 0.8452,
      "num_input_tokens_seen": 73294320,
      "step": 4555
    },
    {
      "epoch": 0.31913980754244486,
      "grad_norm": 4.452631950378418,
      "learning_rate": 6.811597898423819e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 73310256,
      "step": 4556
    },
    {
      "epoch": 0.3192098557881741,
      "grad_norm": 3.572444438934326,
      "learning_rate": 6.810898073555167e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 73326640,
      "step": 4557
    },
    {
      "epoch": 0.31927990403390333,
      "grad_norm": 4.059347629547119,
      "learning_rate": 6.810198248686514e-05,
      "loss": 1.0103,
      "num_input_tokens_seen": 73342096,
      "step": 4558
    },
    {
      "epoch": 0.3193499522796326,
      "grad_norm": 5.144520282745361,
      "learning_rate": 6.809498423817863e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 73358480,
      "step": 4559
    },
    {
      "epoch": 0.31942000052536185,
      "grad_norm": 4.210456848144531,
      "learning_rate": 6.808798598949212e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 73374864,
      "step": 4560
    },
    {
      "epoch": 0.3194900487710911,
      "grad_norm": 5.06007194519043,
      "learning_rate": 6.808098774080561e-05,
      "loss": 0.8933,
      "num_input_tokens_seen": 73391248,
      "step": 4561
    },
    {
      "epoch": 0.3195600970168203,
      "grad_norm": 4.032425403594971,
      "learning_rate": 6.80739894921191e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 73406728,
      "step": 4562
    },
    {
      "epoch": 0.31963014526254957,
      "grad_norm": 4.344507694244385,
      "learning_rate": 6.806699124343259e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 73423112,
      "step": 4563
    },
    {
      "epoch": 0.31970019350827883,
      "grad_norm": 3.7113993167877197,
      "learning_rate": 6.805999299474606e-05,
      "loss": 0.9122,
      "num_input_tokens_seen": 73439496,
      "step": 4564
    },
    {
      "epoch": 0.3197702417540081,
      "grad_norm": 4.160495281219482,
      "learning_rate": 6.805299474605955e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 73455456,
      "step": 4565
    },
    {
      "epoch": 0.3198402899997373,
      "grad_norm": 5.51431131362915,
      "learning_rate": 6.804599649737304e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 73471840,
      "step": 4566
    },
    {
      "epoch": 0.31991033824546655,
      "grad_norm": 4.145261287689209,
      "learning_rate": 6.803899824868651e-05,
      "loss": 0.947,
      "num_input_tokens_seen": 73487688,
      "step": 4567
    },
    {
      "epoch": 0.3199803864911958,
      "grad_norm": 3.917922019958496,
      "learning_rate": 6.8032e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 73504072,
      "step": 4568
    },
    {
      "epoch": 0.32005043473692507,
      "grad_norm": 3.8644864559173584,
      "learning_rate": 6.802500175131349e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 73520344,
      "step": 4569
    },
    {
      "epoch": 0.32012048298265433,
      "grad_norm": 4.043839931488037,
      "learning_rate": 6.801800350262698e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 73536248,
      "step": 4570
    },
    {
      "epoch": 0.32019053122838353,
      "grad_norm": 4.793722629547119,
      "learning_rate": 6.801100525394045e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 73552512,
      "step": 4571
    },
    {
      "epoch": 0.3202605794741128,
      "grad_norm": 3.632899761199951,
      "learning_rate": 6.800400700525394e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 73568896,
      "step": 4572
    },
    {
      "epoch": 0.32033062771984205,
      "grad_norm": 6.236395359039307,
      "learning_rate": 6.799700875656743e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 73585280,
      "step": 4573
    },
    {
      "epoch": 0.3204006759655713,
      "grad_norm": 4.591775417327881,
      "learning_rate": 6.799001050788092e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 73600328,
      "step": 4574
    },
    {
      "epoch": 0.3204707242113005,
      "grad_norm": 3.9546539783477783,
      "learning_rate": 6.79830122591944e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 73616568,
      "step": 4575
    },
    {
      "epoch": 0.3205407724570298,
      "grad_norm": 4.425241470336914,
      "learning_rate": 6.79760140105079e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 73632552,
      "step": 4576
    },
    {
      "epoch": 0.32061082070275904,
      "grad_norm": 3.999953508377075,
      "learning_rate": 6.796901576182137e-05,
      "loss": 1.1854,
      "num_input_tokens_seen": 73648672,
      "step": 4577
    },
    {
      "epoch": 0.3206808689484883,
      "grad_norm": 3.6718766689300537,
      "learning_rate": 6.796201751313485e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 73665056,
      "step": 4578
    },
    {
      "epoch": 0.3207509171942175,
      "grad_norm": 4.37136173248291,
      "learning_rate": 6.795501926444834e-05,
      "loss": 0.9921,
      "num_input_tokens_seen": 73679680,
      "step": 4579
    },
    {
      "epoch": 0.32082096543994676,
      "grad_norm": 5.109454154968262,
      "learning_rate": 6.794802101576182e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 73695768,
      "step": 4580
    },
    {
      "epoch": 0.320891013685676,
      "grad_norm": 4.277298927307129,
      "learning_rate": 6.794102276707531e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 73711840,
      "step": 4581
    },
    {
      "epoch": 0.3209610619314053,
      "grad_norm": 4.598893165588379,
      "learning_rate": 6.79340245183888e-05,
      "loss": 1.1733,
      "num_input_tokens_seen": 73728184,
      "step": 4582
    },
    {
      "epoch": 0.3210311101771345,
      "grad_norm": 5.124484539031982,
      "learning_rate": 6.792702626970229e-05,
      "loss": 0.9399,
      "num_input_tokens_seen": 73744568,
      "step": 4583
    },
    {
      "epoch": 0.32110115842286374,
      "grad_norm": 4.426584243774414,
      "learning_rate": 6.792002802101577e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 73760424,
      "step": 4584
    },
    {
      "epoch": 0.321171206668593,
      "grad_norm": 3.5181384086608887,
      "learning_rate": 6.791302977232924e-05,
      "loss": 0.8025,
      "num_input_tokens_seen": 73776808,
      "step": 4585
    },
    {
      "epoch": 0.32124125491432226,
      "grad_norm": 6.614295482635498,
      "learning_rate": 6.790603152364273e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 73793192,
      "step": 4586
    },
    {
      "epoch": 0.32131130316005146,
      "grad_norm": 5.212308406829834,
      "learning_rate": 6.789903327495622e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 73809576,
      "step": 4587
    },
    {
      "epoch": 0.3213813514057807,
      "grad_norm": 4.7378106117248535,
      "learning_rate": 6.78920350262697e-05,
      "loss": 1.112,
      "num_input_tokens_seen": 73825680,
      "step": 4588
    },
    {
      "epoch": 0.32145139965151,
      "grad_norm": 5.0195136070251465,
      "learning_rate": 6.78850367775832e-05,
      "loss": 1.4437,
      "num_input_tokens_seen": 73841200,
      "step": 4589
    },
    {
      "epoch": 0.32152144789723924,
      "grad_norm": 6.186412811279297,
      "learning_rate": 6.787803852889668e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 73857584,
      "step": 4590
    },
    {
      "epoch": 0.32159149614296845,
      "grad_norm": 6.835412502288818,
      "learning_rate": 6.787104028021016e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 73873624,
      "step": 4591
    },
    {
      "epoch": 0.3216615443886977,
      "grad_norm": 4.3859333992004395,
      "learning_rate": 6.786404203152365e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 73890008,
      "step": 4592
    },
    {
      "epoch": 0.32173159263442697,
      "grad_norm": 3.8230555057525635,
      "learning_rate": 6.785704378283714e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 73906392,
      "step": 4593
    },
    {
      "epoch": 0.3218016408801562,
      "grad_norm": 4.458274841308594,
      "learning_rate": 6.785004553415062e-05,
      "loss": 1.1223,
      "num_input_tokens_seen": 73922776,
      "step": 4594
    },
    {
      "epoch": 0.32187168912588543,
      "grad_norm": 4.006426811218262,
      "learning_rate": 6.78430472854641e-05,
      "loss": 1.3019,
      "num_input_tokens_seen": 73938896,
      "step": 4595
    },
    {
      "epoch": 0.3219417373716147,
      "grad_norm": 4.637386322021484,
      "learning_rate": 6.783604903677759e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 73955280,
      "step": 4596
    },
    {
      "epoch": 0.32201178561734395,
      "grad_norm": 5.13168478012085,
      "learning_rate": 6.782905078809108e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 73971480,
      "step": 4597
    },
    {
      "epoch": 0.3220818338630732,
      "grad_norm": 3.8248770236968994,
      "learning_rate": 6.782205253940455e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 73987712,
      "step": 4598
    },
    {
      "epoch": 0.3221518821088024,
      "grad_norm": 5.167041778564453,
      "learning_rate": 6.781505429071804e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 74004096,
      "step": 4599
    },
    {
      "epoch": 0.32222193035453167,
      "grad_norm": 3.779311180114746,
      "learning_rate": 6.780805604203153e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 74020176,
      "step": 4600
    },
    {
      "epoch": 0.32222193035453167,
      "eval_loss": 1.1318858861923218,
      "eval_runtime": 0.2027,
      "eval_samples_per_second": 4.933,
      "eval_steps_per_second": 4.933,
      "num_input_tokens_seen": 74020176,
      "step": 4600
    },
    {
      "epoch": 0.32229197860026093,
      "grad_norm": 3.8468148708343506,
      "learning_rate": 6.780105779334502e-05,
      "loss": 0.9602,
      "num_input_tokens_seen": 74035664,
      "step": 4601
    },
    {
      "epoch": 0.3223620268459902,
      "grad_norm": 4.28491735458374,
      "learning_rate": 6.779405954465849e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 74050408,
      "step": 4602
    },
    {
      "epoch": 0.3224320750917194,
      "grad_norm": 4.872751712799072,
      "learning_rate": 6.7787061295972e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 74066336,
      "step": 4603
    },
    {
      "epoch": 0.32250212333744865,
      "grad_norm": 4.060647487640381,
      "learning_rate": 6.778006304728547e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 74082720,
      "step": 4604
    },
    {
      "epoch": 0.3225721715831779,
      "grad_norm": 3.607623815536499,
      "learning_rate": 6.777306479859894e-05,
      "loss": 0.9797,
      "num_input_tokens_seen": 74099104,
      "step": 4605
    },
    {
      "epoch": 0.3226422198289072,
      "grad_norm": 3.719801187515259,
      "learning_rate": 6.776606654991243e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 74115488,
      "step": 4606
    },
    {
      "epoch": 0.32271226807463643,
      "grad_norm": 5.072197914123535,
      "learning_rate": 6.775906830122592e-05,
      "loss": 1.1264,
      "num_input_tokens_seen": 74131224,
      "step": 4607
    },
    {
      "epoch": 0.32278231632036564,
      "grad_norm": 6.052949905395508,
      "learning_rate": 6.775207005253941e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 74147608,
      "step": 4608
    },
    {
      "epoch": 0.3228523645660949,
      "grad_norm": 6.214832782745361,
      "learning_rate": 6.77450718038529e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 74162840,
      "step": 4609
    },
    {
      "epoch": 0.32292241281182416,
      "grad_norm": 4.279264450073242,
      "learning_rate": 6.773807355516639e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 74179224,
      "step": 4610
    },
    {
      "epoch": 0.3229924610575534,
      "grad_norm": 3.8564460277557373,
      "learning_rate": 6.773107530647986e-05,
      "loss": 1.0555,
      "num_input_tokens_seen": 74195608,
      "step": 4611
    },
    {
      "epoch": 0.3230625093032826,
      "grad_norm": 4.747770309448242,
      "learning_rate": 6.772407705779334e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 74211664,
      "step": 4612
    },
    {
      "epoch": 0.3231325575490119,
      "grad_norm": 3.5425655841827393,
      "learning_rate": 6.771707880910683e-05,
      "loss": 1.135,
      "num_input_tokens_seen": 74228048,
      "step": 4613
    },
    {
      "epoch": 0.32320260579474114,
      "grad_norm": 3.919851303100586,
      "learning_rate": 6.771008056042033e-05,
      "loss": 0.9791,
      "num_input_tokens_seen": 74243424,
      "step": 4614
    },
    {
      "epoch": 0.3232726540404704,
      "grad_norm": 4.061427593231201,
      "learning_rate": 6.77030823117338e-05,
      "loss": 1.2477,
      "num_input_tokens_seen": 74259696,
      "step": 4615
    },
    {
      "epoch": 0.3233427022861996,
      "grad_norm": 5.14341926574707,
      "learning_rate": 6.769608406304729e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 74274968,
      "step": 4616
    },
    {
      "epoch": 0.32341275053192886,
      "grad_norm": 6.207670211791992,
      "learning_rate": 6.768908581436078e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 74291352,
      "step": 4617
    },
    {
      "epoch": 0.3234827987776581,
      "grad_norm": 5.948925971984863,
      "learning_rate": 6.768208756567426e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 74307000,
      "step": 4618
    },
    {
      "epoch": 0.3235528470233874,
      "grad_norm": 5.205277442932129,
      "learning_rate": 6.767508931698774e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 74323384,
      "step": 4619
    },
    {
      "epoch": 0.3236228952691166,
      "grad_norm": 3.8878557682037354,
      "learning_rate": 6.766809106830123e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 74339768,
      "step": 4620
    },
    {
      "epoch": 0.32369294351484584,
      "grad_norm": 4.9194111824035645,
      "learning_rate": 6.766109281961472e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 74355888,
      "step": 4621
    },
    {
      "epoch": 0.3237629917605751,
      "grad_norm": 3.5212655067443848,
      "learning_rate": 6.76540945709282e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 74372048,
      "step": 4622
    },
    {
      "epoch": 0.32383304000630436,
      "grad_norm": 3.6212568283081055,
      "learning_rate": 6.764709632224168e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 74388432,
      "step": 4623
    },
    {
      "epoch": 0.32390308825203357,
      "grad_norm": 3.795515298843384,
      "learning_rate": 6.764009807355517e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 74404584,
      "step": 4624
    },
    {
      "epoch": 0.3239731364977628,
      "grad_norm": 4.537838935852051,
      "learning_rate": 6.763309982486865e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 74420304,
      "step": 4625
    },
    {
      "epoch": 0.3240431847434921,
      "grad_norm": 4.276764392852783,
      "learning_rate": 6.762610157618214e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 74436688,
      "step": 4626
    },
    {
      "epoch": 0.32411323298922134,
      "grad_norm": 3.9739227294921875,
      "learning_rate": 6.761910332749563e-05,
      "loss": 1.0002,
      "num_input_tokens_seen": 74451824,
      "step": 4627
    },
    {
      "epoch": 0.32418328123495055,
      "grad_norm": 4.176823616027832,
      "learning_rate": 6.761210507880911e-05,
      "loss": 1.2547,
      "num_input_tokens_seen": 74467080,
      "step": 4628
    },
    {
      "epoch": 0.3242533294806798,
      "grad_norm": 4.471405029296875,
      "learning_rate": 6.760510683012259e-05,
      "loss": 0.9694,
      "num_input_tokens_seen": 74483464,
      "step": 4629
    },
    {
      "epoch": 0.32432337772640907,
      "grad_norm": 3.95442271232605,
      "learning_rate": 6.759810858143609e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 74499848,
      "step": 4630
    },
    {
      "epoch": 0.3243934259721383,
      "grad_norm": 5.348501682281494,
      "learning_rate": 6.759111033274957e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 74516232,
      "step": 4631
    },
    {
      "epoch": 0.32446347421786753,
      "grad_norm": 4.405150413513184,
      "learning_rate": 6.758411208406304e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 74531120,
      "step": 4632
    },
    {
      "epoch": 0.3245335224635968,
      "grad_norm": 3.633358955383301,
      "learning_rate": 6.757711383537653e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 74547504,
      "step": 4633
    },
    {
      "epoch": 0.32460357070932605,
      "grad_norm": 4.668785095214844,
      "learning_rate": 6.757011558669003e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 74563888,
      "step": 4634
    },
    {
      "epoch": 0.3246736189550553,
      "grad_norm": 5.222908020019531,
      "learning_rate": 6.756311733800351e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 74580224,
      "step": 4635
    },
    {
      "epoch": 0.3247436672007845,
      "grad_norm": 3.812385082244873,
      "learning_rate": 6.7556119089317e-05,
      "loss": 1.1326,
      "num_input_tokens_seen": 74596608,
      "step": 4636
    },
    {
      "epoch": 0.3248137154465138,
      "grad_norm": 5.080833911895752,
      "learning_rate": 6.754912084063048e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 74612456,
      "step": 4637
    },
    {
      "epoch": 0.32488376369224303,
      "grad_norm": 5.309609413146973,
      "learning_rate": 6.754212259194396e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 74627840,
      "step": 4638
    },
    {
      "epoch": 0.3249538119379723,
      "grad_norm": 4.46236515045166,
      "learning_rate": 6.753512434325743e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 74643800,
      "step": 4639
    },
    {
      "epoch": 0.32502386018370155,
      "grad_norm": 9.981855392456055,
      "learning_rate": 6.752812609457094e-05,
      "loss": 1.2777,
      "num_input_tokens_seen": 74660184,
      "step": 4640
    },
    {
      "epoch": 0.32509390842943076,
      "grad_norm": 5.075852870941162,
      "learning_rate": 6.752112784588443e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 74676568,
      "step": 4641
    },
    {
      "epoch": 0.32516395667516,
      "grad_norm": 3.8985090255737305,
      "learning_rate": 6.75141295971979e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 74692952,
      "step": 4642
    },
    {
      "epoch": 0.3252340049208893,
      "grad_norm": 4.9769673347473145,
      "learning_rate": 6.750713134851139e-05,
      "loss": 1.3023,
      "num_input_tokens_seen": 74709216,
      "step": 4643
    },
    {
      "epoch": 0.32530405316661853,
      "grad_norm": 4.508238315582275,
      "learning_rate": 6.750013309982488e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 74724640,
      "step": 4644
    },
    {
      "epoch": 0.32537410141234774,
      "grad_norm": 4.214225769042969,
      "learning_rate": 6.749313485113835e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 74740696,
      "step": 4645
    },
    {
      "epoch": 0.325444149658077,
      "grad_norm": 4.217604160308838,
      "learning_rate": 6.748613660245184e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 74756520,
      "step": 4646
    },
    {
      "epoch": 0.32551419790380626,
      "grad_norm": 3.5975253582000732,
      "learning_rate": 6.747913835376533e-05,
      "loss": 0.9642,
      "num_input_tokens_seen": 74772904,
      "step": 4647
    },
    {
      "epoch": 0.3255842461495355,
      "grad_norm": 3.5055267810821533,
      "learning_rate": 6.747214010507882e-05,
      "loss": 1.0975,
      "num_input_tokens_seen": 74789288,
      "step": 4648
    },
    {
      "epoch": 0.3256542943952647,
      "grad_norm": 3.8605833053588867,
      "learning_rate": 6.746514185639229e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 74804768,
      "step": 4649
    },
    {
      "epoch": 0.325724342640994,
      "grad_norm": 9.446599006652832,
      "learning_rate": 6.745814360770578e-05,
      "loss": 1.1894,
      "num_input_tokens_seen": 74821152,
      "step": 4650
    },
    {
      "epoch": 0.32579439088672324,
      "grad_norm": 4.161158084869385,
      "learning_rate": 6.745114535901927e-05,
      "loss": 0.984,
      "num_input_tokens_seen": 74836992,
      "step": 4651
    },
    {
      "epoch": 0.3258644391324525,
      "grad_norm": 3.5690324306488037,
      "learning_rate": 6.744414711033275e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 74852896,
      "step": 4652
    },
    {
      "epoch": 0.3259344873781817,
      "grad_norm": 3.5873210430145264,
      "learning_rate": 6.743714886164623e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 74868472,
      "step": 4653
    },
    {
      "epoch": 0.32600453562391096,
      "grad_norm": 4.192559719085693,
      "learning_rate": 6.743015061295972e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 74884856,
      "step": 4654
    },
    {
      "epoch": 0.3260745838696402,
      "grad_norm": 4.633018493652344,
      "learning_rate": 6.742315236427321e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 74900848,
      "step": 4655
    },
    {
      "epoch": 0.3261446321153695,
      "grad_norm": 3.568934440612793,
      "learning_rate": 6.741615411558669e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 74917232,
      "step": 4656
    },
    {
      "epoch": 0.3262146803610987,
      "grad_norm": 3.6099655628204346,
      "learning_rate": 6.740915586690019e-05,
      "loss": 1.0758,
      "num_input_tokens_seen": 74933616,
      "step": 4657
    },
    {
      "epoch": 0.32628472860682795,
      "grad_norm": 4.272975921630859,
      "learning_rate": 6.740215761821366e-05,
      "loss": 1.1901,
      "num_input_tokens_seen": 74950000,
      "step": 4658
    },
    {
      "epoch": 0.3263547768525572,
      "grad_norm": 4.2752251625061035,
      "learning_rate": 6.739515936952714e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 74966032,
      "step": 4659
    },
    {
      "epoch": 0.32642482509828646,
      "grad_norm": 5.06410551071167,
      "learning_rate": 6.738816112084064e-05,
      "loss": 1.2041,
      "num_input_tokens_seen": 74981432,
      "step": 4660
    },
    {
      "epoch": 0.32649487334401567,
      "grad_norm": 6.378856182098389,
      "learning_rate": 6.738116287215413e-05,
      "loss": 1.2996,
      "num_input_tokens_seen": 74997440,
      "step": 4661
    },
    {
      "epoch": 0.3265649215897449,
      "grad_norm": 5.427485466003418,
      "learning_rate": 6.73741646234676e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 75013824,
      "step": 4662
    },
    {
      "epoch": 0.3266349698354742,
      "grad_norm": 4.366839408874512,
      "learning_rate": 6.736716637478109e-05,
      "loss": 1.2077,
      "num_input_tokens_seen": 75030208,
      "step": 4663
    },
    {
      "epoch": 0.32670501808120345,
      "grad_norm": 5.765005588531494,
      "learning_rate": 6.736016812609458e-05,
      "loss": 1.0833,
      "num_input_tokens_seen": 75046592,
      "step": 4664
    },
    {
      "epoch": 0.32677506632693265,
      "grad_norm": 3.4886975288391113,
      "learning_rate": 6.735316987740806e-05,
      "loss": 0.7976,
      "num_input_tokens_seen": 75062976,
      "step": 4665
    },
    {
      "epoch": 0.3268451145726619,
      "grad_norm": 4.1105875968933105,
      "learning_rate": 6.734617162872154e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 75078024,
      "step": 4666
    },
    {
      "epoch": 0.32691516281839117,
      "grad_norm": 3.8737053871154785,
      "learning_rate": 6.733917338003503e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 75094408,
      "step": 4667
    },
    {
      "epoch": 0.32698521106412043,
      "grad_norm": 4.077807426452637,
      "learning_rate": 6.733217513134852e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 75110792,
      "step": 4668
    },
    {
      "epoch": 0.32705525930984963,
      "grad_norm": 4.339305400848389,
      "learning_rate": 6.7325176882662e-05,
      "loss": 0.7132,
      "num_input_tokens_seen": 75126240,
      "step": 4669
    },
    {
      "epoch": 0.3271253075555789,
      "grad_norm": 4.241507053375244,
      "learning_rate": 6.731817863397549e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 75142144,
      "step": 4670
    },
    {
      "epoch": 0.32719535580130815,
      "grad_norm": 7.518558979034424,
      "learning_rate": 6.731118038528897e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 75158528,
      "step": 4671
    },
    {
      "epoch": 0.3272654040470374,
      "grad_norm": 4.342295169830322,
      "learning_rate": 6.730418213660245e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 75174912,
      "step": 4672
    },
    {
      "epoch": 0.3273354522927666,
      "grad_norm": 3.3599188327789307,
      "learning_rate": 6.729718388791594e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 75190720,
      "step": 4673
    },
    {
      "epoch": 0.3274055005384959,
      "grad_norm": 4.393617153167725,
      "learning_rate": 6.729018563922943e-05,
      "loss": 1.1215,
      "num_input_tokens_seen": 75207104,
      "step": 4674
    },
    {
      "epoch": 0.32747554878422513,
      "grad_norm": 3.948538064956665,
      "learning_rate": 6.728318739054292e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 75222736,
      "step": 4675
    },
    {
      "epoch": 0.3275455970299544,
      "grad_norm": 5.3323469161987305,
      "learning_rate": 6.727618914185639e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 75238680,
      "step": 4676
    },
    {
      "epoch": 0.32761564527568365,
      "grad_norm": 4.943187713623047,
      "learning_rate": 6.726919089316988e-05,
      "loss": 0.9327,
      "num_input_tokens_seen": 75255064,
      "step": 4677
    },
    {
      "epoch": 0.32768569352141286,
      "grad_norm": 4.083932399749756,
      "learning_rate": 6.726219264448337e-05,
      "loss": 1.2085,
      "num_input_tokens_seen": 75271448,
      "step": 4678
    },
    {
      "epoch": 0.3277557417671421,
      "grad_norm": 4.682622909545898,
      "learning_rate": 6.725519439579684e-05,
      "loss": 1.0105,
      "num_input_tokens_seen": 75287752,
      "step": 4679
    },
    {
      "epoch": 0.3278257900128714,
      "grad_norm": 4.544816493988037,
      "learning_rate": 6.724819614711033e-05,
      "loss": 1.0422,
      "num_input_tokens_seen": 75304136,
      "step": 4680
    },
    {
      "epoch": 0.32789583825860064,
      "grad_norm": 3.859891176223755,
      "learning_rate": 6.724119789842383e-05,
      "loss": 0.9317,
      "num_input_tokens_seen": 75320520,
      "step": 4681
    },
    {
      "epoch": 0.32796588650432984,
      "grad_norm": 5.739070415496826,
      "learning_rate": 6.723419964973731e-05,
      "loss": 1.1315,
      "num_input_tokens_seen": 75336904,
      "step": 4682
    },
    {
      "epoch": 0.3280359347500591,
      "grad_norm": 4.289483547210693,
      "learning_rate": 6.722720140105078e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 75353288,
      "step": 4683
    },
    {
      "epoch": 0.32810598299578836,
      "grad_norm": 4.03695011138916,
      "learning_rate": 6.722020315236429e-05,
      "loss": 1.0129,
      "num_input_tokens_seen": 75369424,
      "step": 4684
    },
    {
      "epoch": 0.3281760312415176,
      "grad_norm": 3.8941352367401123,
      "learning_rate": 6.721320490367776e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 75385760,
      "step": 4685
    },
    {
      "epoch": 0.3282460794872468,
      "grad_norm": 4.345769882202148,
      "learning_rate": 6.720620665499125e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 75401736,
      "step": 4686
    },
    {
      "epoch": 0.3283161277329761,
      "grad_norm": 5.759182453155518,
      "learning_rate": 6.719920840630474e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 75417928,
      "step": 4687
    },
    {
      "epoch": 0.32838617597870534,
      "grad_norm": 4.947919845581055,
      "learning_rate": 6.719221015761823e-05,
      "loss": 1.0346,
      "num_input_tokens_seen": 75433624,
      "step": 4688
    },
    {
      "epoch": 0.3284562242244346,
      "grad_norm": 3.936934471130371,
      "learning_rate": 6.71852119089317e-05,
      "loss": 1.1684,
      "num_input_tokens_seen": 75450008,
      "step": 4689
    },
    {
      "epoch": 0.3285262724701638,
      "grad_norm": 3.7944555282592773,
      "learning_rate": 6.717821366024519e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 75466392,
      "step": 4690
    },
    {
      "epoch": 0.32859632071589306,
      "grad_norm": 3.8094451427459717,
      "learning_rate": 6.717121541155868e-05,
      "loss": 0.9309,
      "num_input_tokens_seen": 75482776,
      "step": 4691
    },
    {
      "epoch": 0.3286663689616223,
      "grad_norm": 4.426685333251953,
      "learning_rate": 6.716421716287215e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 75497760,
      "step": 4692
    },
    {
      "epoch": 0.3287364172073516,
      "grad_norm": 4.299224376678467,
      "learning_rate": 6.715721891418564e-05,
      "loss": 1.214,
      "num_input_tokens_seen": 75513024,
      "step": 4693
    },
    {
      "epoch": 0.3288064654530808,
      "grad_norm": 3.765477418899536,
      "learning_rate": 6.715022066549913e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 75529304,
      "step": 4694
    },
    {
      "epoch": 0.32887651369881005,
      "grad_norm": 3.991591453552246,
      "learning_rate": 6.714322241681262e-05,
      "loss": 0.8295,
      "num_input_tokens_seen": 75545264,
      "step": 4695
    },
    {
      "epoch": 0.3289465619445393,
      "grad_norm": 3.652726888656616,
      "learning_rate": 6.71362241681261e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 75561648,
      "step": 4696
    },
    {
      "epoch": 0.32901661019026857,
      "grad_norm": 6.083689212799072,
      "learning_rate": 6.712922591943958e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 75578032,
      "step": 4697
    },
    {
      "epoch": 0.32908665843599777,
      "grad_norm": 4.732533931732178,
      "learning_rate": 6.712222767075307e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 75593944,
      "step": 4698
    },
    {
      "epoch": 0.32915670668172703,
      "grad_norm": 5.024901866912842,
      "learning_rate": 6.711522942206655e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 75610328,
      "step": 4699
    },
    {
      "epoch": 0.3292267549274563,
      "grad_norm": 4.663429260253906,
      "learning_rate": 6.710823117338004e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 75626712,
      "step": 4700
    },
    {
      "epoch": 0.32929680317318555,
      "grad_norm": 4.396904945373535,
      "learning_rate": 6.710123292469354e-05,
      "loss": 1.2419,
      "num_input_tokens_seen": 75643096,
      "step": 4701
    },
    {
      "epoch": 0.32936685141891475,
      "grad_norm": 3.7963149547576904,
      "learning_rate": 6.709423467600701e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 75658616,
      "step": 4702
    },
    {
      "epoch": 0.329436899664644,
      "grad_norm": 4.154513835906982,
      "learning_rate": 6.708723642732049e-05,
      "loss": 1.0529,
      "num_input_tokens_seen": 75675000,
      "step": 4703
    },
    {
      "epoch": 0.32950694791037327,
      "grad_norm": 3.8939032554626465,
      "learning_rate": 6.708023817863398e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 75690728,
      "step": 4704
    },
    {
      "epoch": 0.32957699615610253,
      "grad_norm": 4.7678375244140625,
      "learning_rate": 6.707323992994746e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 75707080,
      "step": 4705
    },
    {
      "epoch": 0.32964704440183173,
      "grad_norm": 6.56498384475708,
      "learning_rate": 6.706624168126094e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 75723464,
      "step": 4706
    },
    {
      "epoch": 0.329717092647561,
      "grad_norm": 6.917506694793701,
      "learning_rate": 6.705924343257444e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 75739848,
      "step": 4707
    },
    {
      "epoch": 0.32978714089329025,
      "grad_norm": 3.9431846141815186,
      "learning_rate": 6.705224518388793e-05,
      "loss": 0.9693,
      "num_input_tokens_seen": 75756232,
      "step": 4708
    },
    {
      "epoch": 0.3298571891390195,
      "grad_norm": 4.838469505310059,
      "learning_rate": 6.70452469352014e-05,
      "loss": 1.2367,
      "num_input_tokens_seen": 75772616,
      "step": 4709
    },
    {
      "epoch": 0.3299272373847488,
      "grad_norm": 4.0371012687683105,
      "learning_rate": 6.703824868651488e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 75789000,
      "step": 4710
    },
    {
      "epoch": 0.329997285630478,
      "grad_norm": 3.491875410079956,
      "learning_rate": 6.703125043782838e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 75805384,
      "step": 4711
    },
    {
      "epoch": 0.33006733387620724,
      "grad_norm": 3.5304512977600098,
      "learning_rate": 6.702425218914186e-05,
      "loss": 0.8896,
      "num_input_tokens_seen": 75821104,
      "step": 4712
    },
    {
      "epoch": 0.3301373821219365,
      "grad_norm": 3.642528533935547,
      "learning_rate": 6.701725394045535e-05,
      "loss": 0.9843,
      "num_input_tokens_seen": 75837424,
      "step": 4713
    },
    {
      "epoch": 0.33020743036766576,
      "grad_norm": 6.536950588226318,
      "learning_rate": 6.701025569176883e-05,
      "loss": 0.9545,
      "num_input_tokens_seen": 75853808,
      "step": 4714
    },
    {
      "epoch": 0.33027747861339496,
      "grad_norm": 3.376460075378418,
      "learning_rate": 6.700325744308232e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 75870192,
      "step": 4715
    },
    {
      "epoch": 0.3303475268591242,
      "grad_norm": 4.988052845001221,
      "learning_rate": 6.69962591943958e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 75886576,
      "step": 4716
    },
    {
      "epoch": 0.3304175751048535,
      "grad_norm": 4.724236965179443,
      "learning_rate": 6.698926094570929e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 75902960,
      "step": 4717
    },
    {
      "epoch": 0.33048762335058274,
      "grad_norm": 4.877357006072998,
      "learning_rate": 6.698226269702278e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 75919344,
      "step": 4718
    },
    {
      "epoch": 0.33055767159631194,
      "grad_norm": 3.981224775314331,
      "learning_rate": 6.697526444833625e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 75935728,
      "step": 4719
    },
    {
      "epoch": 0.3306277198420412,
      "grad_norm": 5.456554889678955,
      "learning_rate": 6.696826619964974e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 75951304,
      "step": 4720
    },
    {
      "epoch": 0.33069776808777046,
      "grad_norm": 3.9885287284851074,
      "learning_rate": 6.696126795096323e-05,
      "loss": 1.2658,
      "num_input_tokens_seen": 75967688,
      "step": 4721
    },
    {
      "epoch": 0.3307678163334997,
      "grad_norm": 3.447371482849121,
      "learning_rate": 6.695426970227672e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 75984072,
      "step": 4722
    },
    {
      "epoch": 0.3308378645792289,
      "grad_norm": 4.405709743499756,
      "learning_rate": 6.694727145359019e-05,
      "loss": 1.2445,
      "num_input_tokens_seen": 76000456,
      "step": 4723
    },
    {
      "epoch": 0.3309079128249582,
      "grad_norm": 3.7595372200012207,
      "learning_rate": 6.694027320490368e-05,
      "loss": 1.1851,
      "num_input_tokens_seen": 76016840,
      "step": 4724
    },
    {
      "epoch": 0.33097796107068744,
      "grad_norm": 5.460091590881348,
      "learning_rate": 6.693327495621717e-05,
      "loss": 0.8514,
      "num_input_tokens_seen": 76032344,
      "step": 4725
    },
    {
      "epoch": 0.3310480093164167,
      "grad_norm": 7.111250400543213,
      "learning_rate": 6.692627670753064e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 76048728,
      "step": 4726
    },
    {
      "epoch": 0.3311180575621459,
      "grad_norm": 4.799232482910156,
      "learning_rate": 6.691927845884415e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 76063832,
      "step": 4727
    },
    {
      "epoch": 0.33118810580787517,
      "grad_norm": 4.045900344848633,
      "learning_rate": 6.691228021015763e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 76079792,
      "step": 4728
    },
    {
      "epoch": 0.3312581540536044,
      "grad_norm": 3.9942305088043213,
      "learning_rate": 6.690528196147111e-05,
      "loss": 1.1444,
      "num_input_tokens_seen": 76095992,
      "step": 4729
    },
    {
      "epoch": 0.3313282022993337,
      "grad_norm": 4.173962116241455,
      "learning_rate": 6.689828371278458e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 76111760,
      "step": 4730
    },
    {
      "epoch": 0.3313982505450629,
      "grad_norm": 8.357215881347656,
      "learning_rate": 6.689128546409807e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 76127184,
      "step": 4731
    },
    {
      "epoch": 0.33146829879079215,
      "grad_norm": 3.7359249591827393,
      "learning_rate": 6.688428721541156e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 76143536,
      "step": 4732
    },
    {
      "epoch": 0.3315383470365214,
      "grad_norm": 4.159603595733643,
      "learning_rate": 6.687728896672505e-05,
      "loss": 1.1565,
      "num_input_tokens_seen": 76159640,
      "step": 4733
    },
    {
      "epoch": 0.33160839528225067,
      "grad_norm": 4.893441200256348,
      "learning_rate": 6.687029071803854e-05,
      "loss": 1.191,
      "num_input_tokens_seen": 76176024,
      "step": 4734
    },
    {
      "epoch": 0.33167844352797987,
      "grad_norm": 4.4292426109313965,
      "learning_rate": 6.686329246935203e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 76192408,
      "step": 4735
    },
    {
      "epoch": 0.33174849177370913,
      "grad_norm": 3.612821102142334,
      "learning_rate": 6.68562942206655e-05,
      "loss": 1.0195,
      "num_input_tokens_seen": 76208792,
      "step": 4736
    },
    {
      "epoch": 0.3318185400194384,
      "grad_norm": 3.6046557426452637,
      "learning_rate": 6.684929597197898e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 76225176,
      "step": 4737
    },
    {
      "epoch": 0.33188858826516765,
      "grad_norm": 4.637216567993164,
      "learning_rate": 6.684229772329248e-05,
      "loss": 0.8202,
      "num_input_tokens_seen": 76241560,
      "step": 4738
    },
    {
      "epoch": 0.33195863651089685,
      "grad_norm": 4.83438777923584,
      "learning_rate": 6.683529947460595e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 76257264,
      "step": 4739
    },
    {
      "epoch": 0.3320286847566261,
      "grad_norm": 3.903982400894165,
      "learning_rate": 6.682830122591944e-05,
      "loss": 1.2306,
      "num_input_tokens_seen": 76273608,
      "step": 4740
    },
    {
      "epoch": 0.3320987330023554,
      "grad_norm": 6.24022102355957,
      "learning_rate": 6.682130297723293e-05,
      "loss": 0.9706,
      "num_input_tokens_seen": 76289840,
      "step": 4741
    },
    {
      "epoch": 0.33216878124808463,
      "grad_norm": 5.286207675933838,
      "learning_rate": 6.681430472854642e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 76306088,
      "step": 4742
    },
    {
      "epoch": 0.3322388294938139,
      "grad_norm": 5.145969867706299,
      "learning_rate": 6.68073064798599e-05,
      "loss": 1.2303,
      "num_input_tokens_seen": 76322152,
      "step": 4743
    },
    {
      "epoch": 0.3323088777395431,
      "grad_norm": 3.6806249618530273,
      "learning_rate": 6.680030823117338e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 76338424,
      "step": 4744
    },
    {
      "epoch": 0.33237892598527236,
      "grad_norm": 3.743912696838379,
      "learning_rate": 6.679330998248687e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 76354808,
      "step": 4745
    },
    {
      "epoch": 0.3324489742310016,
      "grad_norm": 5.072415828704834,
      "learning_rate": 6.678631173380035e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 76369696,
      "step": 4746
    },
    {
      "epoch": 0.3325190224767309,
      "grad_norm": 3.366450548171997,
      "learning_rate": 6.677931348511384e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 76385560,
      "step": 4747
    },
    {
      "epoch": 0.3325890707224601,
      "grad_norm": 3.8318989276885986,
      "learning_rate": 6.677231523642732e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 76401168,
      "step": 4748
    },
    {
      "epoch": 0.33265911896818934,
      "grad_norm": 3.9670164585113525,
      "learning_rate": 6.676531698774081e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 76417552,
      "step": 4749
    },
    {
      "epoch": 0.3327291672139186,
      "grad_norm": 4.344585418701172,
      "learning_rate": 6.675831873905429e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 76433936,
      "step": 4750
    },
    {
      "epoch": 0.33279921545964786,
      "grad_norm": 7.547675132751465,
      "learning_rate": 6.675132049036778e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 76450320,
      "step": 4751
    },
    {
      "epoch": 0.33286926370537706,
      "grad_norm": 7.854677677154541,
      "learning_rate": 6.674432224168127e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 76465696,
      "step": 4752
    },
    {
      "epoch": 0.3329393119511063,
      "grad_norm": 4.030972480773926,
      "learning_rate": 6.673732399299475e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 76482080,
      "step": 4753
    },
    {
      "epoch": 0.3330093601968356,
      "grad_norm": 5.990024089813232,
      "learning_rate": 6.673032574430824e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 76498464,
      "step": 4754
    },
    {
      "epoch": 0.33307940844256484,
      "grad_norm": 3.8437137603759766,
      "learning_rate": 6.672332749562173e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 76514848,
      "step": 4755
    },
    {
      "epoch": 0.33314945668829404,
      "grad_norm": 3.883882761001587,
      "learning_rate": 6.67163292469352e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 76531232,
      "step": 4756
    },
    {
      "epoch": 0.3332195049340233,
      "grad_norm": 4.205630779266357,
      "learning_rate": 6.670933099824868e-05,
      "loss": 1.0847,
      "num_input_tokens_seen": 76547616,
      "step": 4757
    },
    {
      "epoch": 0.33328955317975256,
      "grad_norm": 6.173430442810059,
      "learning_rate": 6.670233274956217e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 76564000,
      "step": 4758
    },
    {
      "epoch": 0.3333596014254818,
      "grad_norm": 3.464181661605835,
      "learning_rate": 6.669533450087566e-05,
      "loss": 0.8751,
      "num_input_tokens_seen": 76579904,
      "step": 4759
    },
    {
      "epoch": 0.333429649671211,
      "grad_norm": 3.3506994247436523,
      "learning_rate": 6.668833625218915e-05,
      "loss": 0.8281,
      "num_input_tokens_seen": 76596288,
      "step": 4760
    },
    {
      "epoch": 0.3334996979169403,
      "grad_norm": 7.188508987426758,
      "learning_rate": 6.668133800350264e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 76611632,
      "step": 4761
    },
    {
      "epoch": 0.33356974616266954,
      "grad_norm": 3.916689872741699,
      "learning_rate": 6.667433975481612e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 76626840,
      "step": 4762
    },
    {
      "epoch": 0.3336397944083988,
      "grad_norm": 3.4827966690063477,
      "learning_rate": 6.66673415061296e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 76643024,
      "step": 4763
    },
    {
      "epoch": 0.333709842654128,
      "grad_norm": 4.479428768157959,
      "learning_rate": 6.666034325744307e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 76659408,
      "step": 4764
    },
    {
      "epoch": 0.33377989089985727,
      "grad_norm": 6.008899211883545,
      "learning_rate": 6.665334500875658e-05,
      "loss": 1.2375,
      "num_input_tokens_seen": 76675272,
      "step": 4765
    },
    {
      "epoch": 0.3338499391455865,
      "grad_norm": 4.10992431640625,
      "learning_rate": 6.664634676007005e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 76691000,
      "step": 4766
    },
    {
      "epoch": 0.3339199873913158,
      "grad_norm": 3.953507423400879,
      "learning_rate": 6.663934851138354e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 76707024,
      "step": 4767
    },
    {
      "epoch": 0.333990035637045,
      "grad_norm": 4.237090587615967,
      "learning_rate": 6.663235026269703e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 76723408,
      "step": 4768
    },
    {
      "epoch": 0.33406008388277425,
      "grad_norm": 4.417295932769775,
      "learning_rate": 6.662535201401052e-05,
      "loss": 0.923,
      "num_input_tokens_seen": 76739296,
      "step": 4769
    },
    {
      "epoch": 0.3341301321285035,
      "grad_norm": 3.664970874786377,
      "learning_rate": 6.661835376532399e-05,
      "loss": 0.9556,
      "num_input_tokens_seen": 76755432,
      "step": 4770
    },
    {
      "epoch": 0.33420018037423277,
      "grad_norm": 3.702932834625244,
      "learning_rate": 6.661135551663748e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 76770688,
      "step": 4771
    },
    {
      "epoch": 0.334270228619962,
      "grad_norm": 3.741722822189331,
      "learning_rate": 6.660435726795097e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 76787072,
      "step": 4772
    },
    {
      "epoch": 0.33434027686569123,
      "grad_norm": 3.9605424404144287,
      "learning_rate": 6.659735901926446e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 76803456,
      "step": 4773
    },
    {
      "epoch": 0.3344103251114205,
      "grad_norm": 3.9401822090148926,
      "learning_rate": 6.659036077057793e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 76819840,
      "step": 4774
    },
    {
      "epoch": 0.33448037335714975,
      "grad_norm": 3.8762905597686768,
      "learning_rate": 6.658336252189142e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 76836224,
      "step": 4775
    },
    {
      "epoch": 0.33455042160287896,
      "grad_norm": 4.117221832275391,
      "learning_rate": 6.657636427320491e-05,
      "loss": 1.2631,
      "num_input_tokens_seen": 76852608,
      "step": 4776
    },
    {
      "epoch": 0.3346204698486082,
      "grad_norm": 3.814997434616089,
      "learning_rate": 6.656936602451839e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 76868400,
      "step": 4777
    },
    {
      "epoch": 0.3346905180943375,
      "grad_norm": 3.6070499420166016,
      "learning_rate": 6.656236777583187e-05,
      "loss": 0.8537,
      "num_input_tokens_seen": 76884784,
      "step": 4778
    },
    {
      "epoch": 0.33476056634006673,
      "grad_norm": 6.291281700134277,
      "learning_rate": 6.655536952714536e-05,
      "loss": 1.2195,
      "num_input_tokens_seen": 76901168,
      "step": 4779
    },
    {
      "epoch": 0.334830614585796,
      "grad_norm": 7.043301105499268,
      "learning_rate": 6.654837127845885e-05,
      "loss": 1.1015,
      "num_input_tokens_seen": 76917552,
      "step": 4780
    },
    {
      "epoch": 0.3349006628315252,
      "grad_norm": 3.6702778339385986,
      "learning_rate": 6.654137302977234e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 76933936,
      "step": 4781
    },
    {
      "epoch": 0.33497071107725446,
      "grad_norm": 4.228512287139893,
      "learning_rate": 6.653437478108583e-05,
      "loss": 0.8781,
      "num_input_tokens_seen": 76950320,
      "step": 4782
    },
    {
      "epoch": 0.3350407593229837,
      "grad_norm": 3.9304075241088867,
      "learning_rate": 6.65273765323993e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 76966704,
      "step": 4783
    },
    {
      "epoch": 0.335110807568713,
      "grad_norm": 3.608708620071411,
      "learning_rate": 6.652037828371278e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 76983016,
      "step": 4784
    },
    {
      "epoch": 0.3351808558144422,
      "grad_norm": 4.402626037597656,
      "learning_rate": 6.651338003502627e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 76999344,
      "step": 4785
    },
    {
      "epoch": 0.33525090406017144,
      "grad_norm": 4.109679222106934,
      "learning_rate": 6.650638178633976e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 77014960,
      "step": 4786
    },
    {
      "epoch": 0.3353209523059007,
      "grad_norm": 3.893702507019043,
      "learning_rate": 6.649938353765324e-05,
      "loss": 1.1101,
      "num_input_tokens_seen": 77031344,
      "step": 4787
    },
    {
      "epoch": 0.33539100055162996,
      "grad_norm": 4.326907157897949,
      "learning_rate": 6.649238528896673e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 77047264,
      "step": 4788
    },
    {
      "epoch": 0.33546104879735916,
      "grad_norm": 4.946060657501221,
      "learning_rate": 6.648538704028022e-05,
      "loss": 1.0413,
      "num_input_tokens_seen": 77063648,
      "step": 4789
    },
    {
      "epoch": 0.3355310970430884,
      "grad_norm": 3.5379018783569336,
      "learning_rate": 6.64783887915937e-05,
      "loss": 1.0981,
      "num_input_tokens_seen": 77080032,
      "step": 4790
    },
    {
      "epoch": 0.3356011452888177,
      "grad_norm": 4.117929935455322,
      "learning_rate": 6.647139054290717e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 77096416,
      "step": 4791
    },
    {
      "epoch": 0.33567119353454694,
      "grad_norm": 4.293130397796631,
      "learning_rate": 6.646439229422067e-05,
      "loss": 1.1938,
      "num_input_tokens_seen": 77112800,
      "step": 4792
    },
    {
      "epoch": 0.33574124178027615,
      "grad_norm": 3.8246893882751465,
      "learning_rate": 6.645739404553416e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 77128272,
      "step": 4793
    },
    {
      "epoch": 0.3358112900260054,
      "grad_norm": 4.095324993133545,
      "learning_rate": 6.645039579684764e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 77144008,
      "step": 4794
    },
    {
      "epoch": 0.33588133827173466,
      "grad_norm": 3.7015397548675537,
      "learning_rate": 6.644339754816113e-05,
      "loss": 1.1318,
      "num_input_tokens_seen": 77160392,
      "step": 4795
    },
    {
      "epoch": 0.3359513865174639,
      "grad_norm": 3.7702248096466064,
      "learning_rate": 6.643639929947461e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 77176776,
      "step": 4796
    },
    {
      "epoch": 0.3360214347631931,
      "grad_norm": 6.468194961547852,
      "learning_rate": 6.642940105078809e-05,
      "loss": 1.1144,
      "num_input_tokens_seen": 77192000,
      "step": 4797
    },
    {
      "epoch": 0.3360914830089224,
      "grad_norm": 5.211976528167725,
      "learning_rate": 6.642240280210158e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 77208040,
      "step": 4798
    },
    {
      "epoch": 0.33616153125465165,
      "grad_norm": 4.3227763175964355,
      "learning_rate": 6.641540455341507e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 77224016,
      "step": 4799
    },
    {
      "epoch": 0.3362315795003809,
      "grad_norm": 3.6128039360046387,
      "learning_rate": 6.640840630472856e-05,
      "loss": 1.0222,
      "num_input_tokens_seen": 77240400,
      "step": 4800
    },
    {
      "epoch": 0.3362315795003809,
      "eval_loss": 1.1314613819122314,
      "eval_runtime": 0.2053,
      "eval_samples_per_second": 4.87,
      "eval_steps_per_second": 4.87,
      "num_input_tokens_seen": 77240400,
      "step": 4800
    },
    {
      "epoch": 0.3363016277461101,
      "grad_norm": 3.851407766342163,
      "learning_rate": 6.640140805604203e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 77255896,
      "step": 4801
    },
    {
      "epoch": 0.33637167599183937,
      "grad_norm": 3.8287763595581055,
      "learning_rate": 6.639440980735552e-05,
      "loss": 1.02,
      "num_input_tokens_seen": 77272280,
      "step": 4802
    },
    {
      "epoch": 0.33644172423756863,
      "grad_norm": 4.373470306396484,
      "learning_rate": 6.638741155866901e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 77288664,
      "step": 4803
    },
    {
      "epoch": 0.3365117724832979,
      "grad_norm": 5.384084224700928,
      "learning_rate": 6.638041330998248e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 77304544,
      "step": 4804
    },
    {
      "epoch": 0.3365818207290271,
      "grad_norm": 4.742502212524414,
      "learning_rate": 6.637341506129597e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 77320928,
      "step": 4805
    },
    {
      "epoch": 0.33665186897475635,
      "grad_norm": 3.5294950008392334,
      "learning_rate": 6.636641681260946e-05,
      "loss": 0.827,
      "num_input_tokens_seen": 77337072,
      "step": 4806
    },
    {
      "epoch": 0.3367219172204856,
      "grad_norm": 4.925806999206543,
      "learning_rate": 6.635941856392295e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 77352312,
      "step": 4807
    },
    {
      "epoch": 0.33679196546621487,
      "grad_norm": 4.373791694641113,
      "learning_rate": 6.635242031523644e-05,
      "loss": 1.055,
      "num_input_tokens_seen": 77368696,
      "step": 4808
    },
    {
      "epoch": 0.3368620137119441,
      "grad_norm": 4.921911239624023,
      "learning_rate": 6.634542206654993e-05,
      "loss": 1.0054,
      "num_input_tokens_seen": 77384976,
      "step": 4809
    },
    {
      "epoch": 0.33693206195767333,
      "grad_norm": 6.203757286071777,
      "learning_rate": 6.63384238178634e-05,
      "loss": 1.1801,
      "num_input_tokens_seen": 77401360,
      "step": 4810
    },
    {
      "epoch": 0.3370021102034026,
      "grad_norm": 3.675086498260498,
      "learning_rate": 6.633142556917688e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 77417744,
      "step": 4811
    },
    {
      "epoch": 0.33707215844913185,
      "grad_norm": 5.169121742248535,
      "learning_rate": 6.632442732049036e-05,
      "loss": 0.9541,
      "num_input_tokens_seen": 77434128,
      "step": 4812
    },
    {
      "epoch": 0.3371422066948611,
      "grad_norm": 4.036499977111816,
      "learning_rate": 6.631742907180387e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 77449856,
      "step": 4813
    },
    {
      "epoch": 0.3372122549405903,
      "grad_norm": 4.040637016296387,
      "learning_rate": 6.631043082311734e-05,
      "loss": 1.0704,
      "num_input_tokens_seen": 77465536,
      "step": 4814
    },
    {
      "epoch": 0.3372823031863196,
      "grad_norm": 5.76871395111084,
      "learning_rate": 6.630343257443083e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 77481920,
      "step": 4815
    },
    {
      "epoch": 0.33735235143204884,
      "grad_norm": 5.222348690032959,
      "learning_rate": 6.629643432574432e-05,
      "loss": 0.9411,
      "num_input_tokens_seen": 77497464,
      "step": 4816
    },
    {
      "epoch": 0.3374223996777781,
      "grad_norm": 4.099587440490723,
      "learning_rate": 6.62894360770578e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 77513848,
      "step": 4817
    },
    {
      "epoch": 0.3374924479235073,
      "grad_norm": 4.034639835357666,
      "learning_rate": 6.628243782837127e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 77530000,
      "step": 4818
    },
    {
      "epoch": 0.33756249616923656,
      "grad_norm": 5.280242443084717,
      "learning_rate": 6.627543957968477e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 77546384,
      "step": 4819
    },
    {
      "epoch": 0.3376325444149658,
      "grad_norm": 4.851918697357178,
      "learning_rate": 6.626844133099826e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 77561000,
      "step": 4820
    },
    {
      "epoch": 0.3377025926606951,
      "grad_norm": 3.957601308822632,
      "learning_rate": 6.626144308231173e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 77576512,
      "step": 4821
    },
    {
      "epoch": 0.3377726409064243,
      "grad_norm": 3.9369540214538574,
      "learning_rate": 6.625444483362522e-05,
      "loss": 1.0611,
      "num_input_tokens_seen": 77592896,
      "step": 4822
    },
    {
      "epoch": 0.33784268915215354,
      "grad_norm": 4.397778511047363,
      "learning_rate": 6.624744658493871e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 77609280,
      "step": 4823
    },
    {
      "epoch": 0.3379127373978828,
      "grad_norm": 6.919220447540283,
      "learning_rate": 6.624044833625219e-05,
      "loss": 0.9992,
      "num_input_tokens_seen": 77625664,
      "step": 4824
    },
    {
      "epoch": 0.33798278564361206,
      "grad_norm": 5.501309871673584,
      "learning_rate": 6.623345008756568e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 77642048,
      "step": 4825
    },
    {
      "epoch": 0.33805283388934126,
      "grad_norm": 4.240433216094971,
      "learning_rate": 6.622645183887916e-05,
      "loss": 0.8771,
      "num_input_tokens_seen": 77658432,
      "step": 4826
    },
    {
      "epoch": 0.3381228821350705,
      "grad_norm": 3.9563584327697754,
      "learning_rate": 6.621945359019265e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 77674816,
      "step": 4827
    },
    {
      "epoch": 0.3381929303807998,
      "grad_norm": 6.861433982849121,
      "learning_rate": 6.621245534150613e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 77689712,
      "step": 4828
    },
    {
      "epoch": 0.33826297862652904,
      "grad_norm": 3.951972246170044,
      "learning_rate": 6.620545709281962e-05,
      "loss": 1.2363,
      "num_input_tokens_seen": 77706096,
      "step": 4829
    },
    {
      "epoch": 0.33833302687225825,
      "grad_norm": 4.419849395751953,
      "learning_rate": 6.61984588441331e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 77721168,
      "step": 4830
    },
    {
      "epoch": 0.3384030751179875,
      "grad_norm": 5.075031280517578,
      "learning_rate": 6.619146059544658e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 77737552,
      "step": 4831
    },
    {
      "epoch": 0.33847312336371677,
      "grad_norm": 4.216047763824463,
      "learning_rate": 6.618446234676007e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 77753936,
      "step": 4832
    },
    {
      "epoch": 0.338543171609446,
      "grad_norm": 4.175511360168457,
      "learning_rate": 6.617746409807357e-05,
      "loss": 1.1294,
      "num_input_tokens_seen": 77769848,
      "step": 4833
    },
    {
      "epoch": 0.33861321985517523,
      "grad_norm": 3.894831418991089,
      "learning_rate": 6.617046584938705e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 77785792,
      "step": 4834
    },
    {
      "epoch": 0.3386832681009045,
      "grad_norm": 4.06626033782959,
      "learning_rate": 6.616346760070053e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 77801728,
      "step": 4835
    },
    {
      "epoch": 0.33875331634663375,
      "grad_norm": 6.5680341720581055,
      "learning_rate": 6.615646935201402e-05,
      "loss": 1.2627,
      "num_input_tokens_seen": 77818112,
      "step": 4836
    },
    {
      "epoch": 0.338823364592363,
      "grad_norm": 4.967332363128662,
      "learning_rate": 6.61494711033275e-05,
      "loss": 1.1455,
      "num_input_tokens_seen": 77833464,
      "step": 4837
    },
    {
      "epoch": 0.3388934128380922,
      "grad_norm": 4.244988918304443,
      "learning_rate": 6.614247285464097e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 77849848,
      "step": 4838
    },
    {
      "epoch": 0.33896346108382147,
      "grad_norm": 4.361011028289795,
      "learning_rate": 6.613547460595447e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 77866232,
      "step": 4839
    },
    {
      "epoch": 0.33903350932955073,
      "grad_norm": 6.348353385925293,
      "learning_rate": 6.612847635726796e-05,
      "loss": 1.1202,
      "num_input_tokens_seen": 77882616,
      "step": 4840
    },
    {
      "epoch": 0.33910355757528,
      "grad_norm": 3.8235714435577393,
      "learning_rate": 6.612147810858144e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 77899000,
      "step": 4841
    },
    {
      "epoch": 0.3391736058210092,
      "grad_norm": 4.069207191467285,
      "learning_rate": 6.611447985989493e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 77915384,
      "step": 4842
    },
    {
      "epoch": 0.33924365406673845,
      "grad_norm": 3.8036270141601562,
      "learning_rate": 6.610748161120842e-05,
      "loss": 1.1935,
      "num_input_tokens_seen": 77931768,
      "step": 4843
    },
    {
      "epoch": 0.3393137023124677,
      "grad_norm": 5.776700019836426,
      "learning_rate": 6.610048336252189e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 77948152,
      "step": 4844
    },
    {
      "epoch": 0.339383750558197,
      "grad_norm": 5.484714508056641,
      "learning_rate": 6.609348511383538e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 77964536,
      "step": 4845
    },
    {
      "epoch": 0.3394537988039262,
      "grad_norm": 4.595640659332275,
      "learning_rate": 6.608648686514887e-05,
      "loss": 1.0854,
      "num_input_tokens_seen": 77980648,
      "step": 4846
    },
    {
      "epoch": 0.33952384704965544,
      "grad_norm": 5.0377197265625,
      "learning_rate": 6.607948861646236e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 77995624,
      "step": 4847
    },
    {
      "epoch": 0.3395938952953847,
      "grad_norm": 3.796713352203369,
      "learning_rate": 6.607249036777583e-05,
      "loss": 0.8928,
      "num_input_tokens_seen": 78012008,
      "step": 4848
    },
    {
      "epoch": 0.33966394354111396,
      "grad_norm": 5.138030052185059,
      "learning_rate": 6.606549211908932e-05,
      "loss": 0.9565,
      "num_input_tokens_seen": 78027960,
      "step": 4849
    },
    {
      "epoch": 0.3397339917868432,
      "grad_norm": 5.852467060089111,
      "learning_rate": 6.605849387040281e-05,
      "loss": 1.1297,
      "num_input_tokens_seen": 78044144,
      "step": 4850
    },
    {
      "epoch": 0.3398040400325724,
      "grad_norm": 3.5677835941314697,
      "learning_rate": 6.605149562171628e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 78059096,
      "step": 4851
    },
    {
      "epoch": 0.3398740882783017,
      "grad_norm": 4.033452987670898,
      "learning_rate": 6.604449737302977e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 78075480,
      "step": 4852
    },
    {
      "epoch": 0.33994413652403094,
      "grad_norm": 5.06736421585083,
      "learning_rate": 6.603749912434327e-05,
      "loss": 1.0456,
      "num_input_tokens_seen": 78091864,
      "step": 4853
    },
    {
      "epoch": 0.3400141847697602,
      "grad_norm": 4.851357460021973,
      "learning_rate": 6.603050087565675e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 78108248,
      "step": 4854
    },
    {
      "epoch": 0.3400842330154894,
      "grad_norm": 4.659403324127197,
      "learning_rate": 6.602350262697022e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 78123720,
      "step": 4855
    },
    {
      "epoch": 0.34015428126121866,
      "grad_norm": 4.0248870849609375,
      "learning_rate": 6.601650437828371e-05,
      "loss": 1.0192,
      "num_input_tokens_seen": 78140104,
      "step": 4856
    },
    {
      "epoch": 0.3402243295069479,
      "grad_norm": 3.615807056427002,
      "learning_rate": 6.60095061295972e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 78155768,
      "step": 4857
    },
    {
      "epoch": 0.3402943777526772,
      "grad_norm": 4.032024383544922,
      "learning_rate": 6.600250788091068e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 78171688,
      "step": 4858
    },
    {
      "epoch": 0.3403644259984064,
      "grad_norm": 4.213406085968018,
      "learning_rate": 6.599550963222418e-05,
      "loss": 1.1186,
      "num_input_tokens_seen": 78188072,
      "step": 4859
    },
    {
      "epoch": 0.34043447424413564,
      "grad_norm": 4.343294620513916,
      "learning_rate": 6.598851138353767e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 78203328,
      "step": 4860
    },
    {
      "epoch": 0.3405045224898649,
      "grad_norm": 5.112723350524902,
      "learning_rate": 6.598151313485114e-05,
      "loss": 0.8201,
      "num_input_tokens_seen": 78218480,
      "step": 4861
    },
    {
      "epoch": 0.34057457073559416,
      "grad_norm": 5.0619215965271,
      "learning_rate": 6.597451488616463e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 78234864,
      "step": 4862
    },
    {
      "epoch": 0.34064461898132337,
      "grad_norm": 4.161584377288818,
      "learning_rate": 6.596751663747812e-05,
      "loss": 1.3292,
      "num_input_tokens_seen": 78250664,
      "step": 4863
    },
    {
      "epoch": 0.3407146672270526,
      "grad_norm": 4.402634143829346,
      "learning_rate": 6.59605183887916e-05,
      "loss": 1.2664,
      "num_input_tokens_seen": 78266976,
      "step": 4864
    },
    {
      "epoch": 0.3407847154727819,
      "grad_norm": 4.01839017868042,
      "learning_rate": 6.595352014010508e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 78283360,
      "step": 4865
    },
    {
      "epoch": 0.34085476371851114,
      "grad_norm": 3.6157965660095215,
      "learning_rate": 6.594652189141857e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 78299744,
      "step": 4866
    },
    {
      "epoch": 0.34092481196424035,
      "grad_norm": 4.221523761749268,
      "learning_rate": 6.593952364273206e-05,
      "loss": 1.3689,
      "num_input_tokens_seen": 78314944,
      "step": 4867
    },
    {
      "epoch": 0.3409948602099696,
      "grad_norm": 5.253129482269287,
      "learning_rate": 6.593252539404554e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 78331168,
      "step": 4868
    },
    {
      "epoch": 0.34106490845569887,
      "grad_norm": 4.839991569519043,
      "learning_rate": 6.592552714535902e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 78347200,
      "step": 4869
    },
    {
      "epoch": 0.3411349567014281,
      "grad_norm": 5.994297504425049,
      "learning_rate": 6.591852889667251e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 78363584,
      "step": 4870
    },
    {
      "epoch": 0.34120500494715733,
      "grad_norm": 4.886160850524902,
      "learning_rate": 6.591153064798599e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 78379968,
      "step": 4871
    },
    {
      "epoch": 0.3412750531928866,
      "grad_norm": 8.658349990844727,
      "learning_rate": 6.590453239929948e-05,
      "loss": 1.0145,
      "num_input_tokens_seen": 78395368,
      "step": 4872
    },
    {
      "epoch": 0.34134510143861585,
      "grad_norm": 5.1440935134887695,
      "learning_rate": 6.589753415061298e-05,
      "loss": 0.9584,
      "num_input_tokens_seen": 78411752,
      "step": 4873
    },
    {
      "epoch": 0.3414151496843451,
      "grad_norm": 4.83282995223999,
      "learning_rate": 6.589053590192645e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 78428128,
      "step": 4874
    },
    {
      "epoch": 0.3414851979300743,
      "grad_norm": 3.603290557861328,
      "learning_rate": 6.588353765323993e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 78444512,
      "step": 4875
    },
    {
      "epoch": 0.3415552461758036,
      "grad_norm": 3.8035361766815186,
      "learning_rate": 6.587653940455342e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 78460896,
      "step": 4876
    },
    {
      "epoch": 0.34162529442153283,
      "grad_norm": 4.02992582321167,
      "learning_rate": 6.58695411558669e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 78477280,
      "step": 4877
    },
    {
      "epoch": 0.3416953426672621,
      "grad_norm": 4.898126125335693,
      "learning_rate": 6.586254290718038e-05,
      "loss": 1.0632,
      "num_input_tokens_seen": 78493664,
      "step": 4878
    },
    {
      "epoch": 0.3417653909129913,
      "grad_norm": 4.779463768005371,
      "learning_rate": 6.585554465849388e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 78509280,
      "step": 4879
    },
    {
      "epoch": 0.34183543915872056,
      "grad_norm": 3.7280569076538086,
      "learning_rate": 6.584854640980737e-05,
      "loss": 1.1184,
      "num_input_tokens_seen": 78525664,
      "step": 4880
    },
    {
      "epoch": 0.3419054874044498,
      "grad_norm": 4.691235542297363,
      "learning_rate": 6.584154816112085e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 78542048,
      "step": 4881
    },
    {
      "epoch": 0.3419755356501791,
      "grad_norm": 4.188792705535889,
      "learning_rate": 6.583454991243432e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 78558432,
      "step": 4882
    },
    {
      "epoch": 0.34204558389590833,
      "grad_norm": 3.7049522399902344,
      "learning_rate": 6.582755166374781e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 78574816,
      "step": 4883
    },
    {
      "epoch": 0.34211563214163754,
      "grad_norm": 5.808310508728027,
      "learning_rate": 6.58205534150613e-05,
      "loss": 1.0674,
      "num_input_tokens_seen": 78590992,
      "step": 4884
    },
    {
      "epoch": 0.3421856803873668,
      "grad_norm": 3.877638339996338,
      "learning_rate": 6.581355516637479e-05,
      "loss": 0.9908,
      "num_input_tokens_seen": 78607368,
      "step": 4885
    },
    {
      "epoch": 0.34225572863309606,
      "grad_norm": 3.7855000495910645,
      "learning_rate": 6.580655691768828e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 78622712,
      "step": 4886
    },
    {
      "epoch": 0.3423257768788253,
      "grad_norm": 3.9921584129333496,
      "learning_rate": 6.579955866900176e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 78638840,
      "step": 4887
    },
    {
      "epoch": 0.3423958251245545,
      "grad_norm": 4.037683486938477,
      "learning_rate": 6.579256042031524e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 78655224,
      "step": 4888
    },
    {
      "epoch": 0.3424658733702838,
      "grad_norm": 4.109930515289307,
      "learning_rate": 6.578556217162873e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 78670984,
      "step": 4889
    },
    {
      "epoch": 0.34253592161601304,
      "grad_norm": 5.201082229614258,
      "learning_rate": 6.577856392294222e-05,
      "loss": 1.0844,
      "num_input_tokens_seen": 78686856,
      "step": 4890
    },
    {
      "epoch": 0.3426059698617423,
      "grad_norm": 5.922754764556885,
      "learning_rate": 6.577156567425569e-05,
      "loss": 1.2428,
      "num_input_tokens_seen": 78702688,
      "step": 4891
    },
    {
      "epoch": 0.3426760181074715,
      "grad_norm": 4.052786350250244,
      "learning_rate": 6.576456742556918e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 78719072,
      "step": 4892
    },
    {
      "epoch": 0.34274606635320076,
      "grad_norm": 4.0263671875,
      "learning_rate": 6.575756917688267e-05,
      "loss": 1.2076,
      "num_input_tokens_seen": 78735456,
      "step": 4893
    },
    {
      "epoch": 0.34281611459893,
      "grad_norm": 3.773024082183838,
      "learning_rate": 6.575057092819616e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 78751424,
      "step": 4894
    },
    {
      "epoch": 0.3428861628446593,
      "grad_norm": 3.770413398742676,
      "learning_rate": 6.574357267950963e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 78767808,
      "step": 4895
    },
    {
      "epoch": 0.3429562110903885,
      "grad_norm": 6.26648473739624,
      "learning_rate": 6.573657443082312e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 78783448,
      "step": 4896
    },
    {
      "epoch": 0.34302625933611774,
      "grad_norm": 4.071943283081055,
      "learning_rate": 6.572957618213661e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 78798976,
      "step": 4897
    },
    {
      "epoch": 0.343096307581847,
      "grad_norm": 7.654726505279541,
      "learning_rate": 6.572257793345008e-05,
      "loss": 1.0902,
      "num_input_tokens_seen": 78814664,
      "step": 4898
    },
    {
      "epoch": 0.34316635582757626,
      "grad_norm": 5.928562641143799,
      "learning_rate": 6.571557968476357e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 78830792,
      "step": 4899
    },
    {
      "epoch": 0.34323640407330547,
      "grad_norm": 3.8699424266815186,
      "learning_rate": 6.570858143607708e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 78847176,
      "step": 4900
    },
    {
      "epoch": 0.3433064523190347,
      "grad_norm": 4.177735328674316,
      "learning_rate": 6.570158318739055e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 78863496,
      "step": 4901
    },
    {
      "epoch": 0.343376500564764,
      "grad_norm": 3.5755650997161865,
      "learning_rate": 6.569458493870403e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 78879688,
      "step": 4902
    },
    {
      "epoch": 0.34344654881049325,
      "grad_norm": 5.799609184265137,
      "learning_rate": 6.568758669001751e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 78894744,
      "step": 4903
    },
    {
      "epoch": 0.34351659705622245,
      "grad_norm": 3.7705209255218506,
      "learning_rate": 6.5680588441331e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 78911112,
      "step": 4904
    },
    {
      "epoch": 0.3435866453019517,
      "grad_norm": 4.713012218475342,
      "learning_rate": 6.567359019264449e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 78927496,
      "step": 4905
    },
    {
      "epoch": 0.34365669354768097,
      "grad_norm": 3.8360157012939453,
      "learning_rate": 6.566659194395798e-05,
      "loss": 1.1281,
      "num_input_tokens_seen": 78942712,
      "step": 4906
    },
    {
      "epoch": 0.34372674179341023,
      "grad_norm": 3.6071383953094482,
      "learning_rate": 6.565959369527147e-05,
      "loss": 0.974,
      "num_input_tokens_seen": 78959016,
      "step": 4907
    },
    {
      "epoch": 0.34379679003913943,
      "grad_norm": 4.876083850860596,
      "learning_rate": 6.565259544658494e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 78975400,
      "step": 4908
    },
    {
      "epoch": 0.3438668382848687,
      "grad_norm": 4.011876583099365,
      "learning_rate": 6.564559719789842e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 78991784,
      "step": 4909
    },
    {
      "epoch": 0.34393688653059795,
      "grad_norm": 3.74336576461792,
      "learning_rate": 6.563859894921191e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 79008168,
      "step": 4910
    },
    {
      "epoch": 0.3440069347763272,
      "grad_norm": 4.092207908630371,
      "learning_rate": 6.56316007005254e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 79024200,
      "step": 4911
    },
    {
      "epoch": 0.3440769830220564,
      "grad_norm": 3.771979331970215,
      "learning_rate": 6.562460245183888e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 79040584,
      "step": 4912
    },
    {
      "epoch": 0.3441470312677857,
      "grad_norm": 4.791725158691406,
      "learning_rate": 6.561760420315237e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 79056552,
      "step": 4913
    },
    {
      "epoch": 0.34421707951351493,
      "grad_norm": 4.9150519371032715,
      "learning_rate": 6.561060595446586e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 79072536,
      "step": 4914
    },
    {
      "epoch": 0.3442871277592442,
      "grad_norm": 4.550070285797119,
      "learning_rate": 6.560360770577934e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 79088376,
      "step": 4915
    },
    {
      "epoch": 0.34435717600497345,
      "grad_norm": 4.497488975524902,
      "learning_rate": 6.559660945709283e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 79104760,
      "step": 4916
    },
    {
      "epoch": 0.34442722425070266,
      "grad_norm": 4.417470455169678,
      "learning_rate": 6.558961120840631e-05,
      "loss": 1.0471,
      "num_input_tokens_seen": 79120296,
      "step": 4917
    },
    {
      "epoch": 0.3444972724964319,
      "grad_norm": 4.967655658721924,
      "learning_rate": 6.558261295971979e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 79135936,
      "step": 4918
    },
    {
      "epoch": 0.3445673207421612,
      "grad_norm": 4.973440647125244,
      "learning_rate": 6.557561471103328e-05,
      "loss": 1.1045,
      "num_input_tokens_seen": 79151632,
      "step": 4919
    },
    {
      "epoch": 0.34463736898789044,
      "grad_norm": 5.2282609939575195,
      "learning_rate": 6.556861646234677e-05,
      "loss": 1.2508,
      "num_input_tokens_seen": 79167112,
      "step": 4920
    },
    {
      "epoch": 0.34470741723361964,
      "grad_norm": 4.118466854095459,
      "learning_rate": 6.556161821366025e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 79183496,
      "step": 4921
    },
    {
      "epoch": 0.3447774654793489,
      "grad_norm": 4.74249267578125,
      "learning_rate": 6.555461996497373e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 79199240,
      "step": 4922
    },
    {
      "epoch": 0.34484751372507816,
      "grad_norm": 4.21056604385376,
      "learning_rate": 6.554762171628722e-05,
      "loss": 1.2693,
      "num_input_tokens_seen": 79215560,
      "step": 4923
    },
    {
      "epoch": 0.3449175619708074,
      "grad_norm": 3.584332227706909,
      "learning_rate": 6.55406234676007e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 79231944,
      "step": 4924
    },
    {
      "epoch": 0.3449876102165366,
      "grad_norm": 7.287233829498291,
      "learning_rate": 6.55336252189142e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 79248328,
      "step": 4925
    },
    {
      "epoch": 0.3450576584622659,
      "grad_norm": 6.5669379234313965,
      "learning_rate": 6.552662697022767e-05,
      "loss": 0.8538,
      "num_input_tokens_seen": 79264712,
      "step": 4926
    },
    {
      "epoch": 0.34512770670799514,
      "grad_norm": 4.086475849151611,
      "learning_rate": 6.551962872154117e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 79281096,
      "step": 4927
    },
    {
      "epoch": 0.3451977549537244,
      "grad_norm": 5.543658256530762,
      "learning_rate": 6.551263047285465e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 79297120,
      "step": 4928
    },
    {
      "epoch": 0.3452678031994536,
      "grad_norm": 6.474762439727783,
      "learning_rate": 6.550563222416812e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 79313504,
      "step": 4929
    },
    {
      "epoch": 0.34533785144518286,
      "grad_norm": 3.8226888179779053,
      "learning_rate": 6.549863397548161e-05,
      "loss": 0.9796,
      "num_input_tokens_seen": 79329888,
      "step": 4930
    },
    {
      "epoch": 0.3454078996909121,
      "grad_norm": 3.8926212787628174,
      "learning_rate": 6.54916357267951e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 79346272,
      "step": 4931
    },
    {
      "epoch": 0.3454779479366414,
      "grad_norm": 4.127487659454346,
      "learning_rate": 6.548463747810859e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 79362656,
      "step": 4932
    },
    {
      "epoch": 0.3455479961823706,
      "grad_norm": 6.770711421966553,
      "learning_rate": 6.547763922942208e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 79378544,
      "step": 4933
    },
    {
      "epoch": 0.34561804442809985,
      "grad_norm": 5.547317028045654,
      "learning_rate": 6.547064098073557e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 79394896,
      "step": 4934
    },
    {
      "epoch": 0.3456880926738291,
      "grad_norm": 4.469418048858643,
      "learning_rate": 6.546364273204904e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 79410480,
      "step": 4935
    },
    {
      "epoch": 0.34575814091955837,
      "grad_norm": 4.901472091674805,
      "learning_rate": 6.545664448336252e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 79426864,
      "step": 4936
    },
    {
      "epoch": 0.34582818916528757,
      "grad_norm": 3.60495662689209,
      "learning_rate": 6.5449646234676e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 79443248,
      "step": 4937
    },
    {
      "epoch": 0.34589823741101683,
      "grad_norm": 4.513663291931152,
      "learning_rate": 6.544264798598949e-05,
      "loss": 1.2239,
      "num_input_tokens_seen": 79459632,
      "step": 4938
    },
    {
      "epoch": 0.3459682856567461,
      "grad_norm": 3.6959240436553955,
      "learning_rate": 6.543564973730298e-05,
      "loss": 1.0561,
      "num_input_tokens_seen": 79475320,
      "step": 4939
    },
    {
      "epoch": 0.34603833390247535,
      "grad_norm": 4.071475505828857,
      "learning_rate": 6.542865148861647e-05,
      "loss": 1.1963,
      "num_input_tokens_seen": 79491704,
      "step": 4940
    },
    {
      "epoch": 0.34610838214820455,
      "grad_norm": 3.665421962738037,
      "learning_rate": 6.542165323992996e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 79508088,
      "step": 4941
    },
    {
      "epoch": 0.3461784303939338,
      "grad_norm": 4.1782941818237305,
      "learning_rate": 6.541465499124343e-05,
      "loss": 0.85,
      "num_input_tokens_seen": 79523936,
      "step": 4942
    },
    {
      "epoch": 0.34624847863966307,
      "grad_norm": 4.728964328765869,
      "learning_rate": 6.540765674255691e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 79539848,
      "step": 4943
    },
    {
      "epoch": 0.34631852688539233,
      "grad_norm": 5.39119815826416,
      "learning_rate": 6.540065849387041e-05,
      "loss": 1.185,
      "num_input_tokens_seen": 79555040,
      "step": 4944
    },
    {
      "epoch": 0.34638857513112153,
      "grad_norm": 3.8394956588745117,
      "learning_rate": 6.53936602451839e-05,
      "loss": 0.7774,
      "num_input_tokens_seen": 79570504,
      "step": 4945
    },
    {
      "epoch": 0.3464586233768508,
      "grad_norm": 5.03010368347168,
      "learning_rate": 6.538666199649737e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 79586888,
      "step": 4946
    },
    {
      "epoch": 0.34652867162258005,
      "grad_norm": 3.984548807144165,
      "learning_rate": 6.537966374781086e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 79603128,
      "step": 4947
    },
    {
      "epoch": 0.3465987198683093,
      "grad_norm": 5.096433162689209,
      "learning_rate": 6.537266549912435e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 79618624,
      "step": 4948
    },
    {
      "epoch": 0.3466687681140385,
      "grad_norm": 3.6773791313171387,
      "learning_rate": 6.536566725043783e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 79635008,
      "step": 4949
    },
    {
      "epoch": 0.3467388163597678,
      "grad_norm": 4.050341606140137,
      "learning_rate": 6.535866900175132e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 79651392,
      "step": 4950
    },
    {
      "epoch": 0.34680886460549704,
      "grad_norm": 3.8354263305664062,
      "learning_rate": 6.53516707530648e-05,
      "loss": 1.0264,
      "num_input_tokens_seen": 79667040,
      "step": 4951
    },
    {
      "epoch": 0.3468789128512263,
      "grad_norm": 4.2188873291015625,
      "learning_rate": 6.534467250437829e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 79683152,
      "step": 4952
    },
    {
      "epoch": 0.34694896109695555,
      "grad_norm": 4.75797700881958,
      "learning_rate": 6.533767425569177e-05,
      "loss": 1.2475,
      "num_input_tokens_seen": 79699536,
      "step": 4953
    },
    {
      "epoch": 0.34701900934268476,
      "grad_norm": 3.494459867477417,
      "learning_rate": 6.533067600700527e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 79715920,
      "step": 4954
    },
    {
      "epoch": 0.347089057588414,
      "grad_norm": 3.860872268676758,
      "learning_rate": 6.532367775831874e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 79731832,
      "step": 4955
    },
    {
      "epoch": 0.3471591058341433,
      "grad_norm": 4.188973426818848,
      "learning_rate": 6.531667950963222e-05,
      "loss": 1.0814,
      "num_input_tokens_seen": 79747592,
      "step": 4956
    },
    {
      "epoch": 0.34722915407987254,
      "grad_norm": 5.598564624786377,
      "learning_rate": 6.530968126094571e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 79763048,
      "step": 4957
    },
    {
      "epoch": 0.34729920232560174,
      "grad_norm": 4.153980255126953,
      "learning_rate": 6.53026830122592e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 79777928,
      "step": 4958
    },
    {
      "epoch": 0.347369250571331,
      "grad_norm": 3.875469446182251,
      "learning_rate": 6.529568476357269e-05,
      "loss": 1.1449,
      "num_input_tokens_seen": 79794312,
      "step": 4959
    },
    {
      "epoch": 0.34743929881706026,
      "grad_norm": 5.391599655151367,
      "learning_rate": 6.528868651488617e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 79810696,
      "step": 4960
    },
    {
      "epoch": 0.3475093470627895,
      "grad_norm": 3.3462777137756348,
      "learning_rate": 6.528168826619966e-05,
      "loss": 0.8645,
      "num_input_tokens_seen": 79826208,
      "step": 4961
    },
    {
      "epoch": 0.3475793953085187,
      "grad_norm": 3.5444939136505127,
      "learning_rate": 6.527469001751314e-05,
      "loss": 1.0989,
      "num_input_tokens_seen": 79842592,
      "step": 4962
    },
    {
      "epoch": 0.347649443554248,
      "grad_norm": 4.541754722595215,
      "learning_rate": 6.526769176882661e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 79858976,
      "step": 4963
    },
    {
      "epoch": 0.34771949179997724,
      "grad_norm": 3.728207588195801,
      "learning_rate": 6.52606935201401e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 79874944,
      "step": 4964
    },
    {
      "epoch": 0.3477895400457065,
      "grad_norm": 5.615260601043701,
      "learning_rate": 6.525369527145359e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 79890968,
      "step": 4965
    },
    {
      "epoch": 0.3478595882914357,
      "grad_norm": 4.863505840301514,
      "learning_rate": 6.524669702276708e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 79907352,
      "step": 4966
    },
    {
      "epoch": 0.34792963653716497,
      "grad_norm": 3.6932058334350586,
      "learning_rate": 6.523969877408057e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 79923736,
      "step": 4967
    },
    {
      "epoch": 0.3479996847828942,
      "grad_norm": 4.483904838562012,
      "learning_rate": 6.523270052539406e-05,
      "loss": 1.222,
      "num_input_tokens_seen": 79939360,
      "step": 4968
    },
    {
      "epoch": 0.3480697330286235,
      "grad_norm": 3.540771007537842,
      "learning_rate": 6.522570227670753e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 79955744,
      "step": 4969
    },
    {
      "epoch": 0.3481397812743527,
      "grad_norm": 3.980483293533325,
      "learning_rate": 6.5218704028021e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 79971368,
      "step": 4970
    },
    {
      "epoch": 0.34820982952008195,
      "grad_norm": 5.302091598510742,
      "learning_rate": 6.521170577933451e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 79986688,
      "step": 4971
    },
    {
      "epoch": 0.3482798777658112,
      "grad_norm": 4.176638603210449,
      "learning_rate": 6.5204707530648e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 80003072,
      "step": 4972
    },
    {
      "epoch": 0.34834992601154047,
      "grad_norm": 5.939540386199951,
      "learning_rate": 6.519770928196147e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 80019344,
      "step": 4973
    },
    {
      "epoch": 0.34841997425726967,
      "grad_norm": 4.681301593780518,
      "learning_rate": 6.519071103327496e-05,
      "loss": 1.1121,
      "num_input_tokens_seen": 80034504,
      "step": 4974
    },
    {
      "epoch": 0.34849002250299893,
      "grad_norm": 4.993075847625732,
      "learning_rate": 6.518371278458845e-05,
      "loss": 0.8792,
      "num_input_tokens_seen": 80050488,
      "step": 4975
    },
    {
      "epoch": 0.3485600707487282,
      "grad_norm": 3.87778377532959,
      "learning_rate": 6.517671453590192e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 80066872,
      "step": 4976
    },
    {
      "epoch": 0.34863011899445745,
      "grad_norm": 3.652738332748413,
      "learning_rate": 6.516971628721541e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 80083232,
      "step": 4977
    },
    {
      "epoch": 0.34870016724018665,
      "grad_norm": 3.9958438873291016,
      "learning_rate": 6.51627180385289e-05,
      "loss": 0.8653,
      "num_input_tokens_seen": 80099616,
      "step": 4978
    },
    {
      "epoch": 0.3487702154859159,
      "grad_norm": 4.190839767456055,
      "learning_rate": 6.515571978984239e-05,
      "loss": 1.2081,
      "num_input_tokens_seen": 80116000,
      "step": 4979
    },
    {
      "epoch": 0.3488402637316452,
      "grad_norm": 4.848324298858643,
      "learning_rate": 6.514872154115586e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 80132384,
      "step": 4980
    },
    {
      "epoch": 0.34891031197737443,
      "grad_norm": 4.863750457763672,
      "learning_rate": 6.514172329246937e-05,
      "loss": 1.1181,
      "num_input_tokens_seen": 80148768,
      "step": 4981
    },
    {
      "epoch": 0.34898036022310364,
      "grad_norm": 4.555769443511963,
      "learning_rate": 6.513472504378284e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 80164984,
      "step": 4982
    },
    {
      "epoch": 0.3490504084688329,
      "grad_norm": 5.041413307189941,
      "learning_rate": 6.512772679509632e-05,
      "loss": 1.0183,
      "num_input_tokens_seen": 80181336,
      "step": 4983
    },
    {
      "epoch": 0.34912045671456216,
      "grad_norm": 4.58367395401001,
      "learning_rate": 6.51207285464098e-05,
      "loss": 1.232,
      "num_input_tokens_seen": 80197720,
      "step": 4984
    },
    {
      "epoch": 0.3491905049602914,
      "grad_norm": 3.9667036533355713,
      "learning_rate": 6.51137302977233e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 80212776,
      "step": 4985
    },
    {
      "epoch": 0.3492605532060207,
      "grad_norm": 3.474071979522705,
      "learning_rate": 6.510673204903678e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 80229160,
      "step": 4986
    },
    {
      "epoch": 0.3493306014517499,
      "grad_norm": 3.912496328353882,
      "learning_rate": 6.509973380035027e-05,
      "loss": 0.9695,
      "num_input_tokens_seen": 80245544,
      "step": 4987
    },
    {
      "epoch": 0.34940064969747914,
      "grad_norm": 3.760340690612793,
      "learning_rate": 6.509273555166376e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 80261400,
      "step": 4988
    },
    {
      "epoch": 0.3494706979432084,
      "grad_norm": 4.982266426086426,
      "learning_rate": 6.508573730297723e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 80277784,
      "step": 4989
    },
    {
      "epoch": 0.34954074618893766,
      "grad_norm": 4.6823530197143555,
      "learning_rate": 6.507873905429071e-05,
      "loss": 1.3118,
      "num_input_tokens_seen": 80294168,
      "step": 4990
    },
    {
      "epoch": 0.34961079443466686,
      "grad_norm": 3.768439769744873,
      "learning_rate": 6.50717408056042e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 80310552,
      "step": 4991
    },
    {
      "epoch": 0.3496808426803961,
      "grad_norm": 3.5285451412200928,
      "learning_rate": 6.50647425569177e-05,
      "loss": 0.8937,
      "num_input_tokens_seen": 80326464,
      "step": 4992
    },
    {
      "epoch": 0.3497508909261254,
      "grad_norm": 3.875992774963379,
      "learning_rate": 6.505774430823118e-05,
      "loss": 0.9514,
      "num_input_tokens_seen": 80342848,
      "step": 4993
    },
    {
      "epoch": 0.34982093917185464,
      "grad_norm": 4.061910152435303,
      "learning_rate": 6.505074605954466e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 80359232,
      "step": 4994
    },
    {
      "epoch": 0.34989098741758384,
      "grad_norm": 4.456427097320557,
      "learning_rate": 6.504374781085815e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 80375616,
      "step": 4995
    },
    {
      "epoch": 0.3499610356633131,
      "grad_norm": 4.381276607513428,
      "learning_rate": 6.503674956217163e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 80392000,
      "step": 4996
    },
    {
      "epoch": 0.35003108390904236,
      "grad_norm": 6.789033889770508,
      "learning_rate": 6.50297513134851e-05,
      "loss": 0.9971,
      "num_input_tokens_seen": 80407360,
      "step": 4997
    },
    {
      "epoch": 0.3501011321547716,
      "grad_norm": 3.953124761581421,
      "learning_rate": 6.50227530647986e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 80423744,
      "step": 4998
    },
    {
      "epoch": 0.3501711804005008,
      "grad_norm": 3.498389482498169,
      "learning_rate": 6.50157548161121e-05,
      "loss": 0.9259,
      "num_input_tokens_seen": 80440128,
      "step": 4999
    },
    {
      "epoch": 0.3502412286462301,
      "grad_norm": 5.498814582824707,
      "learning_rate": 6.500875656742557e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 80456512,
      "step": 5000
    },
    {
      "epoch": 0.3502412286462301,
      "eval_loss": 1.1277527809143066,
      "eval_runtime": 0.1909,
      "eval_samples_per_second": 5.238,
      "eval_steps_per_second": 5.238,
      "num_input_tokens_seen": 80456512,
      "step": 5000
    },
    {
      "epoch": 0.35031127689195934,
      "grad_norm": 3.440230131149292,
      "learning_rate": 6.500175831873906e-05,
      "loss": 0.8354,
      "num_input_tokens_seen": 80472456,
      "step": 5001
    },
    {
      "epoch": 0.3503813251376886,
      "grad_norm": 5.069565296173096,
      "learning_rate": 6.499476007005255e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 80488840,
      "step": 5002
    },
    {
      "epoch": 0.3504513733834178,
      "grad_norm": 4.53994607925415,
      "learning_rate": 6.498776182136602e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 80504984,
      "step": 5003
    },
    {
      "epoch": 0.35052142162914707,
      "grad_norm": 4.136146068572998,
      "learning_rate": 6.498076357267951e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 80520448,
      "step": 5004
    },
    {
      "epoch": 0.3505914698748763,
      "grad_norm": 5.609417915344238,
      "learning_rate": 6.4973765323993e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 80536496,
      "step": 5005
    },
    {
      "epoch": 0.3506615181206056,
      "grad_norm": 4.375439643859863,
      "learning_rate": 6.496676707530649e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 80551592,
      "step": 5006
    },
    {
      "epoch": 0.3507315663663348,
      "grad_norm": 3.5269775390625,
      "learning_rate": 6.495976882661996e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 80567976,
      "step": 5007
    },
    {
      "epoch": 0.35080161461206405,
      "grad_norm": 3.9541778564453125,
      "learning_rate": 6.495277057793346e-05,
      "loss": 1.1451,
      "num_input_tokens_seen": 80584360,
      "step": 5008
    },
    {
      "epoch": 0.3508716628577933,
      "grad_norm": 5.544612407684326,
      "learning_rate": 6.494577232924694e-05,
      "loss": 1.3493,
      "num_input_tokens_seen": 80599856,
      "step": 5009
    },
    {
      "epoch": 0.35094171110352257,
      "grad_norm": 4.189836502075195,
      "learning_rate": 6.493877408056041e-05,
      "loss": 1.2096,
      "num_input_tokens_seen": 80615392,
      "step": 5010
    },
    {
      "epoch": 0.3510117593492518,
      "grad_norm": 4.8789825439453125,
      "learning_rate": 6.49317758318739e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 80631776,
      "step": 5011
    },
    {
      "epoch": 0.35108180759498103,
      "grad_norm": 4.271617412567139,
      "learning_rate": 6.49247775831874e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 80648160,
      "step": 5012
    },
    {
      "epoch": 0.3511518558407103,
      "grad_norm": 4.656182765960693,
      "learning_rate": 6.491777933450088e-05,
      "loss": 0.9566,
      "num_input_tokens_seen": 80664424,
      "step": 5013
    },
    {
      "epoch": 0.35122190408643955,
      "grad_norm": 6.627303600311279,
      "learning_rate": 6.491078108581437e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 80680128,
      "step": 5014
    },
    {
      "epoch": 0.35129195233216876,
      "grad_norm": 3.6189517974853516,
      "learning_rate": 6.490378283712786e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 80695848,
      "step": 5015
    },
    {
      "epoch": 0.351362000577898,
      "grad_norm": 3.58449387550354,
      "learning_rate": 6.489678458844133e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 80712232,
      "step": 5016
    },
    {
      "epoch": 0.3514320488236273,
      "grad_norm": 4.014143466949463,
      "learning_rate": 6.488978633975481e-05,
      "loss": 1.1271,
      "num_input_tokens_seen": 80726480,
      "step": 5017
    },
    {
      "epoch": 0.35150209706935653,
      "grad_norm": 4.461588382720947,
      "learning_rate": 6.488278809106831e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 80742776,
      "step": 5018
    },
    {
      "epoch": 0.35157214531508574,
      "grad_norm": 4.534054279327393,
      "learning_rate": 6.48757898423818e-05,
      "loss": 1.1009,
      "num_input_tokens_seen": 80758024,
      "step": 5019
    },
    {
      "epoch": 0.351642193560815,
      "grad_norm": 3.502699613571167,
      "learning_rate": 6.486879159369527e-05,
      "loss": 1.0564,
      "num_input_tokens_seen": 80774152,
      "step": 5020
    },
    {
      "epoch": 0.35171224180654426,
      "grad_norm": 4.463150978088379,
      "learning_rate": 6.486179334500876e-05,
      "loss": 0.9945,
      "num_input_tokens_seen": 80790528,
      "step": 5021
    },
    {
      "epoch": 0.3517822900522735,
      "grad_norm": 4.1127543449401855,
      "learning_rate": 6.485479509632225e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 80805400,
      "step": 5022
    },
    {
      "epoch": 0.3518523382980028,
      "grad_norm": 3.6113109588623047,
      "learning_rate": 6.484779684763572e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 80821584,
      "step": 5023
    },
    {
      "epoch": 0.351922386543732,
      "grad_norm": 4.167325019836426,
      "learning_rate": 6.484079859894921e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 80837968,
      "step": 5024
    },
    {
      "epoch": 0.35199243478946124,
      "grad_norm": 3.9422924518585205,
      "learning_rate": 6.48338003502627e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 80854352,
      "step": 5025
    },
    {
      "epoch": 0.3520624830351905,
      "grad_norm": 4.867110729217529,
      "learning_rate": 6.482680210157619e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 80870648,
      "step": 5026
    },
    {
      "epoch": 0.35213253128091976,
      "grad_norm": 4.714593887329102,
      "learning_rate": 6.481980385288967e-05,
      "loss": 1.0256,
      "num_input_tokens_seen": 80886704,
      "step": 5027
    },
    {
      "epoch": 0.35220257952664896,
      "grad_norm": 3.8926947116851807,
      "learning_rate": 6.481280560420315e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 80902184,
      "step": 5028
    },
    {
      "epoch": 0.3522726277723782,
      "grad_norm": 4.510727405548096,
      "learning_rate": 6.480580735551664e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 80917960,
      "step": 5029
    },
    {
      "epoch": 0.3523426760181075,
      "grad_norm": 3.6175239086151123,
      "learning_rate": 6.479880910683012e-05,
      "loss": 1.0692,
      "num_input_tokens_seen": 80934344,
      "step": 5030
    },
    {
      "epoch": 0.35241272426383674,
      "grad_norm": 4.112790584564209,
      "learning_rate": 6.47918108581436e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 80950336,
      "step": 5031
    },
    {
      "epoch": 0.35248277250956594,
      "grad_norm": 4.372056007385254,
      "learning_rate": 6.478481260945711e-05,
      "loss": 1.0732,
      "num_input_tokens_seen": 80966272,
      "step": 5032
    },
    {
      "epoch": 0.3525528207552952,
      "grad_norm": 5.2401204109191895,
      "learning_rate": 6.477781436077058e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 80981568,
      "step": 5033
    },
    {
      "epoch": 0.35262286900102446,
      "grad_norm": 4.032891273498535,
      "learning_rate": 6.477081611208406e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 80997384,
      "step": 5034
    },
    {
      "epoch": 0.3526929172467537,
      "grad_norm": 5.448423385620117,
      "learning_rate": 6.476381786339756e-05,
      "loss": 1.2136,
      "num_input_tokens_seen": 81013768,
      "step": 5035
    },
    {
      "epoch": 0.3527629654924829,
      "grad_norm": 3.5669469833374023,
      "learning_rate": 6.475681961471104e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 81030152,
      "step": 5036
    },
    {
      "epoch": 0.3528330137382122,
      "grad_norm": 3.4767303466796875,
      "learning_rate": 6.474982136602451e-05,
      "loss": 0.9563,
      "num_input_tokens_seen": 81046536,
      "step": 5037
    },
    {
      "epoch": 0.35290306198394145,
      "grad_norm": 4.859378814697266,
      "learning_rate": 6.474282311733801e-05,
      "loss": 1.2855,
      "num_input_tokens_seen": 81062528,
      "step": 5038
    },
    {
      "epoch": 0.3529731102296707,
      "grad_norm": 5.003366470336914,
      "learning_rate": 6.47358248686515e-05,
      "loss": 1.1317,
      "num_input_tokens_seen": 81078912,
      "step": 5039
    },
    {
      "epoch": 0.3530431584753999,
      "grad_norm": 3.9362549781799316,
      "learning_rate": 6.472882661996498e-05,
      "loss": 1.2051,
      "num_input_tokens_seen": 81095296,
      "step": 5040
    },
    {
      "epoch": 0.35311320672112917,
      "grad_norm": 3.319826364517212,
      "learning_rate": 6.472182837127847e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 81111640,
      "step": 5041
    },
    {
      "epoch": 0.35318325496685843,
      "grad_norm": 3.5816714763641357,
      "learning_rate": 6.471483012259195e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 81128024,
      "step": 5042
    },
    {
      "epoch": 0.3532533032125877,
      "grad_norm": 4.352350234985352,
      "learning_rate": 6.470783187390543e-05,
      "loss": 1.1754,
      "num_input_tokens_seen": 81143992,
      "step": 5043
    },
    {
      "epoch": 0.3533233514583169,
      "grad_norm": 3.4122314453125,
      "learning_rate": 6.470083362521892e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 81160376,
      "step": 5044
    },
    {
      "epoch": 0.35339339970404615,
      "grad_norm": 4.0952324867248535,
      "learning_rate": 6.46938353765324e-05,
      "loss": 0.9727,
      "num_input_tokens_seen": 81175968,
      "step": 5045
    },
    {
      "epoch": 0.3534634479497754,
      "grad_norm": 3.9099533557891846,
      "learning_rate": 6.46868371278459e-05,
      "loss": 1.0624,
      "num_input_tokens_seen": 81192352,
      "step": 5046
    },
    {
      "epoch": 0.35353349619550467,
      "grad_norm": 6.379274845123291,
      "learning_rate": 6.467983887915937e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 81208648,
      "step": 5047
    },
    {
      "epoch": 0.3536035444412339,
      "grad_norm": 3.9650473594665527,
      "learning_rate": 6.467284063047286e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 81224472,
      "step": 5048
    },
    {
      "epoch": 0.35367359268696313,
      "grad_norm": 3.7729573249816895,
      "learning_rate": 6.466584238178635e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 81240232,
      "step": 5049
    },
    {
      "epoch": 0.3537436409326924,
      "grad_norm": 4.012545585632324,
      "learning_rate": 6.465884413309982e-05,
      "loss": 1.0527,
      "num_input_tokens_seen": 81256616,
      "step": 5050
    },
    {
      "epoch": 0.35381368917842165,
      "grad_norm": 3.679382801055908,
      "learning_rate": 6.465184588441331e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 81272888,
      "step": 5051
    },
    {
      "epoch": 0.35388373742415086,
      "grad_norm": 3.897606134414673,
      "learning_rate": 6.464484763572681e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 81289272,
      "step": 5052
    },
    {
      "epoch": 0.3539537856698801,
      "grad_norm": 4.988255023956299,
      "learning_rate": 6.463784938704029e-05,
      "loss": 0.8484,
      "num_input_tokens_seen": 81305656,
      "step": 5053
    },
    {
      "epoch": 0.3540238339156094,
      "grad_norm": 4.226601600646973,
      "learning_rate": 6.463085113835376e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 81320912,
      "step": 5054
    },
    {
      "epoch": 0.35409388216133864,
      "grad_norm": 4.0905070304870605,
      "learning_rate": 6.462385288966725e-05,
      "loss": 1.2044,
      "num_input_tokens_seen": 81337296,
      "step": 5055
    },
    {
      "epoch": 0.3541639304070679,
      "grad_norm": 4.470916748046875,
      "learning_rate": 6.461685464098074e-05,
      "loss": 1.1198,
      "num_input_tokens_seen": 81353680,
      "step": 5056
    },
    {
      "epoch": 0.3542339786527971,
      "grad_norm": 3.8264098167419434,
      "learning_rate": 6.460985639229421e-05,
      "loss": 0.8444,
      "num_input_tokens_seen": 81370064,
      "step": 5057
    },
    {
      "epoch": 0.35430402689852636,
      "grad_norm": 5.07196569442749,
      "learning_rate": 6.460285814360772e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 81386368,
      "step": 5058
    },
    {
      "epoch": 0.3543740751442556,
      "grad_norm": 4.830010414123535,
      "learning_rate": 6.45958598949212e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 81402752,
      "step": 5059
    },
    {
      "epoch": 0.3544441233899849,
      "grad_norm": 3.5972540378570557,
      "learning_rate": 6.458886164623468e-05,
      "loss": 0.9466,
      "num_input_tokens_seen": 81418856,
      "step": 5060
    },
    {
      "epoch": 0.3545141716357141,
      "grad_norm": 4.840418815612793,
      "learning_rate": 6.458186339754816e-05,
      "loss": 1.0174,
      "num_input_tokens_seen": 81434344,
      "step": 5061
    },
    {
      "epoch": 0.35458421988144334,
      "grad_norm": 4.891697883605957,
      "learning_rate": 6.457486514886166e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 81450280,
      "step": 5062
    },
    {
      "epoch": 0.3546542681271726,
      "grad_norm": 3.7236123085021973,
      "learning_rate": 6.456786690017513e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 81466664,
      "step": 5063
    },
    {
      "epoch": 0.35472431637290186,
      "grad_norm": 3.6597838401794434,
      "learning_rate": 6.456086865148862e-05,
      "loss": 0.9648,
      "num_input_tokens_seen": 81483048,
      "step": 5064
    },
    {
      "epoch": 0.35479436461863106,
      "grad_norm": 4.048685073852539,
      "learning_rate": 6.455387040280211e-05,
      "loss": 1.0033,
      "num_input_tokens_seen": 81499080,
      "step": 5065
    },
    {
      "epoch": 0.3548644128643603,
      "grad_norm": 3.683549165725708,
      "learning_rate": 6.45468721541156e-05,
      "loss": 1.054,
      "num_input_tokens_seen": 81515464,
      "step": 5066
    },
    {
      "epoch": 0.3549344611100896,
      "grad_norm": 4.80827522277832,
      "learning_rate": 6.453987390542907e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 81530672,
      "step": 5067
    },
    {
      "epoch": 0.35500450935581884,
      "grad_norm": 3.6255602836608887,
      "learning_rate": 6.453287565674256e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 81546976,
      "step": 5068
    },
    {
      "epoch": 0.35507455760154805,
      "grad_norm": 3.430290460586548,
      "learning_rate": 6.452587740805605e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 81562936,
      "step": 5069
    },
    {
      "epoch": 0.3551446058472773,
      "grad_norm": 5.140942573547363,
      "learning_rate": 6.451887915936953e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 81579120,
      "step": 5070
    },
    {
      "epoch": 0.35521465409300657,
      "grad_norm": 4.5443115234375,
      "learning_rate": 6.451188091068301e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 81595504,
      "step": 5071
    },
    {
      "epoch": 0.3552847023387358,
      "grad_norm": 4.33146333694458,
      "learning_rate": 6.45048826619965e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 81611024,
      "step": 5072
    },
    {
      "epoch": 0.35535475058446503,
      "grad_norm": 4.212037563323975,
      "learning_rate": 6.449788441330999e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 81627208,
      "step": 5073
    },
    {
      "epoch": 0.3554247988301943,
      "grad_norm": 3.714611053466797,
      "learning_rate": 6.449088616462347e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 81642744,
      "step": 5074
    },
    {
      "epoch": 0.35549484707592355,
      "grad_norm": 3.985471487045288,
      "learning_rate": 6.448388791593696e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 81659128,
      "step": 5075
    },
    {
      "epoch": 0.3555648953216528,
      "grad_norm": 4.519073963165283,
      "learning_rate": 6.447688966725044e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 81675512,
      "step": 5076
    },
    {
      "epoch": 0.355634943567382,
      "grad_norm": 4.546297550201416,
      "learning_rate": 6.446989141856392e-05,
      "loss": 1.324,
      "num_input_tokens_seen": 81691528,
      "step": 5077
    },
    {
      "epoch": 0.35570499181311127,
      "grad_norm": 4.023989200592041,
      "learning_rate": 6.446289316987741e-05,
      "loss": 1.068,
      "num_input_tokens_seen": 81707912,
      "step": 5078
    },
    {
      "epoch": 0.35577504005884053,
      "grad_norm": 4.442357540130615,
      "learning_rate": 6.445589492119091e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 81724296,
      "step": 5079
    },
    {
      "epoch": 0.3558450883045698,
      "grad_norm": 3.63273286819458,
      "learning_rate": 6.444889667250438e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 81740112,
      "step": 5080
    },
    {
      "epoch": 0.355915136550299,
      "grad_norm": 3.8844716548919678,
      "learning_rate": 6.444189842381786e-05,
      "loss": 1.1389,
      "num_input_tokens_seen": 81756024,
      "step": 5081
    },
    {
      "epoch": 0.35598518479602825,
      "grad_norm": 3.8603484630584717,
      "learning_rate": 6.443490017513135e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 81772408,
      "step": 5082
    },
    {
      "epoch": 0.3560552330417575,
      "grad_norm": 4.305675029754639,
      "learning_rate": 6.442790192644484e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 81787992,
      "step": 5083
    },
    {
      "epoch": 0.3561252812874868,
      "grad_norm": 5.944203853607178,
      "learning_rate": 6.442090367775833e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 81804032,
      "step": 5084
    },
    {
      "epoch": 0.356195329533216,
      "grad_norm": 5.269783020019531,
      "learning_rate": 6.441390542907181e-05,
      "loss": 1.0697,
      "num_input_tokens_seen": 81820416,
      "step": 5085
    },
    {
      "epoch": 0.35626537777894524,
      "grad_norm": 3.775933027267456,
      "learning_rate": 6.44069071803853e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 81836712,
      "step": 5086
    },
    {
      "epoch": 0.3563354260246745,
      "grad_norm": 4.133227825164795,
      "learning_rate": 6.439990893169878e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 81853096,
      "step": 5087
    },
    {
      "epoch": 0.35640547427040375,
      "grad_norm": 4.418367862701416,
      "learning_rate": 6.439291068301225e-05,
      "loss": 1.1836,
      "num_input_tokens_seen": 81869480,
      "step": 5088
    },
    {
      "epoch": 0.356475522516133,
      "grad_norm": 3.584392786026001,
      "learning_rate": 6.438591243432575e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 81885864,
      "step": 5089
    },
    {
      "epoch": 0.3565455707618622,
      "grad_norm": 4.216940402984619,
      "learning_rate": 6.437891418563923e-05,
      "loss": 0.8602,
      "num_input_tokens_seen": 81902248,
      "step": 5090
    },
    {
      "epoch": 0.3566156190075915,
      "grad_norm": 4.383372783660889,
      "learning_rate": 6.437191593695272e-05,
      "loss": 0.9763,
      "num_input_tokens_seen": 81918464,
      "step": 5091
    },
    {
      "epoch": 0.35668566725332074,
      "grad_norm": 4.06666374206543,
      "learning_rate": 6.436491768826621e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 81934848,
      "step": 5092
    },
    {
      "epoch": 0.35675571549905,
      "grad_norm": 5.485066890716553,
      "learning_rate": 6.43579194395797e-05,
      "loss": 0.9188,
      "num_input_tokens_seen": 81950696,
      "step": 5093
    },
    {
      "epoch": 0.3568257637447792,
      "grad_norm": 6.794841766357422,
      "learning_rate": 6.435092119089317e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 81967080,
      "step": 5094
    },
    {
      "epoch": 0.35689581199050846,
      "grad_norm": 3.531291961669922,
      "learning_rate": 6.434392294220666e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 81983464,
      "step": 5095
    },
    {
      "epoch": 0.3569658602362377,
      "grad_norm": 3.694018840789795,
      "learning_rate": 6.433692469352015e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 81999848,
      "step": 5096
    },
    {
      "epoch": 0.357035908481967,
      "grad_norm": 6.933582305908203,
      "learning_rate": 6.432992644483362e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 82015304,
      "step": 5097
    },
    {
      "epoch": 0.3571059567276962,
      "grad_norm": 5.904866695404053,
      "learning_rate": 6.432292819614711e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 82031688,
      "step": 5098
    },
    {
      "epoch": 0.35717600497342544,
      "grad_norm": 4.199756145477295,
      "learning_rate": 6.43159299474606e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 82047336,
      "step": 5099
    },
    {
      "epoch": 0.3572460532191547,
      "grad_norm": 3.703000783920288,
      "learning_rate": 6.430893169877409e-05,
      "loss": 0.8503,
      "num_input_tokens_seen": 82063720,
      "step": 5100
    },
    {
      "epoch": 0.35731610146488396,
      "grad_norm": 4.844930171966553,
      "learning_rate": 6.430193345008756e-05,
      "loss": 1.0255,
      "num_input_tokens_seen": 82079632,
      "step": 5101
    },
    {
      "epoch": 0.35738614971061317,
      "grad_norm": 3.870488166809082,
      "learning_rate": 6.429493520140105e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 82094864,
      "step": 5102
    },
    {
      "epoch": 0.3574561979563424,
      "grad_norm": 3.9125707149505615,
      "learning_rate": 6.428793695271454e-05,
      "loss": 0.9626,
      "num_input_tokens_seen": 82111136,
      "step": 5103
    },
    {
      "epoch": 0.3575262462020717,
      "grad_norm": 4.347132205963135,
      "learning_rate": 6.428093870402803e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 82127064,
      "step": 5104
    },
    {
      "epoch": 0.35759629444780094,
      "grad_norm": 3.739053964614868,
      "learning_rate": 6.42739404553415e-05,
      "loss": 1.201,
      "num_input_tokens_seen": 82143448,
      "step": 5105
    },
    {
      "epoch": 0.35766634269353015,
      "grad_norm": 4.781857967376709,
      "learning_rate": 6.4266942206655e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 82159832,
      "step": 5106
    },
    {
      "epoch": 0.3577363909392594,
      "grad_norm": 4.3711700439453125,
      "learning_rate": 6.425994395796848e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 82175944,
      "step": 5107
    },
    {
      "epoch": 0.35780643918498867,
      "grad_norm": 3.6916282176971436,
      "learning_rate": 6.425294570928196e-05,
      "loss": 0.787,
      "num_input_tokens_seen": 82192304,
      "step": 5108
    },
    {
      "epoch": 0.3578764874307179,
      "grad_norm": 4.418915271759033,
      "learning_rate": 6.424594746059545e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 82208080,
      "step": 5109
    },
    {
      "epoch": 0.35794653567644713,
      "grad_norm": 3.9138340950012207,
      "learning_rate": 6.423894921190893e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 82224464,
      "step": 5110
    },
    {
      "epoch": 0.3580165839221764,
      "grad_norm": 3.99479079246521,
      "learning_rate": 6.423195096322242e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 82240664,
      "step": 5111
    },
    {
      "epoch": 0.35808663216790565,
      "grad_norm": 4.260537147521973,
      "learning_rate": 6.422495271453591e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 82257048,
      "step": 5112
    },
    {
      "epoch": 0.3581566804136349,
      "grad_norm": 3.5181097984313965,
      "learning_rate": 6.42179544658494e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 82273432,
      "step": 5113
    },
    {
      "epoch": 0.3582267286593641,
      "grad_norm": 5.96913480758667,
      "learning_rate": 6.421095621716287e-05,
      "loss": 0.8867,
      "num_input_tokens_seen": 82289816,
      "step": 5114
    },
    {
      "epoch": 0.3582967769050934,
      "grad_norm": 4.628411769866943,
      "learning_rate": 6.420395796847635e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 82305784,
      "step": 5115
    },
    {
      "epoch": 0.35836682515082263,
      "grad_norm": 3.5981955528259277,
      "learning_rate": 6.419695971978985e-05,
      "loss": 0.9182,
      "num_input_tokens_seen": 82321384,
      "step": 5116
    },
    {
      "epoch": 0.3584368733965519,
      "grad_norm": 4.410891056060791,
      "learning_rate": 6.418996147110333e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 82336184,
      "step": 5117
    },
    {
      "epoch": 0.3585069216422811,
      "grad_norm": 4.316674709320068,
      "learning_rate": 6.418296322241682e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 82351520,
      "step": 5118
    },
    {
      "epoch": 0.35857696988801036,
      "grad_norm": 5.662688255310059,
      "learning_rate": 6.41759649737303e-05,
      "loss": 1.1212,
      "num_input_tokens_seen": 82367904,
      "step": 5119
    },
    {
      "epoch": 0.3586470181337396,
      "grad_norm": 4.5336151123046875,
      "learning_rate": 6.416896672504379e-05,
      "loss": 1.0093,
      "num_input_tokens_seen": 82384288,
      "step": 5120
    },
    {
      "epoch": 0.3587170663794689,
      "grad_norm": 6.43854284286499,
      "learning_rate": 6.416196847635727e-05,
      "loss": 0.9434,
      "num_input_tokens_seen": 82400120,
      "step": 5121
    },
    {
      "epoch": 0.3587871146251981,
      "grad_norm": 3.519869089126587,
      "learning_rate": 6.415497022767076e-05,
      "loss": 0.9704,
      "num_input_tokens_seen": 82416504,
      "step": 5122
    },
    {
      "epoch": 0.35885716287092734,
      "grad_norm": 4.426568508148193,
      "learning_rate": 6.414797197898425e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 82431936,
      "step": 5123
    },
    {
      "epoch": 0.3589272111166566,
      "grad_norm": 10.392409324645996,
      "learning_rate": 6.414097373029773e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 82447232,
      "step": 5124
    },
    {
      "epoch": 0.35899725936238586,
      "grad_norm": 4.133431434631348,
      "learning_rate": 6.413397548161121e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 82462648,
      "step": 5125
    },
    {
      "epoch": 0.3590673076081151,
      "grad_norm": 5.43566370010376,
      "learning_rate": 6.41269772329247e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 82478536,
      "step": 5126
    },
    {
      "epoch": 0.3591373558538443,
      "grad_norm": 4.205079555511475,
      "learning_rate": 6.411997898423819e-05,
      "loss": 1.1152,
      "num_input_tokens_seen": 82494224,
      "step": 5127
    },
    {
      "epoch": 0.3592074040995736,
      "grad_norm": 4.165416240692139,
      "learning_rate": 6.411298073555166e-05,
      "loss": 1.3017,
      "num_input_tokens_seen": 82510608,
      "step": 5128
    },
    {
      "epoch": 0.35927745234530284,
      "grad_norm": 3.7855117321014404,
      "learning_rate": 6.410598248686515e-05,
      "loss": 0.8362,
      "num_input_tokens_seen": 82526992,
      "step": 5129
    },
    {
      "epoch": 0.3593475005910321,
      "grad_norm": 4.406207084655762,
      "learning_rate": 6.409898423817864e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 82543376,
      "step": 5130
    },
    {
      "epoch": 0.3594175488367613,
      "grad_norm": 4.228625774383545,
      "learning_rate": 6.409198598949213e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 82559760,
      "step": 5131
    },
    {
      "epoch": 0.35948759708249056,
      "grad_norm": 3.6679983139038086,
      "learning_rate": 6.40849877408056e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 82575552,
      "step": 5132
    },
    {
      "epoch": 0.3595576453282198,
      "grad_norm": 4.011179447174072,
      "learning_rate": 6.40779894921191e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 82591936,
      "step": 5133
    },
    {
      "epoch": 0.3596276935739491,
      "grad_norm": 4.861363410949707,
      "learning_rate": 6.407099124343258e-05,
      "loss": 1.1077,
      "num_input_tokens_seen": 82608320,
      "step": 5134
    },
    {
      "epoch": 0.3596977418196783,
      "grad_norm": 4.128578186035156,
      "learning_rate": 6.406399299474605e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 82624704,
      "step": 5135
    },
    {
      "epoch": 0.35976779006540754,
      "grad_norm": 4.036421775817871,
      "learning_rate": 6.405699474605954e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 82641088,
      "step": 5136
    },
    {
      "epoch": 0.3598378383111368,
      "grad_norm": 4.536168098449707,
      "learning_rate": 6.404999649737303e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 82657472,
      "step": 5137
    },
    {
      "epoch": 0.35990788655686606,
      "grad_norm": 3.665916681289673,
      "learning_rate": 6.404299824868652e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 82673856,
      "step": 5138
    },
    {
      "epoch": 0.35997793480259527,
      "grad_norm": 3.798205852508545,
      "learning_rate": 6.403600000000001e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 82690240,
      "step": 5139
    },
    {
      "epoch": 0.3600479830483245,
      "grad_norm": 3.9616305828094482,
      "learning_rate": 6.40290017513135e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 82706624,
      "step": 5140
    },
    {
      "epoch": 0.3601180312940538,
      "grad_norm": 4.6059489250183105,
      "learning_rate": 6.402200350262697e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 82723008,
      "step": 5141
    },
    {
      "epoch": 0.36018807953978305,
      "grad_norm": 4.2935943603515625,
      "learning_rate": 6.401500525394045e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 82739392,
      "step": 5142
    },
    {
      "epoch": 0.36025812778551225,
      "grad_norm": 4.02174711227417,
      "learning_rate": 6.400800700525395e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 82755600,
      "step": 5143
    },
    {
      "epoch": 0.3603281760312415,
      "grad_norm": 4.0431599617004395,
      "learning_rate": 6.400100875656744e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 82771592,
      "step": 5144
    },
    {
      "epoch": 0.36039822427697077,
      "grad_norm": 3.6921310424804688,
      "learning_rate": 6.399401050788091e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 82787728,
      "step": 5145
    },
    {
      "epoch": 0.36046827252270003,
      "grad_norm": 4.27170991897583,
      "learning_rate": 6.39870122591944e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 82803152,
      "step": 5146
    },
    {
      "epoch": 0.36053832076842923,
      "grad_norm": 4.670827865600586,
      "learning_rate": 6.398001401050789e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 82819536,
      "step": 5147
    },
    {
      "epoch": 0.3606083690141585,
      "grad_norm": 3.6219654083251953,
      "learning_rate": 6.397301576182136e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 82835920,
      "step": 5148
    },
    {
      "epoch": 0.36067841725988775,
      "grad_norm": 3.53466796875,
      "learning_rate": 6.396601751313485e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 82852304,
      "step": 5149
    },
    {
      "epoch": 0.360748465505617,
      "grad_norm": 4.027638912200928,
      "learning_rate": 6.395901926444834e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 82867816,
      "step": 5150
    },
    {
      "epoch": 0.3608185137513462,
      "grad_norm": 5.701491832733154,
      "learning_rate": 6.395202101576183e-05,
      "loss": 1.2476,
      "num_input_tokens_seen": 82883480,
      "step": 5151
    },
    {
      "epoch": 0.3608885619970755,
      "grad_norm": 4.156428337097168,
      "learning_rate": 6.39450227670753e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 82899608,
      "step": 5152
    },
    {
      "epoch": 0.36095861024280473,
      "grad_norm": 5.278023719787598,
      "learning_rate": 6.39380245183888e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 82915656,
      "step": 5153
    },
    {
      "epoch": 0.361028658488534,
      "grad_norm": 3.6892948150634766,
      "learning_rate": 6.393102626970228e-05,
      "loss": 1.0063,
      "num_input_tokens_seen": 82931632,
      "step": 5154
    },
    {
      "epoch": 0.3610987067342632,
      "grad_norm": 5.179676055908203,
      "learning_rate": 6.392402802101576e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 82947344,
      "step": 5155
    },
    {
      "epoch": 0.36116875497999246,
      "grad_norm": 4.948189735412598,
      "learning_rate": 6.391702977232925e-05,
      "loss": 1.056,
      "num_input_tokens_seen": 82963720,
      "step": 5156
    },
    {
      "epoch": 0.3612388032257217,
      "grad_norm": 4.465184688568115,
      "learning_rate": 6.391003152364274e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 82980048,
      "step": 5157
    },
    {
      "epoch": 0.361308851471451,
      "grad_norm": 4.053642749786377,
      "learning_rate": 6.390303327495622e-05,
      "loss": 1.1481,
      "num_input_tokens_seen": 82996432,
      "step": 5158
    },
    {
      "epoch": 0.36137889971718024,
      "grad_norm": 8.422308921813965,
      "learning_rate": 6.38960350262697e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 83012560,
      "step": 5159
    },
    {
      "epoch": 0.36144894796290944,
      "grad_norm": 3.4304730892181396,
      "learning_rate": 6.38890367775832e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 83028680,
      "step": 5160
    },
    {
      "epoch": 0.3615189962086387,
      "grad_norm": 9.87295913696289,
      "learning_rate": 6.388203852889668e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 83045064,
      "step": 5161
    },
    {
      "epoch": 0.36158904445436796,
      "grad_norm": 3.7000608444213867,
      "learning_rate": 6.387504028021015e-05,
      "loss": 1.0758,
      "num_input_tokens_seen": 83061448,
      "step": 5162
    },
    {
      "epoch": 0.3616590927000972,
      "grad_norm": 3.5490283966064453,
      "learning_rate": 6.386804203152364e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 83077176,
      "step": 5163
    },
    {
      "epoch": 0.3617291409458264,
      "grad_norm": 3.850770950317383,
      "learning_rate": 6.386104378283714e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 83093560,
      "step": 5164
    },
    {
      "epoch": 0.3617991891915557,
      "grad_norm": 5.09017276763916,
      "learning_rate": 6.385404553415062e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 83109752,
      "step": 5165
    },
    {
      "epoch": 0.36186923743728494,
      "grad_norm": 4.801665782928467,
      "learning_rate": 6.38470472854641e-05,
      "loss": 1.0909,
      "num_input_tokens_seen": 83125048,
      "step": 5166
    },
    {
      "epoch": 0.3619392856830142,
      "grad_norm": 3.954345941543579,
      "learning_rate": 6.38400490367776e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 83140808,
      "step": 5167
    },
    {
      "epoch": 0.3620093339287434,
      "grad_norm": 4.874080657958984,
      "learning_rate": 6.383305078809107e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 83157176,
      "step": 5168
    },
    {
      "epoch": 0.36207938217447266,
      "grad_norm": 4.3997111320495605,
      "learning_rate": 6.382605253940454e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 83173560,
      "step": 5169
    },
    {
      "epoch": 0.3621494304202019,
      "grad_norm": 4.431540489196777,
      "learning_rate": 6.381905429071805e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 83189864,
      "step": 5170
    },
    {
      "epoch": 0.3622194786659312,
      "grad_norm": 4.48107385635376,
      "learning_rate": 6.381205604203153e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 83205560,
      "step": 5171
    },
    {
      "epoch": 0.3622895269116604,
      "grad_norm": 4.369350910186768,
      "learning_rate": 6.380505779334501e-05,
      "loss": 1.0877,
      "num_input_tokens_seen": 83221544,
      "step": 5172
    },
    {
      "epoch": 0.36235957515738965,
      "grad_norm": 3.8510024547576904,
      "learning_rate": 6.37980595446585e-05,
      "loss": 0.8895,
      "num_input_tokens_seen": 83237928,
      "step": 5173
    },
    {
      "epoch": 0.3624296234031189,
      "grad_norm": 3.7452402114868164,
      "learning_rate": 6.379106129597199e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 83254168,
      "step": 5174
    },
    {
      "epoch": 0.36249967164884817,
      "grad_norm": 4.53076171875,
      "learning_rate": 6.378406304728546e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 83269568,
      "step": 5175
    },
    {
      "epoch": 0.36256971989457737,
      "grad_norm": 3.729602813720703,
      "learning_rate": 6.377706479859895e-05,
      "loss": 1.2105,
      "num_input_tokens_seen": 83285952,
      "step": 5176
    },
    {
      "epoch": 0.36263976814030663,
      "grad_norm": 4.085333824157715,
      "learning_rate": 6.377006654991244e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 83302200,
      "step": 5177
    },
    {
      "epoch": 0.3627098163860359,
      "grad_norm": 3.9202303886413574,
      "learning_rate": 6.376306830122593e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 83318584,
      "step": 5178
    },
    {
      "epoch": 0.36277986463176515,
      "grad_norm": 4.10648775100708,
      "learning_rate": 6.37560700525394e-05,
      "loss": 1.3052,
      "num_input_tokens_seen": 83334288,
      "step": 5179
    },
    {
      "epoch": 0.36284991287749435,
      "grad_norm": 3.975217580795288,
      "learning_rate": 6.374907180385289e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 83350096,
      "step": 5180
    },
    {
      "epoch": 0.3629199611232236,
      "grad_norm": 4.207096099853516,
      "learning_rate": 6.374207355516638e-05,
      "loss": 1.1396,
      "num_input_tokens_seen": 83366480,
      "step": 5181
    },
    {
      "epoch": 0.36299000936895287,
      "grad_norm": 3.9960830211639404,
      "learning_rate": 6.373507530647986e-05,
      "loss": 1.1971,
      "num_input_tokens_seen": 83381832,
      "step": 5182
    },
    {
      "epoch": 0.36306005761468213,
      "grad_norm": 4.142012596130371,
      "learning_rate": 6.372807705779334e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 83398216,
      "step": 5183
    },
    {
      "epoch": 0.36313010586041133,
      "grad_norm": 3.8692433834075928,
      "learning_rate": 6.372107880910685e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 83414600,
      "step": 5184
    },
    {
      "epoch": 0.3632001541061406,
      "grad_norm": 3.663544178009033,
      "learning_rate": 6.371408056042032e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 83430984,
      "step": 5185
    },
    {
      "epoch": 0.36327020235186985,
      "grad_norm": 4.056418418884277,
      "learning_rate": 6.37070823117338e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 83447368,
      "step": 5186
    },
    {
      "epoch": 0.3633402505975991,
      "grad_norm": 4.209747314453125,
      "learning_rate": 6.37000840630473e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 83463752,
      "step": 5187
    },
    {
      "epoch": 0.3634102988433283,
      "grad_norm": 4.93091344833374,
      "learning_rate": 6.369308581436077e-05,
      "loss": 1.2046,
      "num_input_tokens_seen": 83479424,
      "step": 5188
    },
    {
      "epoch": 0.3634803470890576,
      "grad_norm": 3.6523993015289307,
      "learning_rate": 6.368608756567425e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 83495808,
      "step": 5189
    },
    {
      "epoch": 0.36355039533478684,
      "grad_norm": 4.8949294090271,
      "learning_rate": 6.367908931698775e-05,
      "loss": 0.8928,
      "num_input_tokens_seen": 83511448,
      "step": 5190
    },
    {
      "epoch": 0.3636204435805161,
      "grad_norm": 5.856332778930664,
      "learning_rate": 6.367209106830124e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 83526664,
      "step": 5191
    },
    {
      "epoch": 0.3636904918262453,
      "grad_norm": 3.762014865875244,
      "learning_rate": 6.366509281961471e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 83542792,
      "step": 5192
    },
    {
      "epoch": 0.36376054007197456,
      "grad_norm": 4.075290203094482,
      "learning_rate": 6.36580945709282e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 83558992,
      "step": 5193
    },
    {
      "epoch": 0.3638305883177038,
      "grad_norm": 4.124780178070068,
      "learning_rate": 6.365109632224169e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 83575376,
      "step": 5194
    },
    {
      "epoch": 0.3639006365634331,
      "grad_norm": 6.1159210205078125,
      "learning_rate": 6.364409807355517e-05,
      "loss": 0.9209,
      "num_input_tokens_seen": 83591400,
      "step": 5195
    },
    {
      "epoch": 0.36397068480916234,
      "grad_norm": 3.8839027881622314,
      "learning_rate": 6.363709982486865e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 83607784,
      "step": 5196
    },
    {
      "epoch": 0.36404073305489154,
      "grad_norm": 4.260892391204834,
      "learning_rate": 6.363010157618214e-05,
      "loss": 1.0747,
      "num_input_tokens_seen": 83623944,
      "step": 5197
    },
    {
      "epoch": 0.3641107813006208,
      "grad_norm": 4.111022472381592,
      "learning_rate": 6.362310332749563e-05,
      "loss": 1.2594,
      "num_input_tokens_seen": 83639408,
      "step": 5198
    },
    {
      "epoch": 0.36418082954635006,
      "grad_norm": 3.567676305770874,
      "learning_rate": 6.361610507880911e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 83655496,
      "step": 5199
    },
    {
      "epoch": 0.3642508777920793,
      "grad_norm": 4.935754299163818,
      "learning_rate": 6.36091068301226e-05,
      "loss": 1.2028,
      "num_input_tokens_seen": 83671016,
      "step": 5200
    },
    {
      "epoch": 0.3642508777920793,
      "eval_loss": 1.129547119140625,
      "eval_runtime": 0.1857,
      "eval_samples_per_second": 5.386,
      "eval_steps_per_second": 5.386,
      "num_input_tokens_seen": 83671016,
      "step": 5200
    },
    {
      "epoch": 0.3643209260378085,
      "grad_norm": 3.8546817302703857,
      "learning_rate": 6.360210858143608e-05,
      "loss": 0.9873,
      "num_input_tokens_seen": 83685736,
      "step": 5201
    },
    {
      "epoch": 0.3643909742835378,
      "grad_norm": 3.900425910949707,
      "learning_rate": 6.359511033274956e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 83702120,
      "step": 5202
    },
    {
      "epoch": 0.36446102252926704,
      "grad_norm": 4.270096302032471,
      "learning_rate": 6.358811208406305e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 83718504,
      "step": 5203
    },
    {
      "epoch": 0.3645310707749963,
      "grad_norm": 5.027628421783447,
      "learning_rate": 6.358111383537655e-05,
      "loss": 1.1363,
      "num_input_tokens_seen": 83734888,
      "step": 5204
    },
    {
      "epoch": 0.3646011190207255,
      "grad_norm": 4.843371868133545,
      "learning_rate": 6.357411558669002e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 83749488,
      "step": 5205
    },
    {
      "epoch": 0.36467116726645477,
      "grad_norm": 7.530435562133789,
      "learning_rate": 6.35671173380035e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 83765872,
      "step": 5206
    },
    {
      "epoch": 0.364741215512184,
      "grad_norm": 4.028171062469482,
      "learning_rate": 6.356011908931699e-05,
      "loss": 1.2011,
      "num_input_tokens_seen": 83781936,
      "step": 5207
    },
    {
      "epoch": 0.3648112637579133,
      "grad_norm": 6.744492053985596,
      "learning_rate": 6.355312084063048e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 83797520,
      "step": 5208
    },
    {
      "epoch": 0.3648813120036425,
      "grad_norm": 3.9689910411834717,
      "learning_rate": 6.354612259194395e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 83813872,
      "step": 5209
    },
    {
      "epoch": 0.36495136024937175,
      "grad_norm": 4.990142345428467,
      "learning_rate": 6.353912434325745e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 83830256,
      "step": 5210
    },
    {
      "epoch": 0.365021408495101,
      "grad_norm": 4.547253131866455,
      "learning_rate": 6.353212609457094e-05,
      "loss": 1.1825,
      "num_input_tokens_seen": 83846640,
      "step": 5211
    },
    {
      "epoch": 0.36509145674083027,
      "grad_norm": 4.108243465423584,
      "learning_rate": 6.352512784588442e-05,
      "loss": 1.1827,
      "num_input_tokens_seen": 83863024,
      "step": 5212
    },
    {
      "epoch": 0.36516150498655947,
      "grad_norm": 4.540827751159668,
      "learning_rate": 6.351812959719789e-05,
      "loss": 1.0034,
      "num_input_tokens_seen": 83878976,
      "step": 5213
    },
    {
      "epoch": 0.36523155323228873,
      "grad_norm": 5.3233842849731445,
      "learning_rate": 6.35111313485114e-05,
      "loss": 1.2247,
      "num_input_tokens_seen": 83895360,
      "step": 5214
    },
    {
      "epoch": 0.365301601478018,
      "grad_norm": 5.161661624908447,
      "learning_rate": 6.350413309982487e-05,
      "loss": 1.2067,
      "num_input_tokens_seen": 83910064,
      "step": 5215
    },
    {
      "epoch": 0.36537164972374725,
      "grad_norm": 4.908864498138428,
      "learning_rate": 6.349713485113836e-05,
      "loss": 1.1748,
      "num_input_tokens_seen": 83926448,
      "step": 5216
    },
    {
      "epoch": 0.36544169796947645,
      "grad_norm": 5.954193592071533,
      "learning_rate": 6.349013660245185e-05,
      "loss": 0.99,
      "num_input_tokens_seen": 83942248,
      "step": 5217
    },
    {
      "epoch": 0.3655117462152057,
      "grad_norm": 3.5276272296905518,
      "learning_rate": 6.348313835376534e-05,
      "loss": 0.9637,
      "num_input_tokens_seen": 83958632,
      "step": 5218
    },
    {
      "epoch": 0.365581794460935,
      "grad_norm": 3.736661195755005,
      "learning_rate": 6.347614010507881e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 83975016,
      "step": 5219
    },
    {
      "epoch": 0.36565184270666423,
      "grad_norm": 5.434671401977539,
      "learning_rate": 6.34691418563923e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 83990424,
      "step": 5220
    },
    {
      "epoch": 0.36572189095239344,
      "grad_norm": 3.9301772117614746,
      "learning_rate": 6.346214360770579e-05,
      "loss": 1.0278,
      "num_input_tokens_seen": 84006808,
      "step": 5221
    },
    {
      "epoch": 0.3657919391981227,
      "grad_norm": 5.101827621459961,
      "learning_rate": 6.345514535901926e-05,
      "loss": 1.2129,
      "num_input_tokens_seen": 84022624,
      "step": 5222
    },
    {
      "epoch": 0.36586198744385195,
      "grad_norm": 4.042179584503174,
      "learning_rate": 6.344814711033275e-05,
      "loss": 1.2996,
      "num_input_tokens_seen": 84038688,
      "step": 5223
    },
    {
      "epoch": 0.3659320356895812,
      "grad_norm": 4.2309441566467285,
      "learning_rate": 6.344114886164624e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 84055072,
      "step": 5224
    },
    {
      "epoch": 0.3660020839353104,
      "grad_norm": 6.73452615737915,
      "learning_rate": 6.343415061295973e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 84071456,
      "step": 5225
    },
    {
      "epoch": 0.3660721321810397,
      "grad_norm": 3.684497833251953,
      "learning_rate": 6.34271523642732e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 84087840,
      "step": 5226
    },
    {
      "epoch": 0.36614218042676894,
      "grad_norm": 3.7974796295166016,
      "learning_rate": 6.342015411558669e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 84103456,
      "step": 5227
    },
    {
      "epoch": 0.3662122286724982,
      "grad_norm": 4.681473255157471,
      "learning_rate": 6.341315586690018e-05,
      "loss": 0.9202,
      "num_input_tokens_seen": 84119840,
      "step": 5228
    },
    {
      "epoch": 0.36628227691822746,
      "grad_norm": 4.197212219238281,
      "learning_rate": 6.340615761821366e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 84136224,
      "step": 5229
    },
    {
      "epoch": 0.36635232516395666,
      "grad_norm": 4.1414794921875,
      "learning_rate": 6.339915936952716e-05,
      "loss": 1.1421,
      "num_input_tokens_seen": 84152552,
      "step": 5230
    },
    {
      "epoch": 0.3664223734096859,
      "grad_norm": 4.138907432556152,
      "learning_rate": 6.339216112084065e-05,
      "loss": 1.0841,
      "num_input_tokens_seen": 84168936,
      "step": 5231
    },
    {
      "epoch": 0.3664924216554152,
      "grad_norm": 4.723425388336182,
      "learning_rate": 6.338516287215412e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 84185320,
      "step": 5232
    },
    {
      "epoch": 0.36656246990114444,
      "grad_norm": 4.167308330535889,
      "learning_rate": 6.33781646234676e-05,
      "loss": 1.2636,
      "num_input_tokens_seen": 84201704,
      "step": 5233
    },
    {
      "epoch": 0.36663251814687364,
      "grad_norm": 3.832829236984253,
      "learning_rate": 6.337116637478109e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 84217216,
      "step": 5234
    },
    {
      "epoch": 0.3667025663926029,
      "grad_norm": 6.1642842292785645,
      "learning_rate": 6.336416812609457e-05,
      "loss": 0.8102,
      "num_input_tokens_seen": 84232896,
      "step": 5235
    },
    {
      "epoch": 0.36677261463833216,
      "grad_norm": 3.948350429534912,
      "learning_rate": 6.335716987740806e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 84248448,
      "step": 5236
    },
    {
      "epoch": 0.3668426628840614,
      "grad_norm": 3.6216750144958496,
      "learning_rate": 6.335017162872155e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 84264832,
      "step": 5237
    },
    {
      "epoch": 0.3669127111297906,
      "grad_norm": 5.787931442260742,
      "learning_rate": 6.334317338003504e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 84281216,
      "step": 5238
    },
    {
      "epoch": 0.3669827593755199,
      "grad_norm": 4.830391883850098,
      "learning_rate": 6.333617513134851e-05,
      "loss": 1.3014,
      "num_input_tokens_seen": 84297352,
      "step": 5239
    },
    {
      "epoch": 0.36705280762124914,
      "grad_norm": 3.839425563812256,
      "learning_rate": 6.332917688266199e-05,
      "loss": 1.031,
      "num_input_tokens_seen": 84313608,
      "step": 5240
    },
    {
      "epoch": 0.3671228558669784,
      "grad_norm": 3.963012456893921,
      "learning_rate": 6.332217863397549e-05,
      "loss": 1.0232,
      "num_input_tokens_seen": 84329680,
      "step": 5241
    },
    {
      "epoch": 0.3671929041127076,
      "grad_norm": 3.4596047401428223,
      "learning_rate": 6.331518038528897e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 84346064,
      "step": 5242
    },
    {
      "epoch": 0.36726295235843687,
      "grad_norm": 5.7928290367126465,
      "learning_rate": 6.330818213660246e-05,
      "loss": 1.2292,
      "num_input_tokens_seen": 84361800,
      "step": 5243
    },
    {
      "epoch": 0.3673330006041661,
      "grad_norm": 3.5012640953063965,
      "learning_rate": 6.330118388791594e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 84378184,
      "step": 5244
    },
    {
      "epoch": 0.3674030488498954,
      "grad_norm": 4.464978218078613,
      "learning_rate": 6.329418563922943e-05,
      "loss": 1.2258,
      "num_input_tokens_seen": 84394568,
      "step": 5245
    },
    {
      "epoch": 0.3674730970956246,
      "grad_norm": 3.4716012477874756,
      "learning_rate": 6.328718739054291e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 84410584,
      "step": 5246
    },
    {
      "epoch": 0.36754314534135385,
      "grad_norm": 4.010568618774414,
      "learning_rate": 6.32801891418564e-05,
      "loss": 1.0823,
      "num_input_tokens_seen": 84426968,
      "step": 5247
    },
    {
      "epoch": 0.3676131935870831,
      "grad_norm": 3.763718605041504,
      "learning_rate": 6.327319089316989e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 84443352,
      "step": 5248
    },
    {
      "epoch": 0.36768324183281237,
      "grad_norm": 5.381477355957031,
      "learning_rate": 6.326619264448336e-05,
      "loss": 1.0387,
      "num_input_tokens_seen": 84459736,
      "step": 5249
    },
    {
      "epoch": 0.3677532900785416,
      "grad_norm": 3.6646018028259277,
      "learning_rate": 6.325919439579685e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 84476120,
      "step": 5250
    },
    {
      "epoch": 0.36782333832427083,
      "grad_norm": 4.005465507507324,
      "learning_rate": 6.325219614711034e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 84492400,
      "step": 5251
    },
    {
      "epoch": 0.3678933865700001,
      "grad_norm": 3.4287807941436768,
      "learning_rate": 6.324519789842383e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 84508720,
      "step": 5252
    },
    {
      "epoch": 0.36796343481572935,
      "grad_norm": 3.8715076446533203,
      "learning_rate": 6.32381996497373e-05,
      "loss": 1.3025,
      "num_input_tokens_seen": 84524592,
      "step": 5253
    },
    {
      "epoch": 0.36803348306145856,
      "grad_norm": 3.4789586067199707,
      "learning_rate": 6.323120140105079e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 84540176,
      "step": 5254
    },
    {
      "epoch": 0.3681035313071878,
      "grad_norm": 3.992988348007202,
      "learning_rate": 6.322420315236428e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 84556560,
      "step": 5255
    },
    {
      "epoch": 0.3681735795529171,
      "grad_norm": 4.3957743644714355,
      "learning_rate": 6.321720490367775e-05,
      "loss": 1.2542,
      "num_input_tokens_seen": 84572240,
      "step": 5256
    },
    {
      "epoch": 0.36824362779864633,
      "grad_norm": 3.7909469604492188,
      "learning_rate": 6.321020665499126e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 84587400,
      "step": 5257
    },
    {
      "epoch": 0.36831367604437554,
      "grad_norm": 3.747345209121704,
      "learning_rate": 6.320320840630474e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 84603240,
      "step": 5258
    },
    {
      "epoch": 0.3683837242901048,
      "grad_norm": 3.6753249168395996,
      "learning_rate": 6.319621015761822e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 84619624,
      "step": 5259
    },
    {
      "epoch": 0.36845377253583406,
      "grad_norm": 3.6952924728393555,
      "learning_rate": 6.31892119089317e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 84636008,
      "step": 5260
    },
    {
      "epoch": 0.3685238207815633,
      "grad_norm": 4.606325149536133,
      "learning_rate": 6.318221366024518e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 84652392,
      "step": 5261
    },
    {
      "epoch": 0.3685938690272925,
      "grad_norm": 3.749755382537842,
      "learning_rate": 6.317521541155867e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 84667832,
      "step": 5262
    },
    {
      "epoch": 0.3686639172730218,
      "grad_norm": 3.7973029613494873,
      "learning_rate": 6.316821716287216e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 84683904,
      "step": 5263
    },
    {
      "epoch": 0.36873396551875104,
      "grad_norm": 4.264857769012451,
      "learning_rate": 6.316121891418565e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 84700288,
      "step": 5264
    },
    {
      "epoch": 0.3688040137644803,
      "grad_norm": 3.4577653408050537,
      "learning_rate": 6.315422066549914e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 84716672,
      "step": 5265
    },
    {
      "epoch": 0.36887406201020956,
      "grad_norm": 4.049471378326416,
      "learning_rate": 6.314722241681261e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 84732976,
      "step": 5266
    },
    {
      "epoch": 0.36894411025593876,
      "grad_norm": 4.293907165527344,
      "learning_rate": 6.314022416812609e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 84747480,
      "step": 5267
    },
    {
      "epoch": 0.369014158501668,
      "grad_norm": 7.115272045135498,
      "learning_rate": 6.313322591943959e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 84763864,
      "step": 5268
    },
    {
      "epoch": 0.3690842067473973,
      "grad_norm": 6.85962438583374,
      "learning_rate": 6.312622767075306e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 84778648,
      "step": 5269
    },
    {
      "epoch": 0.36915425499312654,
      "grad_norm": 4.24301290512085,
      "learning_rate": 6.311922942206655e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 84794440,
      "step": 5270
    },
    {
      "epoch": 0.36922430323885574,
      "grad_norm": 3.533189535140991,
      "learning_rate": 6.311223117338004e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 84810824,
      "step": 5271
    },
    {
      "epoch": 0.369294351484585,
      "grad_norm": 4.706559658050537,
      "learning_rate": 6.310523292469353e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 84827208,
      "step": 5272
    },
    {
      "epoch": 0.36936439973031426,
      "grad_norm": 3.492366075515747,
      "learning_rate": 6.3098234676007e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 84842744,
      "step": 5273
    },
    {
      "epoch": 0.3694344479760435,
      "grad_norm": 4.733495712280273,
      "learning_rate": 6.30912364273205e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 84858432,
      "step": 5274
    },
    {
      "epoch": 0.3695044962217727,
      "grad_norm": 3.6145412921905518,
      "learning_rate": 6.308423817863398e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 84874496,
      "step": 5275
    },
    {
      "epoch": 0.369574544467502,
      "grad_norm": 3.764568328857422,
      "learning_rate": 6.307723992994746e-05,
      "loss": 1.1604,
      "num_input_tokens_seen": 84890880,
      "step": 5276
    },
    {
      "epoch": 0.36964459271323125,
      "grad_norm": 5.0368428230285645,
      "learning_rate": 6.307024168126095e-05,
      "loss": 1.0006,
      "num_input_tokens_seen": 84907264,
      "step": 5277
    },
    {
      "epoch": 0.3697146409589605,
      "grad_norm": 3.9158520698547363,
      "learning_rate": 6.306324343257443e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 84923648,
      "step": 5278
    },
    {
      "epoch": 0.3697846892046897,
      "grad_norm": 4.663973808288574,
      "learning_rate": 6.305624518388792e-05,
      "loss": 0.9818,
      "num_input_tokens_seen": 84939976,
      "step": 5279
    },
    {
      "epoch": 0.36985473745041897,
      "grad_norm": 4.3741455078125,
      "learning_rate": 6.30492469352014e-05,
      "loss": 1.2902,
      "num_input_tokens_seen": 84956184,
      "step": 5280
    },
    {
      "epoch": 0.36992478569614823,
      "grad_norm": 5.071192264556885,
      "learning_rate": 6.304224868651489e-05,
      "loss": 1.0856,
      "num_input_tokens_seen": 84972024,
      "step": 5281
    },
    {
      "epoch": 0.3699948339418775,
      "grad_norm": 3.5479323863983154,
      "learning_rate": 6.303525043782838e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 84988408,
      "step": 5282
    },
    {
      "epoch": 0.3700648821876067,
      "grad_norm": 4.6933465003967285,
      "learning_rate": 6.302825218914186e-05,
      "loss": 1.1826,
      "num_input_tokens_seen": 85004720,
      "step": 5283
    },
    {
      "epoch": 0.37013493043333595,
      "grad_norm": 3.594067096710205,
      "learning_rate": 6.302125394045535e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 85020456,
      "step": 5284
    },
    {
      "epoch": 0.3702049786790652,
      "grad_norm": 3.972480535507202,
      "learning_rate": 6.301425569176884e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 85036840,
      "step": 5285
    },
    {
      "epoch": 0.37027502692479447,
      "grad_norm": 4.674763202667236,
      "learning_rate": 6.300725744308232e-05,
      "loss": 1.029,
      "num_input_tokens_seen": 85053224,
      "step": 5286
    },
    {
      "epoch": 0.3703450751705237,
      "grad_norm": 4.716235160827637,
      "learning_rate": 6.300025919439579e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 85068624,
      "step": 5287
    },
    {
      "epoch": 0.37041512341625293,
      "grad_norm": 5.01246452331543,
      "learning_rate": 6.299326094570928e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 85085008,
      "step": 5288
    },
    {
      "epoch": 0.3704851716619822,
      "grad_norm": 5.020605087280273,
      "learning_rate": 6.298626269702277e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 85101392,
      "step": 5289
    },
    {
      "epoch": 0.37055521990771145,
      "grad_norm": 5.841190814971924,
      "learning_rate": 6.297926444833626e-05,
      "loss": 1.3302,
      "num_input_tokens_seen": 85117776,
      "step": 5290
    },
    {
      "epoch": 0.37062526815344066,
      "grad_norm": 4.592007637023926,
      "learning_rate": 6.297226619964975e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 85134160,
      "step": 5291
    },
    {
      "epoch": 0.3706953163991699,
      "grad_norm": 3.678398609161377,
      "learning_rate": 6.296526795096323e-05,
      "loss": 0.9809,
      "num_input_tokens_seen": 85150544,
      "step": 5292
    },
    {
      "epoch": 0.3707653646448992,
      "grad_norm": 3.9148921966552734,
      "learning_rate": 6.295826970227671e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 85166208,
      "step": 5293
    },
    {
      "epoch": 0.37083541289062844,
      "grad_norm": 3.83375883102417,
      "learning_rate": 6.295127145359018e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 85182592,
      "step": 5294
    },
    {
      "epoch": 0.37090546113635764,
      "grad_norm": 6.339621067047119,
      "learning_rate": 6.294427320490369e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 85197512,
      "step": 5295
    },
    {
      "epoch": 0.3709755093820869,
      "grad_norm": 3.931565046310425,
      "learning_rate": 6.293727495621716e-05,
      "loss": 0.9326,
      "num_input_tokens_seen": 85213800,
      "step": 5296
    },
    {
      "epoch": 0.37104555762781616,
      "grad_norm": 4.46995210647583,
      "learning_rate": 6.293027670753065e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 85229528,
      "step": 5297
    },
    {
      "epoch": 0.3711156058735454,
      "grad_norm": 4.4390363693237305,
      "learning_rate": 6.292327845884414e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 85245912,
      "step": 5298
    },
    {
      "epoch": 0.3711856541192747,
      "grad_norm": 4.089926719665527,
      "learning_rate": 6.291628021015763e-05,
      "loss": 1.037,
      "num_input_tokens_seen": 85262296,
      "step": 5299
    },
    {
      "epoch": 0.3712557023650039,
      "grad_norm": 4.190539360046387,
      "learning_rate": 6.29092819614711e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 85278560,
      "step": 5300
    },
    {
      "epoch": 0.37132575061073314,
      "grad_norm": 5.1102166175842285,
      "learning_rate": 6.290228371278459e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 85294944,
      "step": 5301
    },
    {
      "epoch": 0.3713957988564624,
      "grad_norm": 4.174960136413574,
      "learning_rate": 6.289528546409808e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 85311328,
      "step": 5302
    },
    {
      "epoch": 0.37146584710219166,
      "grad_norm": 3.8785698413848877,
      "learning_rate": 6.288828721541157e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 85326784,
      "step": 5303
    },
    {
      "epoch": 0.37153589534792086,
      "grad_norm": 3.728626251220703,
      "learning_rate": 6.288128896672504e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 85343168,
      "step": 5304
    },
    {
      "epoch": 0.3716059435936501,
      "grad_norm": 5.1877312660217285,
      "learning_rate": 6.287429071803853e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 85359552,
      "step": 5305
    },
    {
      "epoch": 0.3716759918393794,
      "grad_norm": 5.751648902893066,
      "learning_rate": 6.286729246935202e-05,
      "loss": 1.2662,
      "num_input_tokens_seen": 85375136,
      "step": 5306
    },
    {
      "epoch": 0.37174604008510864,
      "grad_norm": 3.7917258739471436,
      "learning_rate": 6.28602942206655e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 85391520,
      "step": 5307
    },
    {
      "epoch": 0.37181608833083785,
      "grad_norm": 4.268946647644043,
      "learning_rate": 6.285329597197898e-05,
      "loss": 1.0928,
      "num_input_tokens_seen": 85406848,
      "step": 5308
    },
    {
      "epoch": 0.3718861365765671,
      "grad_norm": 4.350981712341309,
      "learning_rate": 6.284629772329247e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 85423232,
      "step": 5309
    },
    {
      "epoch": 0.37195618482229637,
      "grad_norm": 3.8072032928466797,
      "learning_rate": 6.283929947460596e-05,
      "loss": 0.9999,
      "num_input_tokens_seen": 85439616,
      "step": 5310
    },
    {
      "epoch": 0.3720262330680256,
      "grad_norm": 4.0531697273254395,
      "learning_rate": 6.283230122591945e-05,
      "loss": 0.9389,
      "num_input_tokens_seen": 85456000,
      "step": 5311
    },
    {
      "epoch": 0.37209628131375483,
      "grad_norm": 5.18675422668457,
      "learning_rate": 6.282530297723294e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 85472384,
      "step": 5312
    },
    {
      "epoch": 0.3721663295594841,
      "grad_norm": 4.675386428833008,
      "learning_rate": 6.281830472854641e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 85488544,
      "step": 5313
    },
    {
      "epoch": 0.37223637780521335,
      "grad_norm": 6.522333145141602,
      "learning_rate": 6.281130647985989e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 85504352,
      "step": 5314
    },
    {
      "epoch": 0.3723064260509426,
      "grad_norm": 3.9266233444213867,
      "learning_rate": 6.280430823117338e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 85520688,
      "step": 5315
    },
    {
      "epoch": 0.3723764742966718,
      "grad_norm": 4.6428093910217285,
      "learning_rate": 6.279730998248687e-05,
      "loss": 0.9396,
      "num_input_tokens_seen": 85537072,
      "step": 5316
    },
    {
      "epoch": 0.37244652254240107,
      "grad_norm": 3.6043691635131836,
      "learning_rate": 6.279031173380035e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 85553456,
      "step": 5317
    },
    {
      "epoch": 0.37251657078813033,
      "grad_norm": 3.4878151416778564,
      "learning_rate": 6.278331348511384e-05,
      "loss": 1.101,
      "num_input_tokens_seen": 85569824,
      "step": 5318
    },
    {
      "epoch": 0.3725866190338596,
      "grad_norm": 4.275106906890869,
      "learning_rate": 6.277631523642733e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 85586208,
      "step": 5319
    },
    {
      "epoch": 0.3726566672795888,
      "grad_norm": 7.615388870239258,
      "learning_rate": 6.27693169877408e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 85600984,
      "step": 5320
    },
    {
      "epoch": 0.37272671552531805,
      "grad_norm": 4.4750752449035645,
      "learning_rate": 6.276231873905428e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 85617368,
      "step": 5321
    },
    {
      "epoch": 0.3727967637710473,
      "grad_norm": 3.7900373935699463,
      "learning_rate": 6.275532049036778e-05,
      "loss": 1.0727,
      "num_input_tokens_seen": 85633304,
      "step": 5322
    },
    {
      "epoch": 0.37286681201677657,
      "grad_norm": 8.58016300201416,
      "learning_rate": 6.274832224168127e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 85648592,
      "step": 5323
    },
    {
      "epoch": 0.3729368602625058,
      "grad_norm": 3.847476005554199,
      "learning_rate": 6.274132399299475e-05,
      "loss": 1.2543,
      "num_input_tokens_seen": 85664976,
      "step": 5324
    },
    {
      "epoch": 0.37300690850823504,
      "grad_norm": 3.68683123588562,
      "learning_rate": 6.273432574430824e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 85681360,
      "step": 5325
    },
    {
      "epoch": 0.3730769567539643,
      "grad_norm": 4.07316255569458,
      "learning_rate": 6.272732749562172e-05,
      "loss": 1.1859,
      "num_input_tokens_seen": 85697744,
      "step": 5326
    },
    {
      "epoch": 0.37314700499969355,
      "grad_norm": 3.7817749977111816,
      "learning_rate": 6.27203292469352e-05,
      "loss": 1.128,
      "num_input_tokens_seen": 85713680,
      "step": 5327
    },
    {
      "epoch": 0.37321705324542276,
      "grad_norm": 3.8322465419769287,
      "learning_rate": 6.271333099824869e-05,
      "loss": 1.1804,
      "num_input_tokens_seen": 85730064,
      "step": 5328
    },
    {
      "epoch": 0.373287101491152,
      "grad_norm": 5.689653396606445,
      "learning_rate": 6.270633274956218e-05,
      "loss": 1.0848,
      "num_input_tokens_seen": 85745904,
      "step": 5329
    },
    {
      "epoch": 0.3733571497368813,
      "grad_norm": 5.568809509277344,
      "learning_rate": 6.269933450087566e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 85762288,
      "step": 5330
    },
    {
      "epoch": 0.37342719798261054,
      "grad_norm": 3.982375383377075,
      "learning_rate": 6.269233625218914e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 85778672,
      "step": 5331
    },
    {
      "epoch": 0.3734972462283398,
      "grad_norm": 3.430204391479492,
      "learning_rate": 6.268533800350263e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 85795056,
      "step": 5332
    },
    {
      "epoch": 0.373567294474069,
      "grad_norm": 3.465724229812622,
      "learning_rate": 6.267833975481612e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 85811392,
      "step": 5333
    },
    {
      "epoch": 0.37363734271979826,
      "grad_norm": 3.837188482284546,
      "learning_rate": 6.267134150612959e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 85827016,
      "step": 5334
    },
    {
      "epoch": 0.3737073909655275,
      "grad_norm": 7.360764980316162,
      "learning_rate": 6.266434325744308e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 85842040,
      "step": 5335
    },
    {
      "epoch": 0.3737774392112568,
      "grad_norm": 3.567553997039795,
      "learning_rate": 6.265734500875657e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 85858424,
      "step": 5336
    },
    {
      "epoch": 0.373847487456986,
      "grad_norm": 4.564986705780029,
      "learning_rate": 6.265034676007006e-05,
      "loss": 1.0178,
      "num_input_tokens_seen": 85874808,
      "step": 5337
    },
    {
      "epoch": 0.37391753570271524,
      "grad_norm": 3.4568405151367188,
      "learning_rate": 6.264334851138355e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 85890672,
      "step": 5338
    },
    {
      "epoch": 0.3739875839484445,
      "grad_norm": 3.723557233810425,
      "learning_rate": 6.263635026269704e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 85906920,
      "step": 5339
    },
    {
      "epoch": 0.37405763219417376,
      "grad_norm": 3.5800676345825195,
      "learning_rate": 6.262935201401051e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 85923304,
      "step": 5340
    },
    {
      "epoch": 0.37412768043990297,
      "grad_norm": 3.8996667861938477,
      "learning_rate": 6.262235376532399e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 85938984,
      "step": 5341
    },
    {
      "epoch": 0.3741977286856322,
      "grad_norm": 3.417182207107544,
      "learning_rate": 6.261535551663747e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 85955368,
      "step": 5342
    },
    {
      "epoch": 0.3742677769313615,
      "grad_norm": 4.214803695678711,
      "learning_rate": 6.260835726795098e-05,
      "loss": 1.1107,
      "num_input_tokens_seen": 85971320,
      "step": 5343
    },
    {
      "epoch": 0.37433782517709074,
      "grad_norm": 3.7782840728759766,
      "learning_rate": 6.260135901926445e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 85987704,
      "step": 5344
    },
    {
      "epoch": 0.37440787342281995,
      "grad_norm": 3.6186842918395996,
      "learning_rate": 6.259436077057794e-05,
      "loss": 1.0682,
      "num_input_tokens_seen": 86004088,
      "step": 5345
    },
    {
      "epoch": 0.3744779216685492,
      "grad_norm": 4.2028913497924805,
      "learning_rate": 6.258736252189143e-05,
      "loss": 1.2203,
      "num_input_tokens_seen": 86020472,
      "step": 5346
    },
    {
      "epoch": 0.37454796991427847,
      "grad_norm": 4.17422342300415,
      "learning_rate": 6.25803642732049e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 86036856,
      "step": 5347
    },
    {
      "epoch": 0.3746180181600077,
      "grad_norm": 3.3578243255615234,
      "learning_rate": 6.257336602451838e-05,
      "loss": 1.0315,
      "num_input_tokens_seen": 86053224,
      "step": 5348
    },
    {
      "epoch": 0.37468806640573693,
      "grad_norm": 4.105921268463135,
      "learning_rate": 6.256636777583188e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 86069272,
      "step": 5349
    },
    {
      "epoch": 0.3747581146514662,
      "grad_norm": 3.7420692443847656,
      "learning_rate": 6.255936952714537e-05,
      "loss": 1.0672,
      "num_input_tokens_seen": 86085656,
      "step": 5350
    },
    {
      "epoch": 0.37482816289719545,
      "grad_norm": 5.1573872566223145,
      "learning_rate": 6.255237127845884e-05,
      "loss": 1.376,
      "num_input_tokens_seen": 86102040,
      "step": 5351
    },
    {
      "epoch": 0.3748982111429247,
      "grad_norm": 3.9844436645507812,
      "learning_rate": 6.254537302977233e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 86117976,
      "step": 5352
    },
    {
      "epoch": 0.3749682593886539,
      "grad_norm": 3.6582653522491455,
      "learning_rate": 6.253837478108582e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 86134360,
      "step": 5353
    },
    {
      "epoch": 0.3750383076343832,
      "grad_norm": 4.814766883850098,
      "learning_rate": 6.25313765323993e-05,
      "loss": 1.2574,
      "num_input_tokens_seen": 86150208,
      "step": 5354
    },
    {
      "epoch": 0.37510835588011243,
      "grad_norm": 4.7514262199401855,
      "learning_rate": 6.252437828371278e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 86165672,
      "step": 5355
    },
    {
      "epoch": 0.3751784041258417,
      "grad_norm": 3.9450578689575195,
      "learning_rate": 6.251738003502627e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 86182056,
      "step": 5356
    },
    {
      "epoch": 0.3752484523715709,
      "grad_norm": 3.5215647220611572,
      "learning_rate": 6.251038178633976e-05,
      "loss": 1.04,
      "num_input_tokens_seen": 86198440,
      "step": 5357
    },
    {
      "epoch": 0.37531850061730015,
      "grad_norm": 3.805070161819458,
      "learning_rate": 6.250338353765324e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 86214824,
      "step": 5358
    },
    {
      "epoch": 0.3753885488630294,
      "grad_norm": 4.033730983734131,
      "learning_rate": 6.249638528896673e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 86231208,
      "step": 5359
    },
    {
      "epoch": 0.3754585971087587,
      "grad_norm": 3.8157355785369873,
      "learning_rate": 6.248938704028021e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 86247392,
      "step": 5360
    },
    {
      "epoch": 0.3755286453544879,
      "grad_norm": 4.832013130187988,
      "learning_rate": 6.248238879159369e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 86263776,
      "step": 5361
    },
    {
      "epoch": 0.37559869360021714,
      "grad_norm": 3.753471612930298,
      "learning_rate": 6.247539054290718e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 86279912,
      "step": 5362
    },
    {
      "epoch": 0.3756687418459464,
      "grad_norm": 8.569518089294434,
      "learning_rate": 6.246839229422068e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 86296296,
      "step": 5363
    },
    {
      "epoch": 0.37573879009167566,
      "grad_norm": 4.399802207946777,
      "learning_rate": 6.246139404553416e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 86312680,
      "step": 5364
    },
    {
      "epoch": 0.37580883833740486,
      "grad_norm": 4.230834484100342,
      "learning_rate": 6.245439579684764e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 86329064,
      "step": 5365
    },
    {
      "epoch": 0.3758788865831341,
      "grad_norm": 4.750765800476074,
      "learning_rate": 6.244739754816113e-05,
      "loss": 1.2126,
      "num_input_tokens_seen": 86345448,
      "step": 5366
    },
    {
      "epoch": 0.3759489348288634,
      "grad_norm": 6.567142963409424,
      "learning_rate": 6.244039929947461e-05,
      "loss": 1.314,
      "num_input_tokens_seen": 86361272,
      "step": 5367
    },
    {
      "epoch": 0.37601898307459264,
      "grad_norm": 3.9668781757354736,
      "learning_rate": 6.243340105078808e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 86377448,
      "step": 5368
    },
    {
      "epoch": 0.3760890313203219,
      "grad_norm": 4.619864463806152,
      "learning_rate": 6.242640280210158e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 86393600,
      "step": 5369
    },
    {
      "epoch": 0.3761590795660511,
      "grad_norm": 6.837228298187256,
      "learning_rate": 6.241940455341507e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 86409896,
      "step": 5370
    },
    {
      "epoch": 0.37622912781178036,
      "grad_norm": 4.634070873260498,
      "learning_rate": 6.241240630472855e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 86426280,
      "step": 5371
    },
    {
      "epoch": 0.3762991760575096,
      "grad_norm": 3.944580554962158,
      "learning_rate": 6.240540805604204e-05,
      "loss": 1.144,
      "num_input_tokens_seen": 86442640,
      "step": 5372
    },
    {
      "epoch": 0.3763692243032389,
      "grad_norm": 7.016427516937256,
      "learning_rate": 6.239840980735553e-05,
      "loss": 1.0016,
      "num_input_tokens_seen": 86459024,
      "step": 5373
    },
    {
      "epoch": 0.3764392725489681,
      "grad_norm": 3.9997384548187256,
      "learning_rate": 6.2391411558669e-05,
      "loss": 0.9382,
      "num_input_tokens_seen": 86475408,
      "step": 5374
    },
    {
      "epoch": 0.37650932079469734,
      "grad_norm": 4.016181945800781,
      "learning_rate": 6.238441330998249e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 86491680,
      "step": 5375
    },
    {
      "epoch": 0.3765793690404266,
      "grad_norm": 4.19748592376709,
      "learning_rate": 6.237741506129598e-05,
      "loss": 1.161,
      "num_input_tokens_seen": 86507768,
      "step": 5376
    },
    {
      "epoch": 0.37664941728615586,
      "grad_norm": 4.579540252685547,
      "learning_rate": 6.237041681260947e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 86524040,
      "step": 5377
    },
    {
      "epoch": 0.37671946553188507,
      "grad_norm": 3.784952402114868,
      "learning_rate": 6.236341856392294e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 86540424,
      "step": 5378
    },
    {
      "epoch": 0.3767895137776143,
      "grad_norm": 5.813356876373291,
      "learning_rate": 6.235642031523643e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 86556360,
      "step": 5379
    },
    {
      "epoch": 0.3768595620233436,
      "grad_norm": 4.314088344573975,
      "learning_rate": 6.234942206654992e-05,
      "loss": 1.2318,
      "num_input_tokens_seen": 86572744,
      "step": 5380
    },
    {
      "epoch": 0.37692961026907285,
      "grad_norm": 3.898298740386963,
      "learning_rate": 6.23424238178634e-05,
      "loss": 1.1217,
      "num_input_tokens_seen": 86588888,
      "step": 5381
    },
    {
      "epoch": 0.37699965851480205,
      "grad_norm": 3.514692544937134,
      "learning_rate": 6.233542556917688e-05,
      "loss": 0.9526,
      "num_input_tokens_seen": 86605272,
      "step": 5382
    },
    {
      "epoch": 0.3770697067605313,
      "grad_norm": 3.7073886394500732,
      "learning_rate": 6.232842732049038e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 86621656,
      "step": 5383
    },
    {
      "epoch": 0.37713975500626057,
      "grad_norm": 3.9826815128326416,
      "learning_rate": 6.232142907180386e-05,
      "loss": 1.1417,
      "num_input_tokens_seen": 86638040,
      "step": 5384
    },
    {
      "epoch": 0.37720980325198983,
      "grad_norm": 3.6563196182250977,
      "learning_rate": 6.231443082311733e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 86654424,
      "step": 5385
    },
    {
      "epoch": 0.37727985149771903,
      "grad_norm": 3.5995571613311768,
      "learning_rate": 6.230743257443082e-05,
      "loss": 1.0457,
      "num_input_tokens_seen": 86670328,
      "step": 5386
    },
    {
      "epoch": 0.3773498997434483,
      "grad_norm": 4.254338264465332,
      "learning_rate": 6.230043432574431e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 86685960,
      "step": 5387
    },
    {
      "epoch": 0.37741994798917755,
      "grad_norm": 3.689716100692749,
      "learning_rate": 6.229343607705779e-05,
      "loss": 0.81,
      "num_input_tokens_seen": 86702008,
      "step": 5388
    },
    {
      "epoch": 0.3774899962349068,
      "grad_norm": 3.4042210578918457,
      "learning_rate": 6.228643782837129e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 86718392,
      "step": 5389
    },
    {
      "epoch": 0.377560044480636,
      "grad_norm": 4.607806205749512,
      "learning_rate": 6.227943957968478e-05,
      "loss": 1.2891,
      "num_input_tokens_seen": 86734624,
      "step": 5390
    },
    {
      "epoch": 0.3776300927263653,
      "grad_norm": 3.951362133026123,
      "learning_rate": 6.227244133099825e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 86749816,
      "step": 5391
    },
    {
      "epoch": 0.37770014097209453,
      "grad_norm": 3.535480260848999,
      "learning_rate": 6.226544308231174e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 86765800,
      "step": 5392
    },
    {
      "epoch": 0.3777701892178238,
      "grad_norm": 5.398930549621582,
      "learning_rate": 6.225844483362523e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 86782184,
      "step": 5393
    },
    {
      "epoch": 0.377840237463553,
      "grad_norm": 4.456240177154541,
      "learning_rate": 6.22514465849387e-05,
      "loss": 1.1725,
      "num_input_tokens_seen": 86798568,
      "step": 5394
    },
    {
      "epoch": 0.37791028570928226,
      "grad_norm": 3.8764703273773193,
      "learning_rate": 6.224444833625219e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 86814824,
      "step": 5395
    },
    {
      "epoch": 0.3779803339550115,
      "grad_norm": 3.8746144771575928,
      "learning_rate": 6.223745008756568e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 86831208,
      "step": 5396
    },
    {
      "epoch": 0.3780503822007408,
      "grad_norm": 4.3454742431640625,
      "learning_rate": 6.223045183887917e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 86846872,
      "step": 5397
    },
    {
      "epoch": 0.37812043044647,
      "grad_norm": 5.006749153137207,
      "learning_rate": 6.222345359019265e-05,
      "loss": 0.8317,
      "num_input_tokens_seen": 86863256,
      "step": 5398
    },
    {
      "epoch": 0.37819047869219924,
      "grad_norm": 3.7388808727264404,
      "learning_rate": 6.221645534150613e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 86879640,
      "step": 5399
    },
    {
      "epoch": 0.3782605269379285,
      "grad_norm": 4.515074253082275,
      "learning_rate": 6.220945709281962e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 86896024,
      "step": 5400
    },
    {
      "epoch": 0.3782605269379285,
      "eval_loss": 1.1279726028442383,
      "eval_runtime": 0.2024,
      "eval_samples_per_second": 4.94,
      "eval_steps_per_second": 4.94,
      "num_input_tokens_seen": 86896024,
      "step": 5400
    },
    {
      "epoch": 0.37833057518365776,
      "grad_norm": 3.5468356609344482,
      "learning_rate": 6.22024588441331e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 86912032,
      "step": 5401
    },
    {
      "epoch": 0.378400623429387,
      "grad_norm": 4.281546115875244,
      "learning_rate": 6.219546059544659e-05,
      "loss": 1.2335,
      "num_input_tokens_seen": 86928080,
      "step": 5402
    },
    {
      "epoch": 0.3784706716751162,
      "grad_norm": 4.247570037841797,
      "learning_rate": 6.218846234676009e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 86944424,
      "step": 5403
    },
    {
      "epoch": 0.3785407199208455,
      "grad_norm": 3.78439998626709,
      "learning_rate": 6.218146409807356e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 86960808,
      "step": 5404
    },
    {
      "epoch": 0.37861076816657474,
      "grad_norm": 4.174613952636719,
      "learning_rate": 6.217446584938704e-05,
      "loss": 1.2858,
      "num_input_tokens_seen": 86976472,
      "step": 5405
    },
    {
      "epoch": 0.378680816412304,
      "grad_norm": 4.759533882141113,
      "learning_rate": 6.216746760070053e-05,
      "loss": 0.9813,
      "num_input_tokens_seen": 86992856,
      "step": 5406
    },
    {
      "epoch": 0.3787508646580332,
      "grad_norm": 5.2616801261901855,
      "learning_rate": 6.216046935201402e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 87007936,
      "step": 5407
    },
    {
      "epoch": 0.37882091290376246,
      "grad_norm": 4.626899719238281,
      "learning_rate": 6.215347110332749e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 87023888,
      "step": 5408
    },
    {
      "epoch": 0.3788909611494917,
      "grad_norm": 3.7142221927642822,
      "learning_rate": 6.214647285464099e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 87040272,
      "step": 5409
    },
    {
      "epoch": 0.378961009395221,
      "grad_norm": 6.228342056274414,
      "learning_rate": 6.213947460595448e-05,
      "loss": 1.0807,
      "num_input_tokens_seen": 87056656,
      "step": 5410
    },
    {
      "epoch": 0.3790310576409502,
      "grad_norm": 3.7979259490966797,
      "learning_rate": 6.213247635726796e-05,
      "loss": 1.0051,
      "num_input_tokens_seen": 87073040,
      "step": 5411
    },
    {
      "epoch": 0.37910110588667945,
      "grad_norm": 3.903106927871704,
      "learning_rate": 6.212547810858143e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 87089344,
      "step": 5412
    },
    {
      "epoch": 0.3791711541324087,
      "grad_norm": 3.966651201248169,
      "learning_rate": 6.211847985989492e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 87105144,
      "step": 5413
    },
    {
      "epoch": 0.37924120237813796,
      "grad_norm": 4.070274829864502,
      "learning_rate": 6.211148161120841e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 87121528,
      "step": 5414
    },
    {
      "epoch": 0.37931125062386717,
      "grad_norm": 3.516997814178467,
      "learning_rate": 6.21044833625219e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 87137752,
      "step": 5415
    },
    {
      "epoch": 0.37938129886959643,
      "grad_norm": 4.28290319442749,
      "learning_rate": 6.209748511383539e-05,
      "loss": 1.2252,
      "num_input_tokens_seen": 87154136,
      "step": 5416
    },
    {
      "epoch": 0.3794513471153257,
      "grad_norm": 4.765808582305908,
      "learning_rate": 6.209048686514887e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 87170520,
      "step": 5417
    },
    {
      "epoch": 0.37952139536105495,
      "grad_norm": 3.8507494926452637,
      "learning_rate": 6.208348861646235e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 87186904,
      "step": 5418
    },
    {
      "epoch": 0.37959144360678415,
      "grad_norm": 7.46950626373291,
      "learning_rate": 6.207649036777584e-05,
      "loss": 1.1376,
      "num_input_tokens_seen": 87203288,
      "step": 5419
    },
    {
      "epoch": 0.3796614918525134,
      "grad_norm": 5.770944595336914,
      "learning_rate": 6.206949211908933e-05,
      "loss": 1.23,
      "num_input_tokens_seen": 87219552,
      "step": 5420
    },
    {
      "epoch": 0.37973154009824267,
      "grad_norm": 3.752936363220215,
      "learning_rate": 6.20624938704028e-05,
      "loss": 0.8285,
      "num_input_tokens_seen": 87235736,
      "step": 5421
    },
    {
      "epoch": 0.37980158834397193,
      "grad_norm": 3.8336403369903564,
      "learning_rate": 6.205549562171629e-05,
      "loss": 0.8416,
      "num_input_tokens_seen": 87252120,
      "step": 5422
    },
    {
      "epoch": 0.37987163658970113,
      "grad_norm": 7.380855083465576,
      "learning_rate": 6.204849737302978e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 87268504,
      "step": 5423
    },
    {
      "epoch": 0.3799416848354304,
      "grad_norm": 4.780874729156494,
      "learning_rate": 6.204149912434327e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 87284888,
      "step": 5424
    },
    {
      "epoch": 0.38001173308115965,
      "grad_norm": 4.691160202026367,
      "learning_rate": 6.203450087565674e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 87301272,
      "step": 5425
    },
    {
      "epoch": 0.3800817813268889,
      "grad_norm": 3.592348098754883,
      "learning_rate": 6.202750262697023e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 87317288,
      "step": 5426
    },
    {
      "epoch": 0.3801518295726181,
      "grad_norm": 4.750811576843262,
      "learning_rate": 6.202050437828372e-05,
      "loss": 1.2781,
      "num_input_tokens_seen": 87332488,
      "step": 5427
    },
    {
      "epoch": 0.3802218778183474,
      "grad_norm": 4.564239501953125,
      "learning_rate": 6.20135061295972e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 87348264,
      "step": 5428
    },
    {
      "epoch": 0.38029192606407664,
      "grad_norm": 4.697380065917969,
      "learning_rate": 6.200650788091068e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 87363920,
      "step": 5429
    },
    {
      "epoch": 0.3803619743098059,
      "grad_norm": 4.026552677154541,
      "learning_rate": 6.199950963222419e-05,
      "loss": 1.2177,
      "num_input_tokens_seen": 87379920,
      "step": 5430
    },
    {
      "epoch": 0.3804320225555351,
      "grad_norm": 5.023289203643799,
      "learning_rate": 6.199251138353766e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 87395848,
      "step": 5431
    },
    {
      "epoch": 0.38050207080126436,
      "grad_norm": 3.6005523204803467,
      "learning_rate": 6.198551313485114e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 87411632,
      "step": 5432
    },
    {
      "epoch": 0.3805721190469936,
      "grad_norm": 4.094357967376709,
      "learning_rate": 6.197851488616462e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 87426912,
      "step": 5433
    },
    {
      "epoch": 0.3806421672927229,
      "grad_norm": 4.452909469604492,
      "learning_rate": 6.197151663747811e-05,
      "loss": 1.248,
      "num_input_tokens_seen": 87443296,
      "step": 5434
    },
    {
      "epoch": 0.3807122155384521,
      "grad_norm": 3.975532054901123,
      "learning_rate": 6.19645183887916e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 87459680,
      "step": 5435
    },
    {
      "epoch": 0.38078226378418134,
      "grad_norm": 4.745920181274414,
      "learning_rate": 6.195752014010509e-05,
      "loss": 1.2534,
      "num_input_tokens_seen": 87476064,
      "step": 5436
    },
    {
      "epoch": 0.3808523120299106,
      "grad_norm": 3.8793790340423584,
      "learning_rate": 6.195052189141858e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 87492448,
      "step": 5437
    },
    {
      "epoch": 0.38092236027563986,
      "grad_norm": 4.695518493652344,
      "learning_rate": 6.194352364273205e-05,
      "loss": 0.93,
      "num_input_tokens_seen": 87508832,
      "step": 5438
    },
    {
      "epoch": 0.3809924085213691,
      "grad_norm": 3.5820047855377197,
      "learning_rate": 6.193652539404553e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 87524728,
      "step": 5439
    },
    {
      "epoch": 0.3810624567670983,
      "grad_norm": 5.76292610168457,
      "learning_rate": 6.192952714535902e-05,
      "loss": 1.1919,
      "num_input_tokens_seen": 87540752,
      "step": 5440
    },
    {
      "epoch": 0.3811325050128276,
      "grad_norm": 4.334653377532959,
      "learning_rate": 6.19225288966725e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 87556384,
      "step": 5441
    },
    {
      "epoch": 0.38120255325855684,
      "grad_norm": 3.7438180446624756,
      "learning_rate": 6.1915530647986e-05,
      "loss": 1.0083,
      "num_input_tokens_seen": 87572320,
      "step": 5442
    },
    {
      "epoch": 0.3812726015042861,
      "grad_norm": 4.082560062408447,
      "learning_rate": 6.190853239929948e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 87588704,
      "step": 5443
    },
    {
      "epoch": 0.3813426497500153,
      "grad_norm": 3.9324755668640137,
      "learning_rate": 6.190153415061297e-05,
      "loss": 1.0991,
      "num_input_tokens_seen": 87605088,
      "step": 5444
    },
    {
      "epoch": 0.38141269799574457,
      "grad_norm": 5.329967498779297,
      "learning_rate": 6.189453590192645e-05,
      "loss": 1.1817,
      "num_input_tokens_seen": 87621472,
      "step": 5445
    },
    {
      "epoch": 0.3814827462414738,
      "grad_norm": 3.627267837524414,
      "learning_rate": 6.188753765323993e-05,
      "loss": 0.9407,
      "num_input_tokens_seen": 87637856,
      "step": 5446
    },
    {
      "epoch": 0.3815527944872031,
      "grad_norm": 3.6728835105895996,
      "learning_rate": 6.188053940455342e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 87653720,
      "step": 5447
    },
    {
      "epoch": 0.3816228427329323,
      "grad_norm": 3.556185245513916,
      "learning_rate": 6.18735411558669e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 87670104,
      "step": 5448
    },
    {
      "epoch": 0.38169289097866155,
      "grad_norm": 4.075231552124023,
      "learning_rate": 6.186654290718039e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 87686488,
      "step": 5449
    },
    {
      "epoch": 0.3817629392243908,
      "grad_norm": 3.981752395629883,
      "learning_rate": 6.185954465849388e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 87702872,
      "step": 5450
    },
    {
      "epoch": 0.38183298747012007,
      "grad_norm": 4.75683069229126,
      "learning_rate": 6.185254640980736e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 87718912,
      "step": 5451
    },
    {
      "epoch": 0.38190303571584927,
      "grad_norm": 6.081716060638428,
      "learning_rate": 6.184554816112084e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 87735160,
      "step": 5452
    },
    {
      "epoch": 0.38197308396157853,
      "grad_norm": 6.651247978210449,
      "learning_rate": 6.183854991243433e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 87749232,
      "step": 5453
    },
    {
      "epoch": 0.3820431322073078,
      "grad_norm": 4.12028694152832,
      "learning_rate": 6.183155166374782e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 87765328,
      "step": 5454
    },
    {
      "epoch": 0.38211318045303705,
      "grad_norm": 6.3344645500183105,
      "learning_rate": 6.18245534150613e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 87781712,
      "step": 5455
    },
    {
      "epoch": 0.38218322869876625,
      "grad_norm": 3.745476007461548,
      "learning_rate": 6.181755516637478e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 87798032,
      "step": 5456
    },
    {
      "epoch": 0.3822532769444955,
      "grad_norm": 3.515174388885498,
      "learning_rate": 6.181055691768828e-05,
      "loss": 1.1138,
      "num_input_tokens_seen": 87814416,
      "step": 5457
    },
    {
      "epoch": 0.38232332519022477,
      "grad_norm": 4.101998329162598,
      "learning_rate": 6.180355866900176e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 87830504,
      "step": 5458
    },
    {
      "epoch": 0.38239337343595403,
      "grad_norm": 4.045940399169922,
      "learning_rate": 6.179656042031523e-05,
      "loss": 1.1278,
      "num_input_tokens_seen": 87846264,
      "step": 5459
    },
    {
      "epoch": 0.38246342168168324,
      "grad_norm": 8.09753131866455,
      "learning_rate": 6.178956217162872e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 87861856,
      "step": 5460
    },
    {
      "epoch": 0.3825334699274125,
      "grad_norm": 5.395979404449463,
      "learning_rate": 6.178256392294221e-05,
      "loss": 1.0364,
      "num_input_tokens_seen": 87878024,
      "step": 5461
    },
    {
      "epoch": 0.38260351817314175,
      "grad_norm": 3.452855110168457,
      "learning_rate": 6.17755656742557e-05,
      "loss": 0.8875,
      "num_input_tokens_seen": 87894408,
      "step": 5462
    },
    {
      "epoch": 0.382673566418871,
      "grad_norm": 3.9877512454986572,
      "learning_rate": 6.176856742556919e-05,
      "loss": 1.0349,
      "num_input_tokens_seen": 87910400,
      "step": 5463
    },
    {
      "epoch": 0.3827436146646002,
      "grad_norm": 3.9095492362976074,
      "learning_rate": 6.176156917688268e-05,
      "loss": 1.063,
      "num_input_tokens_seen": 87926040,
      "step": 5464
    },
    {
      "epoch": 0.3828136629103295,
      "grad_norm": 4.558162212371826,
      "learning_rate": 6.175457092819615e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 87942424,
      "step": 5465
    },
    {
      "epoch": 0.38288371115605874,
      "grad_norm": 3.786123275756836,
      "learning_rate": 6.174757267950963e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 87958808,
      "step": 5466
    },
    {
      "epoch": 0.382953759401788,
      "grad_norm": 4.0291056632995605,
      "learning_rate": 6.174057443082311e-05,
      "loss": 1.0603,
      "num_input_tokens_seen": 87975192,
      "step": 5467
    },
    {
      "epoch": 0.3830238076475172,
      "grad_norm": 3.698666572570801,
      "learning_rate": 6.17335761821366e-05,
      "loss": 0.9187,
      "num_input_tokens_seen": 87991504,
      "step": 5468
    },
    {
      "epoch": 0.38309385589324646,
      "grad_norm": 3.7802882194519043,
      "learning_rate": 6.172657793345009e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 88007888,
      "step": 5469
    },
    {
      "epoch": 0.3831639041389757,
      "grad_norm": 4.754447937011719,
      "learning_rate": 6.171957968476358e-05,
      "loss": 1.1788,
      "num_input_tokens_seen": 88023832,
      "step": 5470
    },
    {
      "epoch": 0.383233952384705,
      "grad_norm": 3.502560615539551,
      "learning_rate": 6.171258143607707e-05,
      "loss": 0.9993,
      "num_input_tokens_seen": 88040216,
      "step": 5471
    },
    {
      "epoch": 0.38330400063043424,
      "grad_norm": 4.379989147186279,
      "learning_rate": 6.170558318739054e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 88055768,
      "step": 5472
    },
    {
      "epoch": 0.38337404887616344,
      "grad_norm": 3.3798177242279053,
      "learning_rate": 6.169858493870403e-05,
      "loss": 0.6884,
      "num_input_tokens_seen": 88072152,
      "step": 5473
    },
    {
      "epoch": 0.3834440971218927,
      "grad_norm": 4.265483856201172,
      "learning_rate": 6.169158669001752e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 88087816,
      "step": 5474
    },
    {
      "epoch": 0.38351414536762196,
      "grad_norm": 4.468397617340088,
      "learning_rate": 6.168458844133101e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 88103160,
      "step": 5475
    },
    {
      "epoch": 0.3835841936133512,
      "grad_norm": 3.888359546661377,
      "learning_rate": 6.167759019264448e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 88119544,
      "step": 5476
    },
    {
      "epoch": 0.3836542418590804,
      "grad_norm": 3.7953927516937256,
      "learning_rate": 6.167059194395797e-05,
      "loss": 1.1585,
      "num_input_tokens_seen": 88135928,
      "step": 5477
    },
    {
      "epoch": 0.3837242901048097,
      "grad_norm": 3.7742021083831787,
      "learning_rate": 6.166359369527146e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 88151928,
      "step": 5478
    },
    {
      "epoch": 0.38379433835053894,
      "grad_norm": 3.811535120010376,
      "learning_rate": 6.165659544658494e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 88168312,
      "step": 5479
    },
    {
      "epoch": 0.3838643865962682,
      "grad_norm": 5.1758646965026855,
      "learning_rate": 6.164959719789842e-05,
      "loss": 1.2087,
      "num_input_tokens_seen": 88184064,
      "step": 5480
    },
    {
      "epoch": 0.3839344348419974,
      "grad_norm": 4.529813289642334,
      "learning_rate": 6.164259894921191e-05,
      "loss": 0.9142,
      "num_input_tokens_seen": 88200216,
      "step": 5481
    },
    {
      "epoch": 0.38400448308772667,
      "grad_norm": 4.426999568939209,
      "learning_rate": 6.16356007005254e-05,
      "loss": 1.0046,
      "num_input_tokens_seen": 88215568,
      "step": 5482
    },
    {
      "epoch": 0.3840745313334559,
      "grad_norm": 4.73276948928833,
      "learning_rate": 6.162860245183888e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 88231952,
      "step": 5483
    },
    {
      "epoch": 0.3841445795791852,
      "grad_norm": 3.6280384063720703,
      "learning_rate": 6.162160420315238e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 88247728,
      "step": 5484
    },
    {
      "epoch": 0.3842146278249144,
      "grad_norm": 3.6947717666625977,
      "learning_rate": 6.161460595446585e-05,
      "loss": 0.9671,
      "num_input_tokens_seen": 88263472,
      "step": 5485
    },
    {
      "epoch": 0.38428467607064365,
      "grad_norm": 3.683591842651367,
      "learning_rate": 6.160760770577933e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 88279856,
      "step": 5486
    },
    {
      "epoch": 0.3843547243163729,
      "grad_norm": 6.020013332366943,
      "learning_rate": 6.160060945709282e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 88295864,
      "step": 5487
    },
    {
      "epoch": 0.38442477256210217,
      "grad_norm": 8.429437637329102,
      "learning_rate": 6.15936112084063e-05,
      "loss": 1.143,
      "num_input_tokens_seen": 88311752,
      "step": 5488
    },
    {
      "epoch": 0.3844948208078314,
      "grad_norm": 3.679159164428711,
      "learning_rate": 6.15866129597198e-05,
      "loss": 0.9212,
      "num_input_tokens_seen": 88327784,
      "step": 5489
    },
    {
      "epoch": 0.38456486905356063,
      "grad_norm": 4.131216526031494,
      "learning_rate": 6.157961471103328e-05,
      "loss": 1.0983,
      "num_input_tokens_seen": 88343480,
      "step": 5490
    },
    {
      "epoch": 0.3846349172992899,
      "grad_norm": 4.294956684112549,
      "learning_rate": 6.157261646234677e-05,
      "loss": 1.224,
      "num_input_tokens_seen": 88359864,
      "step": 5491
    },
    {
      "epoch": 0.38470496554501915,
      "grad_norm": 4.683321952819824,
      "learning_rate": 6.156561821366025e-05,
      "loss": 1.0377,
      "num_input_tokens_seen": 88375184,
      "step": 5492
    },
    {
      "epoch": 0.38477501379074835,
      "grad_norm": 3.932366371154785,
      "learning_rate": 6.155861996497372e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 88391568,
      "step": 5493
    },
    {
      "epoch": 0.3848450620364776,
      "grad_norm": 4.191849231719971,
      "learning_rate": 6.155162171628721e-05,
      "loss": 1.2247,
      "num_input_tokens_seen": 88407680,
      "step": 5494
    },
    {
      "epoch": 0.3849151102822069,
      "grad_norm": 3.983915090560913,
      "learning_rate": 6.154462346760071e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 88422888,
      "step": 5495
    },
    {
      "epoch": 0.38498515852793613,
      "grad_norm": 4.163250923156738,
      "learning_rate": 6.153762521891419e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 88439272,
      "step": 5496
    },
    {
      "epoch": 0.38505520677366534,
      "grad_norm": 6.113068580627441,
      "learning_rate": 6.153062697022768e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 88455656,
      "step": 5497
    },
    {
      "epoch": 0.3851252550193946,
      "grad_norm": 5.32371187210083,
      "learning_rate": 6.152362872154117e-05,
      "loss": 1.1886,
      "num_input_tokens_seen": 88472040,
      "step": 5498
    },
    {
      "epoch": 0.38519530326512386,
      "grad_norm": 6.110095500946045,
      "learning_rate": 6.151663047285464e-05,
      "loss": 0.9587,
      "num_input_tokens_seen": 88487720,
      "step": 5499
    },
    {
      "epoch": 0.3852653515108531,
      "grad_norm": 3.9656851291656494,
      "learning_rate": 6.150963222416813e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 88504104,
      "step": 5500
    },
    {
      "epoch": 0.3853353997565823,
      "grad_norm": 4.601620197296143,
      "learning_rate": 6.150263397548162e-05,
      "loss": 1.2542,
      "num_input_tokens_seen": 88520160,
      "step": 5501
    },
    {
      "epoch": 0.3854054480023116,
      "grad_norm": 4.273797988891602,
      "learning_rate": 6.14956357267951e-05,
      "loss": 1.3405,
      "num_input_tokens_seen": 88535832,
      "step": 5502
    },
    {
      "epoch": 0.38547549624804084,
      "grad_norm": 4.023514747619629,
      "learning_rate": 6.148863747810858e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 88552000,
      "step": 5503
    },
    {
      "epoch": 0.3855455444937701,
      "grad_norm": 3.7229719161987305,
      "learning_rate": 6.148163922942207e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 88567600,
      "step": 5504
    },
    {
      "epoch": 0.38561559273949936,
      "grad_norm": 4.696394920349121,
      "learning_rate": 6.147464098073556e-05,
      "loss": 0.8945,
      "num_input_tokens_seen": 88583224,
      "step": 5505
    },
    {
      "epoch": 0.38568564098522856,
      "grad_norm": 5.354174613952637,
      "learning_rate": 6.146764273204903e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 88599608,
      "step": 5506
    },
    {
      "epoch": 0.3857556892309578,
      "grad_norm": 4.717334747314453,
      "learning_rate": 6.146064448336252e-05,
      "loss": 1.1293,
      "num_input_tokens_seen": 88615048,
      "step": 5507
    },
    {
      "epoch": 0.3858257374766871,
      "grad_norm": 5.373983383178711,
      "learning_rate": 6.145364623467601e-05,
      "loss": 0.9625,
      "num_input_tokens_seen": 88630888,
      "step": 5508
    },
    {
      "epoch": 0.38589578572241634,
      "grad_norm": 4.338916301727295,
      "learning_rate": 6.14466479859895e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 88647072,
      "step": 5509
    },
    {
      "epoch": 0.38596583396814554,
      "grad_norm": 3.898721694946289,
      "learning_rate": 6.143964973730297e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 88663128,
      "step": 5510
    },
    {
      "epoch": 0.3860358822138748,
      "grad_norm": 4.614948749542236,
      "learning_rate": 6.143265148861648e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 88679512,
      "step": 5511
    },
    {
      "epoch": 0.38610593045960406,
      "grad_norm": 5.1157732009887695,
      "learning_rate": 6.142565323992995e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 88695664,
      "step": 5512
    },
    {
      "epoch": 0.3861759787053333,
      "grad_norm": 4.29611873626709,
      "learning_rate": 6.141865499124343e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 88711560,
      "step": 5513
    },
    {
      "epoch": 0.3862460269510625,
      "grad_norm": 3.9464735984802246,
      "learning_rate": 6.141165674255692e-05,
      "loss": 1.1907,
      "num_input_tokens_seen": 88727464,
      "step": 5514
    },
    {
      "epoch": 0.3863160751967918,
      "grad_norm": 3.8381590843200684,
      "learning_rate": 6.140465849387042e-05,
      "loss": 1.1416,
      "num_input_tokens_seen": 88743848,
      "step": 5515
    },
    {
      "epoch": 0.38638612344252105,
      "grad_norm": 3.573434829711914,
      "learning_rate": 6.139766024518389e-05,
      "loss": 0.9316,
      "num_input_tokens_seen": 88759312,
      "step": 5516
    },
    {
      "epoch": 0.3864561716882503,
      "grad_norm": 4.257131576538086,
      "learning_rate": 6.139066199649738e-05,
      "loss": 0.9534,
      "num_input_tokens_seen": 88775112,
      "step": 5517
    },
    {
      "epoch": 0.3865262199339795,
      "grad_norm": 4.2985310554504395,
      "learning_rate": 6.138366374781087e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 88791496,
      "step": 5518
    },
    {
      "epoch": 0.38659626817970877,
      "grad_norm": 3.7012977600097656,
      "learning_rate": 6.137666549912434e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 88807880,
      "step": 5519
    },
    {
      "epoch": 0.38666631642543803,
      "grad_norm": 5.4860453605651855,
      "learning_rate": 6.136966725043782e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 88823392,
      "step": 5520
    },
    {
      "epoch": 0.3867363646711673,
      "grad_norm": 4.165813446044922,
      "learning_rate": 6.136266900175132e-05,
      "loss": 1.0184,
      "num_input_tokens_seen": 88839352,
      "step": 5521
    },
    {
      "epoch": 0.3868064129168965,
      "grad_norm": 3.6253862380981445,
      "learning_rate": 6.135567075306481e-05,
      "loss": 0.9544,
      "num_input_tokens_seen": 88855736,
      "step": 5522
    },
    {
      "epoch": 0.38687646116262575,
      "grad_norm": 3.834057331085205,
      "learning_rate": 6.134867250437829e-05,
      "loss": 1.1863,
      "num_input_tokens_seen": 88871952,
      "step": 5523
    },
    {
      "epoch": 0.386946509408355,
      "grad_norm": 4.534783363342285,
      "learning_rate": 6.134167425569177e-05,
      "loss": 0.85,
      "num_input_tokens_seen": 88888336,
      "step": 5524
    },
    {
      "epoch": 0.38701655765408427,
      "grad_norm": 5.4073381423950195,
      "learning_rate": 6.133467600700526e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 88904256,
      "step": 5525
    },
    {
      "epoch": 0.3870866058998135,
      "grad_norm": 3.819841146469116,
      "learning_rate": 6.132767775831874e-05,
      "loss": 1.0911,
      "num_input_tokens_seen": 88920640,
      "step": 5526
    },
    {
      "epoch": 0.38715665414554273,
      "grad_norm": 3.814857244491577,
      "learning_rate": 6.132067950963223e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 88937024,
      "step": 5527
    },
    {
      "epoch": 0.387226702391272,
      "grad_norm": 3.682535171508789,
      "learning_rate": 6.131368126094571e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 88952712,
      "step": 5528
    },
    {
      "epoch": 0.38729675063700125,
      "grad_norm": 3.5657262802124023,
      "learning_rate": 6.13066830122592e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 88969096,
      "step": 5529
    },
    {
      "epoch": 0.38736679888273046,
      "grad_norm": 3.851977825164795,
      "learning_rate": 6.129968476357268e-05,
      "loss": 1.0546,
      "num_input_tokens_seen": 88985480,
      "step": 5530
    },
    {
      "epoch": 0.3874368471284597,
      "grad_norm": 4.079189777374268,
      "learning_rate": 6.129268651488617e-05,
      "loss": 0.856,
      "num_input_tokens_seen": 89001104,
      "step": 5531
    },
    {
      "epoch": 0.387506895374189,
      "grad_norm": 4.388980865478516,
      "learning_rate": 6.128568826619966e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 89017232,
      "step": 5532
    },
    {
      "epoch": 0.38757694361991823,
      "grad_norm": 3.6747231483459473,
      "learning_rate": 6.127869001751313e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 89033576,
      "step": 5533
    },
    {
      "epoch": 0.38764699186564744,
      "grad_norm": 4.62367057800293,
      "learning_rate": 6.127169176882662e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 89049224,
      "step": 5534
    },
    {
      "epoch": 0.3877170401113767,
      "grad_norm": 3.8601040840148926,
      "learning_rate": 6.126469352014011e-05,
      "loss": 1.0254,
      "num_input_tokens_seen": 89064968,
      "step": 5535
    },
    {
      "epoch": 0.38778708835710596,
      "grad_norm": 5.132208347320557,
      "learning_rate": 6.12576952714536e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 89081352,
      "step": 5536
    },
    {
      "epoch": 0.3878571366028352,
      "grad_norm": 3.9259984493255615,
      "learning_rate": 6.125069702276707e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 89097696,
      "step": 5537
    },
    {
      "epoch": 0.3879271848485644,
      "grad_norm": 3.9004077911376953,
      "learning_rate": 6.124369877408057e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 89114080,
      "step": 5538
    },
    {
      "epoch": 0.3879972330942937,
      "grad_norm": 4.657776355743408,
      "learning_rate": 6.123670052539405e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 89129584,
      "step": 5539
    },
    {
      "epoch": 0.38806728134002294,
      "grad_norm": 3.4758501052856445,
      "learning_rate": 6.122970227670752e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 89145968,
      "step": 5540
    },
    {
      "epoch": 0.3881373295857522,
      "grad_norm": 3.949275255203247,
      "learning_rate": 6.122270402802101e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 89161904,
      "step": 5541
    },
    {
      "epoch": 0.38820737783148146,
      "grad_norm": 5.620425224304199,
      "learning_rate": 6.121570577933451e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 89178032,
      "step": 5542
    },
    {
      "epoch": 0.38827742607721066,
      "grad_norm": 5.1215643882751465,
      "learning_rate": 6.120870753064799e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 89193568,
      "step": 5543
    },
    {
      "epoch": 0.3883474743229399,
      "grad_norm": 3.994556427001953,
      "learning_rate": 6.120170928196148e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 89209952,
      "step": 5544
    },
    {
      "epoch": 0.3884175225686692,
      "grad_norm": 6.86944055557251,
      "learning_rate": 6.119471103327497e-05,
      "loss": 1.208,
      "num_input_tokens_seen": 89226336,
      "step": 5545
    },
    {
      "epoch": 0.38848757081439844,
      "grad_norm": 3.72501540184021,
      "learning_rate": 6.118771278458844e-05,
      "loss": 1.0198,
      "num_input_tokens_seen": 89242720,
      "step": 5546
    },
    {
      "epoch": 0.38855761906012765,
      "grad_norm": 3.6887834072113037,
      "learning_rate": 6.118071453590193e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 89258536,
      "step": 5547
    },
    {
      "epoch": 0.3886276673058569,
      "grad_norm": 5.15130615234375,
      "learning_rate": 6.117371628721542e-05,
      "loss": 1.0193,
      "num_input_tokens_seen": 89274920,
      "step": 5548
    },
    {
      "epoch": 0.38869771555158616,
      "grad_norm": 3.7503981590270996,
      "learning_rate": 6.116671803852891e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 89291304,
      "step": 5549
    },
    {
      "epoch": 0.3887677637973154,
      "grad_norm": 4.851298809051514,
      "learning_rate": 6.115971978984238e-05,
      "loss": 1.132,
      "num_input_tokens_seen": 89307080,
      "step": 5550
    },
    {
      "epoch": 0.38883781204304463,
      "grad_norm": 3.72981858253479,
      "learning_rate": 6.115272154115587e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 89323464,
      "step": 5551
    },
    {
      "epoch": 0.3889078602887739,
      "grad_norm": 4.1301140785217285,
      "learning_rate": 6.114572329246936e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 89339696,
      "step": 5552
    },
    {
      "epoch": 0.38897790853450315,
      "grad_norm": 4.225720405578613,
      "learning_rate": 6.113872504378283e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 89356080,
      "step": 5553
    },
    {
      "epoch": 0.3890479567802324,
      "grad_norm": 3.7197327613830566,
      "learning_rate": 6.113172679509632e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 89372464,
      "step": 5554
    },
    {
      "epoch": 0.3891180050259616,
      "grad_norm": 4.626903057098389,
      "learning_rate": 6.112472854640981e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 89388848,
      "step": 5555
    },
    {
      "epoch": 0.38918805327169087,
      "grad_norm": 4.229621410369873,
      "learning_rate": 6.111773029772329e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 89405112,
      "step": 5556
    },
    {
      "epoch": 0.38925810151742013,
      "grad_norm": 3.5502984523773193,
      "learning_rate": 6.111073204903678e-05,
      "loss": 0.9966,
      "num_input_tokens_seen": 89421496,
      "step": 5557
    },
    {
      "epoch": 0.3893281497631494,
      "grad_norm": 4.251241207122803,
      "learning_rate": 6.110373380035026e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 89437880,
      "step": 5558
    },
    {
      "epoch": 0.3893981980088786,
      "grad_norm": 5.076200485229492,
      "learning_rate": 6.109673555166375e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 89454264,
      "step": 5559
    },
    {
      "epoch": 0.38946824625460785,
      "grad_norm": 4.018000602722168,
      "learning_rate": 6.108973730297723e-05,
      "loss": 1.197,
      "num_input_tokens_seen": 89470648,
      "step": 5560
    },
    {
      "epoch": 0.3895382945003371,
      "grad_norm": 4.3367180824279785,
      "learning_rate": 6.108273905429072e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 89487032,
      "step": 5561
    },
    {
      "epoch": 0.38960834274606637,
      "grad_norm": 6.027153015136719,
      "learning_rate": 6.10757408056042e-05,
      "loss": 0.7955,
      "num_input_tokens_seen": 89502064,
      "step": 5562
    },
    {
      "epoch": 0.3896783909917956,
      "grad_norm": 3.499268054962158,
      "learning_rate": 6.10687425569177e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 89518448,
      "step": 5563
    },
    {
      "epoch": 0.38974843923752484,
      "grad_norm": 3.3691868782043457,
      "learning_rate": 6.106174430823117e-05,
      "loss": 0.9249,
      "num_input_tokens_seen": 89534640,
      "step": 5564
    },
    {
      "epoch": 0.3898184874832541,
      "grad_norm": 3.4140114784240723,
      "learning_rate": 6.105474605954467e-05,
      "loss": 0.9594,
      "num_input_tokens_seen": 89551024,
      "step": 5565
    },
    {
      "epoch": 0.38988853572898335,
      "grad_norm": 4.049834728240967,
      "learning_rate": 6.104774781085815e-05,
      "loss": 1.1637,
      "num_input_tokens_seen": 89567408,
      "step": 5566
    },
    {
      "epoch": 0.38995858397471256,
      "grad_norm": 4.825027942657471,
      "learning_rate": 6.104074956217162e-05,
      "loss": 1.0514,
      "num_input_tokens_seen": 89583792,
      "step": 5567
    },
    {
      "epoch": 0.3900286322204418,
      "grad_norm": 5.281174659729004,
      "learning_rate": 6.103375131348512e-05,
      "loss": 1.0534,
      "num_input_tokens_seen": 89600176,
      "step": 5568
    },
    {
      "epoch": 0.3900986804661711,
      "grad_norm": 3.567270278930664,
      "learning_rate": 6.1026753064798605e-05,
      "loss": 0.9217,
      "num_input_tokens_seen": 89616560,
      "step": 5569
    },
    {
      "epoch": 0.39016872871190034,
      "grad_norm": 5.449852466583252,
      "learning_rate": 6.1019754816112086e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 89631968,
      "step": 5570
    },
    {
      "epoch": 0.39023877695762954,
      "grad_norm": 4.016347885131836,
      "learning_rate": 6.1012756567425575e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 89646712,
      "step": 5571
    },
    {
      "epoch": 0.3903088252033588,
      "grad_norm": 3.8826510906219482,
      "learning_rate": 6.100575831873906e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 89662776,
      "step": 5572
    },
    {
      "epoch": 0.39037887344908806,
      "grad_norm": 3.80755615234375,
      "learning_rate": 6.099876007005254e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 89679096,
      "step": 5573
    },
    {
      "epoch": 0.3904489216948173,
      "grad_norm": 3.7274065017700195,
      "learning_rate": 6.099176182136602e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 89695048,
      "step": 5574
    },
    {
      "epoch": 0.3905189699405466,
      "grad_norm": 4.335930824279785,
      "learning_rate": 6.09847635726795e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 89711432,
      "step": 5575
    },
    {
      "epoch": 0.3905890181862758,
      "grad_norm": 3.8839964866638184,
      "learning_rate": 6.0977765323993e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 89727712,
      "step": 5576
    },
    {
      "epoch": 0.39065906643200504,
      "grad_norm": 4.8028035163879395,
      "learning_rate": 6.097076707530648e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 89744096,
      "step": 5577
    },
    {
      "epoch": 0.3907291146777343,
      "grad_norm": 4.042201519012451,
      "learning_rate": 6.096376882661997e-05,
      "loss": 0.8609,
      "num_input_tokens_seen": 89758688,
      "step": 5578
    },
    {
      "epoch": 0.39079916292346356,
      "grad_norm": 4.1316986083984375,
      "learning_rate": 6.095677057793345e-05,
      "loss": 0.9509,
      "num_input_tokens_seen": 89774720,
      "step": 5579
    },
    {
      "epoch": 0.39086921116919277,
      "grad_norm": 5.164004802703857,
      "learning_rate": 6.094977232924693e-05,
      "loss": 1.1927,
      "num_input_tokens_seen": 89788480,
      "step": 5580
    },
    {
      "epoch": 0.390939259414922,
      "grad_norm": 4.125234127044678,
      "learning_rate": 6.094277408056043e-05,
      "loss": 0.9237,
      "num_input_tokens_seen": 89803000,
      "step": 5581
    },
    {
      "epoch": 0.3910093076606513,
      "grad_norm": 4.798699855804443,
      "learning_rate": 6.093577583187392e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 89818288,
      "step": 5582
    },
    {
      "epoch": 0.39107935590638054,
      "grad_norm": 4.383975028991699,
      "learning_rate": 6.0928777583187404e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 89833616,
      "step": 5583
    },
    {
      "epoch": 0.39114940415210975,
      "grad_norm": 4.20830774307251,
      "learning_rate": 6.092177933450087e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 89849200,
      "step": 5584
    },
    {
      "epoch": 0.391219452397839,
      "grad_norm": 4.470288276672363,
      "learning_rate": 6.091478108581437e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 89865184,
      "step": 5585
    },
    {
      "epoch": 0.39128950064356827,
      "grad_norm": 4.8457112312316895,
      "learning_rate": 6.090778283712785e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 89881184,
      "step": 5586
    },
    {
      "epoch": 0.3913595488892975,
      "grad_norm": 4.112722873687744,
      "learning_rate": 6.090078458844133e-05,
      "loss": 1.0189,
      "num_input_tokens_seen": 89897568,
      "step": 5587
    },
    {
      "epoch": 0.39142959713502673,
      "grad_norm": 4.070732116699219,
      "learning_rate": 6.089378633975482e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 89913952,
      "step": 5588
    },
    {
      "epoch": 0.391499645380756,
      "grad_norm": 4.092299461364746,
      "learning_rate": 6.08867880910683e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 89929216,
      "step": 5589
    },
    {
      "epoch": 0.39156969362648525,
      "grad_norm": 3.8092305660247803,
      "learning_rate": 6.08797898423818e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 89945376,
      "step": 5590
    },
    {
      "epoch": 0.3916397418722145,
      "grad_norm": 5.461154937744141,
      "learning_rate": 6.0872791593695265e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 89960296,
      "step": 5591
    },
    {
      "epoch": 0.3917097901179437,
      "grad_norm": 5.2103190422058105,
      "learning_rate": 6.086579334500877e-05,
      "loss": 1.079,
      "num_input_tokens_seen": 89975864,
      "step": 5592
    },
    {
      "epoch": 0.39177983836367297,
      "grad_norm": 3.6308488845825195,
      "learning_rate": 6.085879509632225e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 89991880,
      "step": 5593
    },
    {
      "epoch": 0.39184988660940223,
      "grad_norm": 4.632900714874268,
      "learning_rate": 6.0851796847635724e-05,
      "loss": 0.8255,
      "num_input_tokens_seen": 90006560,
      "step": 5594
    },
    {
      "epoch": 0.3919199348551315,
      "grad_norm": 3.8614165782928467,
      "learning_rate": 6.084479859894921e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 90022136,
      "step": 5595
    },
    {
      "epoch": 0.3919899831008607,
      "grad_norm": 3.768287420272827,
      "learning_rate": 6.0837800350262695e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 90038520,
      "step": 5596
    },
    {
      "epoch": 0.39206003134658995,
      "grad_norm": 3.355902671813965,
      "learning_rate": 6.083080210157618e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 90054592,
      "step": 5597
    },
    {
      "epoch": 0.3921300795923192,
      "grad_norm": 3.2001609802246094,
      "learning_rate": 6.082380385288967e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 90070976,
      "step": 5598
    },
    {
      "epoch": 0.3922001278380485,
      "grad_norm": 3.74692440032959,
      "learning_rate": 6.081680560420317e-05,
      "loss": 0.9807,
      "num_input_tokens_seen": 90087360,
      "step": 5599
    },
    {
      "epoch": 0.3922701760837777,
      "grad_norm": 5.602208614349365,
      "learning_rate": 6.080980735551665e-05,
      "loss": 1.2451,
      "num_input_tokens_seen": 90103744,
      "step": 5600
    },
    {
      "epoch": 0.3922701760837777,
      "eval_loss": 1.127113699913025,
      "eval_runtime": 0.2033,
      "eval_samples_per_second": 4.919,
      "eval_steps_per_second": 4.919,
      "num_input_tokens_seen": 90103744,
      "step": 5600
    },
    {
      "epoch": 0.39234022432950694,
      "grad_norm": 3.727559804916382,
      "learning_rate": 6.080280910683012e-05,
      "loss": 1.0777,
      "num_input_tokens_seen": 90119400,
      "step": 5601
    },
    {
      "epoch": 0.3924102725752362,
      "grad_norm": 4.577515125274658,
      "learning_rate": 6.0795810858143606e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 90135168,
      "step": 5602
    },
    {
      "epoch": 0.39248032082096546,
      "grad_norm": 5.225588798522949,
      "learning_rate": 6.07888126094571e-05,
      "loss": 1.0964,
      "num_input_tokens_seen": 90151480,
      "step": 5603
    },
    {
      "epoch": 0.39255036906669466,
      "grad_norm": 3.6131844520568848,
      "learning_rate": 6.078181436077057e-05,
      "loss": 0.9255,
      "num_input_tokens_seen": 90167864,
      "step": 5604
    },
    {
      "epoch": 0.3926204173124239,
      "grad_norm": 4.127248287200928,
      "learning_rate": 6.0774816112084065e-05,
      "loss": 1.1939,
      "num_input_tokens_seen": 90184248,
      "step": 5605
    },
    {
      "epoch": 0.3926904655581532,
      "grad_norm": 4.599911689758301,
      "learning_rate": 6.076781786339756e-05,
      "loss": 1.2819,
      "num_input_tokens_seen": 90199816,
      "step": 5606
    },
    {
      "epoch": 0.39276051380388244,
      "grad_norm": 3.7179722785949707,
      "learning_rate": 6.076081961471104e-05,
      "loss": 0.8559,
      "num_input_tokens_seen": 90215640,
      "step": 5607
    },
    {
      "epoch": 0.39283056204961164,
      "grad_norm": 3.872941493988037,
      "learning_rate": 6.0753821366024524e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 90232024,
      "step": 5608
    },
    {
      "epoch": 0.3929006102953409,
      "grad_norm": 3.7326548099517822,
      "learning_rate": 6.074682311733801e-05,
      "loss": 1.0295,
      "num_input_tokens_seen": 90247536,
      "step": 5609
    },
    {
      "epoch": 0.39297065854107016,
      "grad_norm": 4.05418062210083,
      "learning_rate": 6.0739824868651494e-05,
      "loss": 1.2199,
      "num_input_tokens_seen": 90263920,
      "step": 5610
    },
    {
      "epoch": 0.3930407067867994,
      "grad_norm": 5.326319694519043,
      "learning_rate": 6.073282661996497e-05,
      "loss": 1.0705,
      "num_input_tokens_seen": 90280240,
      "step": 5611
    },
    {
      "epoch": 0.3931107550325287,
      "grad_norm": 4.132864952087402,
      "learning_rate": 6.072582837127846e-05,
      "loss": 1.2377,
      "num_input_tokens_seen": 90296128,
      "step": 5612
    },
    {
      "epoch": 0.3931808032782579,
      "grad_norm": 3.7307562828063965,
      "learning_rate": 6.0718830122591953e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 90312512,
      "step": 5613
    },
    {
      "epoch": 0.39325085152398714,
      "grad_norm": 6.35123872756958,
      "learning_rate": 6.0711831873905435e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 90328896,
      "step": 5614
    },
    {
      "epoch": 0.3933208997697164,
      "grad_norm": 5.536827564239502,
      "learning_rate": 6.070483362521892e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 90344648,
      "step": 5615
    },
    {
      "epoch": 0.39339094801544566,
      "grad_norm": 3.6394944190979004,
      "learning_rate": 6.0697835376532406e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 90361032,
      "step": 5616
    },
    {
      "epoch": 0.39346099626117487,
      "grad_norm": 3.4719443321228027,
      "learning_rate": 6.069083712784589e-05,
      "loss": 0.9859,
      "num_input_tokens_seen": 90377192,
      "step": 5617
    },
    {
      "epoch": 0.3935310445069041,
      "grad_norm": 6.195781230926514,
      "learning_rate": 6.068383887915936e-05,
      "loss": 0.9482,
      "num_input_tokens_seen": 90393576,
      "step": 5618
    },
    {
      "epoch": 0.3936010927526334,
      "grad_norm": 3.5569331645965576,
      "learning_rate": 6.0676840630472865e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 90409960,
      "step": 5619
    },
    {
      "epoch": 0.39367114099836265,
      "grad_norm": 5.535704135894775,
      "learning_rate": 6.0669842381786346e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 90426216,
      "step": 5620
    },
    {
      "epoch": 0.39374118924409185,
      "grad_norm": 3.81278395652771,
      "learning_rate": 6.0662844133099815e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 90442320,
      "step": 5621
    },
    {
      "epoch": 0.3938112374898211,
      "grad_norm": 4.093146324157715,
      "learning_rate": 6.065584588441331e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 90458704,
      "step": 5622
    },
    {
      "epoch": 0.39388128573555037,
      "grad_norm": 6.061161518096924,
      "learning_rate": 6.06488476357268e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 90474408,
      "step": 5623
    },
    {
      "epoch": 0.39395133398127963,
      "grad_norm": 3.763059616088867,
      "learning_rate": 6.064184938704028e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 90490592,
      "step": 5624
    },
    {
      "epoch": 0.39402138222700883,
      "grad_norm": 5.287941932678223,
      "learning_rate": 6.063485113835376e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 90506568,
      "step": 5625
    },
    {
      "epoch": 0.3940914304727381,
      "grad_norm": 3.5837693214416504,
      "learning_rate": 6.062785288966726e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 90522952,
      "step": 5626
    },
    {
      "epoch": 0.39416147871846735,
      "grad_norm": 3.5523111820220947,
      "learning_rate": 6.062085464098074e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 90539336,
      "step": 5627
    },
    {
      "epoch": 0.3942315269641966,
      "grad_norm": 4.3880934715271,
      "learning_rate": 6.061385639229421e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 90555720,
      "step": 5628
    },
    {
      "epoch": 0.3943015752099258,
      "grad_norm": 4.26425313949585,
      "learning_rate": 6.060685814360772e-05,
      "loss": 1.08,
      "num_input_tokens_seen": 90572104,
      "step": 5629
    },
    {
      "epoch": 0.3943716234556551,
      "grad_norm": 3.8837990760803223,
      "learning_rate": 6.05998598949212e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 90588488,
      "step": 5630
    },
    {
      "epoch": 0.39444167170138433,
      "grad_norm": 3.704282522201538,
      "learning_rate": 6.059286164623468e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 90604872,
      "step": 5631
    },
    {
      "epoch": 0.3945117199471136,
      "grad_norm": 5.933957099914551,
      "learning_rate": 6.058586339754816e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 90621256,
      "step": 5632
    },
    {
      "epoch": 0.3945817681928428,
      "grad_norm": 4.185206413269043,
      "learning_rate": 6.057886514886165e-05,
      "loss": 1.0044,
      "num_input_tokens_seen": 90637640,
      "step": 5633
    },
    {
      "epoch": 0.39465181643857206,
      "grad_norm": 3.97603440284729,
      "learning_rate": 6.057186690017513e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 90654024,
      "step": 5634
    },
    {
      "epoch": 0.3947218646843013,
      "grad_norm": 3.394630193710327,
      "learning_rate": 6.056486865148863e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 90670008,
      "step": 5635
    },
    {
      "epoch": 0.3947919129300306,
      "grad_norm": 3.810899019241333,
      "learning_rate": 6.055787040280211e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 90686392,
      "step": 5636
    },
    {
      "epoch": 0.3948619611757598,
      "grad_norm": 4.237402439117432,
      "learning_rate": 6.055087215411559e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 90702776,
      "step": 5637
    },
    {
      "epoch": 0.39493200942148904,
      "grad_norm": 5.481308937072754,
      "learning_rate": 6.054387390542907e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 90718312,
      "step": 5638
    },
    {
      "epoch": 0.3950020576672183,
      "grad_norm": 3.582808017730713,
      "learning_rate": 6.053687565674256e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 90733856,
      "step": 5639
    },
    {
      "epoch": 0.39507210591294756,
      "grad_norm": 3.9277966022491455,
      "learning_rate": 6.0529877408056044e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 90750240,
      "step": 5640
    },
    {
      "epoch": 0.39514215415867676,
      "grad_norm": 4.61000394821167,
      "learning_rate": 6.0522879159369526e-05,
      "loss": 1.1705,
      "num_input_tokens_seen": 90766352,
      "step": 5641
    },
    {
      "epoch": 0.395212202404406,
      "grad_norm": 4.445149898529053,
      "learning_rate": 6.051588091068301e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 90782736,
      "step": 5642
    },
    {
      "epoch": 0.3952822506501353,
      "grad_norm": 4.652968406677246,
      "learning_rate": 6.0508882661996516e-05,
      "loss": 0.9761,
      "num_input_tokens_seen": 90799120,
      "step": 5643
    },
    {
      "epoch": 0.39535229889586454,
      "grad_norm": 4.172330856323242,
      "learning_rate": 6.0501884413309985e-05,
      "loss": 1.0637,
      "num_input_tokens_seen": 90815504,
      "step": 5644
    },
    {
      "epoch": 0.3954223471415938,
      "grad_norm": 3.647385358810425,
      "learning_rate": 6.0494886164623466e-05,
      "loss": 0.9284,
      "num_input_tokens_seen": 90831888,
      "step": 5645
    },
    {
      "epoch": 0.395492395387323,
      "grad_norm": 3.9353525638580322,
      "learning_rate": 6.0487887915936955e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 90848048,
      "step": 5646
    },
    {
      "epoch": 0.39556244363305226,
      "grad_norm": 4.216567039489746,
      "learning_rate": 6.048088966725044e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 90863576,
      "step": 5647
    },
    {
      "epoch": 0.3956324918787815,
      "grad_norm": 5.031260013580322,
      "learning_rate": 6.047389141856392e-05,
      "loss": 1.1314,
      "num_input_tokens_seen": 90879960,
      "step": 5648
    },
    {
      "epoch": 0.3957025401245108,
      "grad_norm": 4.927192211151123,
      "learning_rate": 6.0466893169877414e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 90896344,
      "step": 5649
    },
    {
      "epoch": 0.39577258837024,
      "grad_norm": 4.589445114135742,
      "learning_rate": 6.045989492119091e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 90912728,
      "step": 5650
    },
    {
      "epoch": 0.39584263661596925,
      "grad_norm": 3.5707035064697266,
      "learning_rate": 6.045289667250438e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 90929112,
      "step": 5651
    },
    {
      "epoch": 0.3959126848616985,
      "grad_norm": 3.637237787246704,
      "learning_rate": 6.044589842381787e-05,
      "loss": 1.2063,
      "num_input_tokens_seen": 90945376,
      "step": 5652
    },
    {
      "epoch": 0.39598273310742776,
      "grad_norm": 4.068975448608398,
      "learning_rate": 6.043890017513136e-05,
      "loss": 1.0301,
      "num_input_tokens_seen": 90961448,
      "step": 5653
    },
    {
      "epoch": 0.39605278135315697,
      "grad_norm": 3.8378570079803467,
      "learning_rate": 6.0431901926444837e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 90977832,
      "step": 5654
    },
    {
      "epoch": 0.39612282959888623,
      "grad_norm": 4.5788092613220215,
      "learning_rate": 6.042490367775832e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 90993296,
      "step": 5655
    },
    {
      "epoch": 0.3961928778446155,
      "grad_norm": 3.7392847537994385,
      "learning_rate": 6.041790542907181e-05,
      "loss": 1.0081,
      "num_input_tokens_seen": 91009680,
      "step": 5656
    },
    {
      "epoch": 0.39626292609034475,
      "grad_norm": 5.830812931060791,
      "learning_rate": 6.04109071803853e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 91025400,
      "step": 5657
    },
    {
      "epoch": 0.39633297433607395,
      "grad_norm": 3.7372663021087646,
      "learning_rate": 6.040390893169877e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 91041784,
      "step": 5658
    },
    {
      "epoch": 0.3964030225818032,
      "grad_norm": 3.756762981414795,
      "learning_rate": 6.0396910683012266e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 91058168,
      "step": 5659
    },
    {
      "epoch": 0.39647307082753247,
      "grad_norm": 3.659280776977539,
      "learning_rate": 6.0389912434325755e-05,
      "loss": 1.0396,
      "num_input_tokens_seen": 91074288,
      "step": 5660
    },
    {
      "epoch": 0.39654311907326173,
      "grad_norm": 4.339829921722412,
      "learning_rate": 6.038291418563923e-05,
      "loss": 1.2101,
      "num_input_tokens_seen": 91090024,
      "step": 5661
    },
    {
      "epoch": 0.39661316731899093,
      "grad_norm": 4.062867641448975,
      "learning_rate": 6.037591593695272e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 91106408,
      "step": 5662
    },
    {
      "epoch": 0.3966832155647202,
      "grad_norm": 4.45166015625,
      "learning_rate": 6.0368917688266214e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 91122648,
      "step": 5663
    },
    {
      "epoch": 0.39675326381044945,
      "grad_norm": 3.4350759983062744,
      "learning_rate": 6.036191943957968e-05,
      "loss": 0.8876,
      "num_input_tokens_seen": 91138200,
      "step": 5664
    },
    {
      "epoch": 0.3968233120561787,
      "grad_norm": 3.5637154579162598,
      "learning_rate": 6.0354921190893164e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 91154584,
      "step": 5665
    },
    {
      "epoch": 0.3968933603019079,
      "grad_norm": 3.8793985843658447,
      "learning_rate": 6.034792294220666e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 91170968,
      "step": 5666
    },
    {
      "epoch": 0.3969634085476372,
      "grad_norm": 4.1613545417785645,
      "learning_rate": 6.0340924693520154e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 91185856,
      "step": 5667
    },
    {
      "epoch": 0.39703345679336643,
      "grad_norm": 8.646449089050293,
      "learning_rate": 6.033392644483362e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 91201088,
      "step": 5668
    },
    {
      "epoch": 0.3971035050390957,
      "grad_norm": 4.862243175506592,
      "learning_rate": 6.032692819614711e-05,
      "loss": 1.1871,
      "num_input_tokens_seen": 91217472,
      "step": 5669
    },
    {
      "epoch": 0.3971735532848249,
      "grad_norm": 4.013809680938721,
      "learning_rate": 6.0319929947460607e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 91233760,
      "step": 5670
    },
    {
      "epoch": 0.39724360153055416,
      "grad_norm": 4.664083480834961,
      "learning_rate": 6.0312931698774075e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 91250144,
      "step": 5671
    },
    {
      "epoch": 0.3973136497762834,
      "grad_norm": 4.330606937408447,
      "learning_rate": 6.030593345008756e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 91264592,
      "step": 5672
    },
    {
      "epoch": 0.3973836980220127,
      "grad_norm": 4.158743381500244,
      "learning_rate": 6.0298935201401066e-05,
      "loss": 0.8608,
      "num_input_tokens_seen": 91280520,
      "step": 5673
    },
    {
      "epoch": 0.3974537462677419,
      "grad_norm": 3.800955057144165,
      "learning_rate": 6.029193695271455e-05,
      "loss": 0.9451,
      "num_input_tokens_seen": 91296472,
      "step": 5674
    },
    {
      "epoch": 0.39752379451347114,
      "grad_norm": 4.307434558868408,
      "learning_rate": 6.028493870402803e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 91312856,
      "step": 5675
    },
    {
      "epoch": 0.3975938427592004,
      "grad_norm": 4.052398204803467,
      "learning_rate": 6.027794045534151e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 91329216,
      "step": 5676
    },
    {
      "epoch": 0.39766389100492966,
      "grad_norm": 4.665764331817627,
      "learning_rate": 6.0270942206655e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 91345184,
      "step": 5677
    },
    {
      "epoch": 0.3977339392506589,
      "grad_norm": 4.070000648498535,
      "learning_rate": 6.0263943957968475e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 91361568,
      "step": 5678
    },
    {
      "epoch": 0.3978039874963881,
      "grad_norm": 4.304214954376221,
      "learning_rate": 6.025694570928198e-05,
      "loss": 0.8688,
      "num_input_tokens_seen": 91376656,
      "step": 5679
    },
    {
      "epoch": 0.3978740357421174,
      "grad_norm": 4.051540374755859,
      "learning_rate": 6.024994746059546e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 91393040,
      "step": 5680
    },
    {
      "epoch": 0.39794408398784664,
      "grad_norm": 3.986542224884033,
      "learning_rate": 6.024294921190894e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 91408208,
      "step": 5681
    },
    {
      "epoch": 0.3980141322335759,
      "grad_norm": 4.302114963531494,
      "learning_rate": 6.023595096322242e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 91424592,
      "step": 5682
    },
    {
      "epoch": 0.3980841804793051,
      "grad_norm": 6.019785404205322,
      "learning_rate": 6.0228952714535904e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 91440976,
      "step": 5683
    },
    {
      "epoch": 0.39815422872503436,
      "grad_norm": 3.667469024658203,
      "learning_rate": 6.022195446584939e-05,
      "loss": 1.0587,
      "num_input_tokens_seen": 91457360,
      "step": 5684
    },
    {
      "epoch": 0.3982242769707636,
      "grad_norm": 4.30043363571167,
      "learning_rate": 6.0214956217162874e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 91473744,
      "step": 5685
    },
    {
      "epoch": 0.3982943252164929,
      "grad_norm": 3.7446558475494385,
      "learning_rate": 6.0207957968476356e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 91489544,
      "step": 5686
    },
    {
      "epoch": 0.3983643734622221,
      "grad_norm": 3.722567081451416,
      "learning_rate": 6.020095971978985e-05,
      "loss": 0.9913,
      "num_input_tokens_seen": 91505584,
      "step": 5687
    },
    {
      "epoch": 0.39843442170795135,
      "grad_norm": 4.311237812042236,
      "learning_rate": 6.019396147110332e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 91520656,
      "step": 5688
    },
    {
      "epoch": 0.3985044699536806,
      "grad_norm": 5.650984764099121,
      "learning_rate": 6.018696322241683e-05,
      "loss": 1.2826,
      "num_input_tokens_seen": 91537040,
      "step": 5689
    },
    {
      "epoch": 0.39857451819940987,
      "grad_norm": 4.106716632843018,
      "learning_rate": 6.017996497373031e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 91553424,
      "step": 5690
    },
    {
      "epoch": 0.39864456644513907,
      "grad_norm": 3.893007516860962,
      "learning_rate": 6.0172966725043786e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 91569184,
      "step": 5691
    },
    {
      "epoch": 0.39871461469086833,
      "grad_norm": 3.6435177326202393,
      "learning_rate": 6.016596847635727e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 91584832,
      "step": 5692
    },
    {
      "epoch": 0.3987846629365976,
      "grad_norm": 3.639324188232422,
      "learning_rate": 6.015897022767075e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 91601216,
      "step": 5693
    },
    {
      "epoch": 0.39885471118232685,
      "grad_norm": 3.680997848510742,
      "learning_rate": 6.0151971978984245e-05,
      "loss": 1.1352,
      "num_input_tokens_seen": 91617600,
      "step": 5694
    },
    {
      "epoch": 0.39892475942805605,
      "grad_norm": 5.37217903137207,
      "learning_rate": 6.0144973730297726e-05,
      "loss": 1.1507,
      "num_input_tokens_seen": 91633760,
      "step": 5695
    },
    {
      "epoch": 0.3989948076737853,
      "grad_norm": 3.6297101974487305,
      "learning_rate": 6.013797548161122e-05,
      "loss": 0.8412,
      "num_input_tokens_seen": 91649536,
      "step": 5696
    },
    {
      "epoch": 0.39906485591951457,
      "grad_norm": 8.193422317504883,
      "learning_rate": 6.0130977232924704e-05,
      "loss": 1.1533,
      "num_input_tokens_seen": 91665920,
      "step": 5697
    },
    {
      "epoch": 0.39913490416524383,
      "grad_norm": 3.6126644611358643,
      "learning_rate": 6.0123978984238185e-05,
      "loss": 0.8261,
      "num_input_tokens_seen": 91682272,
      "step": 5698
    },
    {
      "epoch": 0.39920495241097304,
      "grad_norm": 4.277047634124756,
      "learning_rate": 6.011698073555167e-05,
      "loss": 1.0904,
      "num_input_tokens_seen": 91698656,
      "step": 5699
    },
    {
      "epoch": 0.3992750006567023,
      "grad_norm": 4.661556720733643,
      "learning_rate": 6.0109982486865156e-05,
      "loss": 1.1948,
      "num_input_tokens_seen": 91715040,
      "step": 5700
    },
    {
      "epoch": 0.39934504890243155,
      "grad_norm": 4.143563270568848,
      "learning_rate": 6.010298423817864e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 91731424,
      "step": 5701
    },
    {
      "epoch": 0.3994150971481608,
      "grad_norm": 5.750835418701172,
      "learning_rate": 6.009598598949212e-05,
      "loss": 1.0594,
      "num_input_tokens_seen": 91747808,
      "step": 5702
    },
    {
      "epoch": 0.39948514539389,
      "grad_norm": 4.048924446105957,
      "learning_rate": 6.0088987740805615e-05,
      "loss": 1.0488,
      "num_input_tokens_seen": 91764192,
      "step": 5703
    },
    {
      "epoch": 0.3995551936396193,
      "grad_norm": 3.7284796237945557,
      "learning_rate": 6.00819894921191e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 91780576,
      "step": 5704
    },
    {
      "epoch": 0.39962524188534854,
      "grad_norm": 4.041873931884766,
      "learning_rate": 6.007499124343258e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 91796960,
      "step": 5705
    },
    {
      "epoch": 0.3996952901310778,
      "grad_norm": 4.37992000579834,
      "learning_rate": 6.006799299474607e-05,
      "loss": 1.3032,
      "num_input_tokens_seen": 91813344,
      "step": 5706
    },
    {
      "epoch": 0.399765338376807,
      "grad_norm": 4.49924373626709,
      "learning_rate": 6.006099474605955e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 91829728,
      "step": 5707
    },
    {
      "epoch": 0.39983538662253626,
      "grad_norm": 4.609983444213867,
      "learning_rate": 6.005399649737303e-05,
      "loss": 1.0591,
      "num_input_tokens_seen": 91844968,
      "step": 5708
    },
    {
      "epoch": 0.3999054348682655,
      "grad_norm": 4.1750006675720215,
      "learning_rate": 6.004699824868651e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 91860464,
      "step": 5709
    },
    {
      "epoch": 0.3999754831139948,
      "grad_norm": 4.009062767028809,
      "learning_rate": 6.0039999999999994e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 91876848,
      "step": 5710
    },
    {
      "epoch": 0.400045531359724,
      "grad_norm": 4.0535078048706055,
      "learning_rate": 6.003300175131349e-05,
      "loss": 1.0851,
      "num_input_tokens_seen": 91892192,
      "step": 5711
    },
    {
      "epoch": 0.40011557960545324,
      "grad_norm": 4.592657089233398,
      "learning_rate": 6.002600350262697e-05,
      "loss": 0.8823,
      "num_input_tokens_seen": 91907040,
      "step": 5712
    },
    {
      "epoch": 0.4001856278511825,
      "grad_norm": 3.87369966506958,
      "learning_rate": 6.001900525394046e-05,
      "loss": 1.153,
      "num_input_tokens_seen": 91922712,
      "step": 5713
    },
    {
      "epoch": 0.40025567609691176,
      "grad_norm": 3.93766713142395,
      "learning_rate": 6.001200700525394e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 91939096,
      "step": 5714
    },
    {
      "epoch": 0.400325724342641,
      "grad_norm": 3.823153018951416,
      "learning_rate": 6.0005008756567424e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 91955016,
      "step": 5715
    },
    {
      "epoch": 0.4003957725883702,
      "grad_norm": 3.5592081546783447,
      "learning_rate": 5.999801050788092e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 91971400,
      "step": 5716
    },
    {
      "epoch": 0.4004658208340995,
      "grad_norm": 3.8749887943267822,
      "learning_rate": 5.9991012259194414e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 91987784,
      "step": 5717
    },
    {
      "epoch": 0.40053586907982874,
      "grad_norm": 4.749402046203613,
      "learning_rate": 5.998401401050788e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 92003592,
      "step": 5718
    },
    {
      "epoch": 0.400605917325558,
      "grad_norm": 4.114437580108643,
      "learning_rate": 5.9977015761821365e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 92019640,
      "step": 5719
    },
    {
      "epoch": 0.4006759655712872,
      "grad_norm": 3.3297617435455322,
      "learning_rate": 5.997001751313486e-05,
      "loss": 0.8776,
      "num_input_tokens_seen": 92035544,
      "step": 5720
    },
    {
      "epoch": 0.40074601381701647,
      "grad_norm": 4.204908847808838,
      "learning_rate": 5.996301926444834e-05,
      "loss": 1.141,
      "num_input_tokens_seen": 92051840,
      "step": 5721
    },
    {
      "epoch": 0.4008160620627457,
      "grad_norm": 4.198369979858398,
      "learning_rate": 5.9956021015761824e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 92068224,
      "step": 5722
    },
    {
      "epoch": 0.400886110308475,
      "grad_norm": 4.46641206741333,
      "learning_rate": 5.994902276707531e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 92083656,
      "step": 5723
    },
    {
      "epoch": 0.4009561585542042,
      "grad_norm": 4.2217535972595215,
      "learning_rate": 5.994202451838881e-05,
      "loss": 1.066,
      "num_input_tokens_seen": 92100040,
      "step": 5724
    },
    {
      "epoch": 0.40102620679993345,
      "grad_norm": 4.484360218048096,
      "learning_rate": 5.993502626970229e-05,
      "loss": 1.1031,
      "num_input_tokens_seen": 92115592,
      "step": 5725
    },
    {
      "epoch": 0.4010962550456627,
      "grad_norm": 4.69040060043335,
      "learning_rate": 5.992802802101576e-05,
      "loss": 1.1487,
      "num_input_tokens_seen": 92131280,
      "step": 5726
    },
    {
      "epoch": 0.40116630329139197,
      "grad_norm": 3.8119077682495117,
      "learning_rate": 5.992102977232926e-05,
      "loss": 1.1336,
      "num_input_tokens_seen": 92147664,
      "step": 5727
    },
    {
      "epoch": 0.40123635153712117,
      "grad_norm": 4.186896800994873,
      "learning_rate": 5.9914031523642735e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 92164048,
      "step": 5728
    },
    {
      "epoch": 0.40130639978285043,
      "grad_norm": 4.658702850341797,
      "learning_rate": 5.9907033274956217e-05,
      "loss": 1.1733,
      "num_input_tokens_seen": 92180432,
      "step": 5729
    },
    {
      "epoch": 0.4013764480285797,
      "grad_norm": 3.8305857181549072,
      "learning_rate": 5.9900035026269705e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 92196816,
      "step": 5730
    },
    {
      "epoch": 0.40144649627430895,
      "grad_norm": 4.063295364379883,
      "learning_rate": 5.989303677758319e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 92212928,
      "step": 5731
    },
    {
      "epoch": 0.40151654452003815,
      "grad_norm": 6.850064277648926,
      "learning_rate": 5.988603852889667e-05,
      "loss": 1.0967,
      "num_input_tokens_seen": 92229312,
      "step": 5732
    },
    {
      "epoch": 0.4015865927657674,
      "grad_norm": 5.065973281860352,
      "learning_rate": 5.9879040280210164e-05,
      "loss": 1.0513,
      "num_input_tokens_seen": 92245696,
      "step": 5733
    },
    {
      "epoch": 0.4016566410114967,
      "grad_norm": 4.02882719039917,
      "learning_rate": 5.987204203152366e-05,
      "loss": 0.934,
      "num_input_tokens_seen": 92261936,
      "step": 5734
    },
    {
      "epoch": 0.40172668925722593,
      "grad_norm": 3.9505250453948975,
      "learning_rate": 5.986504378283714e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 92278320,
      "step": 5735
    },
    {
      "epoch": 0.40179673750295514,
      "grad_norm": 4.949488162994385,
      "learning_rate": 5.985804553415061e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 92294664,
      "step": 5736
    },
    {
      "epoch": 0.4018667857486844,
      "grad_norm": 3.614008903503418,
      "learning_rate": 5.98510472854641e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 92311048,
      "step": 5737
    },
    {
      "epoch": 0.40193683399441366,
      "grad_norm": 3.739224433898926,
      "learning_rate": 5.984404903677758e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 92327432,
      "step": 5738
    },
    {
      "epoch": 0.4020068822401429,
      "grad_norm": 3.6126298904418945,
      "learning_rate": 5.983705078809106e-05,
      "loss": 1.0062,
      "num_input_tokens_seen": 92343816,
      "step": 5739
    },
    {
      "epoch": 0.4020769304858721,
      "grad_norm": 4.304609298706055,
      "learning_rate": 5.983005253940456e-05,
      "loss": 1.0691,
      "num_input_tokens_seen": 92358872,
      "step": 5740
    },
    {
      "epoch": 0.4021469787316014,
      "grad_norm": 4.121729850769043,
      "learning_rate": 5.982305429071805e-05,
      "loss": 1.1047,
      "num_input_tokens_seen": 92374960,
      "step": 5741
    },
    {
      "epoch": 0.40221702697733064,
      "grad_norm": 4.137178421020508,
      "learning_rate": 5.9816056042031534e-05,
      "loss": 1.0809,
      "num_input_tokens_seen": 92391344,
      "step": 5742
    },
    {
      "epoch": 0.4022870752230599,
      "grad_norm": 8.537243843078613,
      "learning_rate": 5.9809057793345016e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 92406096,
      "step": 5743
    },
    {
      "epoch": 0.4023571234687891,
      "grad_norm": 4.613489627838135,
      "learning_rate": 5.9802059544658505e-05,
      "loss": 1.126,
      "num_input_tokens_seen": 92422480,
      "step": 5744
    },
    {
      "epoch": 0.40242717171451836,
      "grad_norm": 4.812812805175781,
      "learning_rate": 5.9795061295971987e-05,
      "loss": 1.2987,
      "num_input_tokens_seen": 92437472,
      "step": 5745
    },
    {
      "epoch": 0.4024972199602476,
      "grad_norm": 4.785153865814209,
      "learning_rate": 5.978806304728546e-05,
      "loss": 0.9517,
      "num_input_tokens_seen": 92452560,
      "step": 5746
    },
    {
      "epoch": 0.4025672682059769,
      "grad_norm": 4.450865268707275,
      "learning_rate": 5.978106479859895e-05,
      "loss": 0.9144,
      "num_input_tokens_seen": 92468312,
      "step": 5747
    },
    {
      "epoch": 0.40263731645170614,
      "grad_norm": 4.854867935180664,
      "learning_rate": 5.9774066549912446e-05,
      "loss": 1.1128,
      "num_input_tokens_seen": 92484616,
      "step": 5748
    },
    {
      "epoch": 0.40270736469743534,
      "grad_norm": 4.00141716003418,
      "learning_rate": 5.976706830122593e-05,
      "loss": 1.1298,
      "num_input_tokens_seen": 92500568,
      "step": 5749
    },
    {
      "epoch": 0.4027774129431646,
      "grad_norm": 3.819101333618164,
      "learning_rate": 5.976007005253941e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 92515952,
      "step": 5750
    },
    {
      "epoch": 0.40284746118889386,
      "grad_norm": 3.7954423427581787,
      "learning_rate": 5.97530718038529e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 92532336,
      "step": 5751
    },
    {
      "epoch": 0.4029175094346231,
      "grad_norm": 4.081971645355225,
      "learning_rate": 5.974607355516638e-05,
      "loss": 1.1075,
      "num_input_tokens_seen": 92548720,
      "step": 5752
    },
    {
      "epoch": 0.4029875576803523,
      "grad_norm": 3.834063768386841,
      "learning_rate": 5.9739075306479855e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 92564648,
      "step": 5753
    },
    {
      "epoch": 0.4030576059260816,
      "grad_norm": 3.6766366958618164,
      "learning_rate": 5.973207705779336e-05,
      "loss": 1.1064,
      "num_input_tokens_seen": 92581032,
      "step": 5754
    },
    {
      "epoch": 0.40312765417181085,
      "grad_norm": 4.423589706420898,
      "learning_rate": 5.972507880910684e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 92596520,
      "step": 5755
    },
    {
      "epoch": 0.4031977024175401,
      "grad_norm": 4.0820207595825195,
      "learning_rate": 5.971808056042031e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 92612448,
      "step": 5756
    },
    {
      "epoch": 0.4032677506632693,
      "grad_norm": 6.634023189544678,
      "learning_rate": 5.97110823117338e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 92627736,
      "step": 5757
    },
    {
      "epoch": 0.40333779890899857,
      "grad_norm": 3.517611026763916,
      "learning_rate": 5.970408406304729e-05,
      "loss": 0.9292,
      "num_input_tokens_seen": 92643216,
      "step": 5758
    },
    {
      "epoch": 0.4034078471547278,
      "grad_norm": 3.7658562660217285,
      "learning_rate": 5.969708581436077e-05,
      "loss": 1.052,
      "num_input_tokens_seen": 92659248,
      "step": 5759
    },
    {
      "epoch": 0.4034778954004571,
      "grad_norm": 4.128793716430664,
      "learning_rate": 5.9690087565674254e-05,
      "loss": 1.1067,
      "num_input_tokens_seen": 92675632,
      "step": 5760
    },
    {
      "epoch": 0.4035479436461863,
      "grad_norm": 4.200130939483643,
      "learning_rate": 5.968308931698775e-05,
      "loss": 0.9264,
      "num_input_tokens_seen": 92691408,
      "step": 5761
    },
    {
      "epoch": 0.40361799189191555,
      "grad_norm": 4.131740093231201,
      "learning_rate": 5.967609106830123e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 92706568,
      "step": 5762
    },
    {
      "epoch": 0.4036880401376448,
      "grad_norm": 4.074241638183594,
      "learning_rate": 5.96690928196147e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 92722952,
      "step": 5763
    },
    {
      "epoch": 0.40375808838337407,
      "grad_norm": 4.351722717285156,
      "learning_rate": 5.9662094570928195e-05,
      "loss": 0.8518,
      "num_input_tokens_seen": 92739248,
      "step": 5764
    },
    {
      "epoch": 0.4038281366291033,
      "grad_norm": 4.0495734214782715,
      "learning_rate": 5.965509632224169e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 92755040,
      "step": 5765
    },
    {
      "epoch": 0.40389818487483253,
      "grad_norm": 3.881098747253418,
      "learning_rate": 5.964809807355517e-05,
      "loss": 0.9899,
      "num_input_tokens_seen": 92771424,
      "step": 5766
    },
    {
      "epoch": 0.4039682331205618,
      "grad_norm": 3.964268922805786,
      "learning_rate": 5.9641099824868654e-05,
      "loss": 1.1636,
      "num_input_tokens_seen": 92787808,
      "step": 5767
    },
    {
      "epoch": 0.40403828136629105,
      "grad_norm": 4.126365661621094,
      "learning_rate": 5.963410157618214e-05,
      "loss": 1.273,
      "num_input_tokens_seen": 92804192,
      "step": 5768
    },
    {
      "epoch": 0.40410832961202026,
      "grad_norm": 4.142693519592285,
      "learning_rate": 5.9627103327495625e-05,
      "loss": 1.2161,
      "num_input_tokens_seen": 92819920,
      "step": 5769
    },
    {
      "epoch": 0.4041783778577495,
      "grad_norm": 4.906876087188721,
      "learning_rate": 5.962010507880912e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 92836304,
      "step": 5770
    },
    {
      "epoch": 0.4042484261034788,
      "grad_norm": 4.597287654876709,
      "learning_rate": 5.96131068301226e-05,
      "loss": 1.13,
      "num_input_tokens_seen": 92852144,
      "step": 5771
    },
    {
      "epoch": 0.40431847434920803,
      "grad_norm": 3.525669813156128,
      "learning_rate": 5.9606108581436084e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 92867944,
      "step": 5772
    },
    {
      "epoch": 0.40438852259493724,
      "grad_norm": 4.908353328704834,
      "learning_rate": 5.9599110332749565e-05,
      "loss": 1.2495,
      "num_input_tokens_seen": 92884328,
      "step": 5773
    },
    {
      "epoch": 0.4044585708406665,
      "grad_norm": 4.31436824798584,
      "learning_rate": 5.959211208406305e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 92900152,
      "step": 5774
    },
    {
      "epoch": 0.40452861908639576,
      "grad_norm": 3.652494430541992,
      "learning_rate": 5.9585113835376536e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 92916416,
      "step": 5775
    },
    {
      "epoch": 0.404598667332125,
      "grad_norm": 3.9569268226623535,
      "learning_rate": 5.957811558669002e-05,
      "loss": 1.0284,
      "num_input_tokens_seen": 92932088,
      "step": 5776
    },
    {
      "epoch": 0.4046687155778542,
      "grad_norm": 4.301011562347412,
      "learning_rate": 5.95711173380035e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 92948472,
      "step": 5777
    },
    {
      "epoch": 0.4047387638235835,
      "grad_norm": 4.0318474769592285,
      "learning_rate": 5.9564119089316995e-05,
      "loss": 1.0636,
      "num_input_tokens_seen": 92964856,
      "step": 5778
    },
    {
      "epoch": 0.40480881206931274,
      "grad_norm": 4.05795955657959,
      "learning_rate": 5.955712084063048e-05,
      "loss": 0.9947,
      "num_input_tokens_seen": 92980040,
      "step": 5779
    },
    {
      "epoch": 0.404878860315042,
      "grad_norm": 5.828601837158203,
      "learning_rate": 5.9550122591943945e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 92996424,
      "step": 5780
    },
    {
      "epoch": 0.4049489085607712,
      "grad_norm": 3.6617836952209473,
      "learning_rate": 5.954312434325745e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 93012400,
      "step": 5781
    },
    {
      "epoch": 0.40501895680650046,
      "grad_norm": 3.8668923377990723,
      "learning_rate": 5.953612609457093e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 93028176,
      "step": 5782
    },
    {
      "epoch": 0.4050890050522297,
      "grad_norm": 6.206475257873535,
      "learning_rate": 5.952912784588441e-05,
      "loss": 1.133,
      "num_input_tokens_seen": 93044560,
      "step": 5783
    },
    {
      "epoch": 0.405159053297959,
      "grad_norm": 3.6768481731414795,
      "learning_rate": 5.952212959719789e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 93060944,
      "step": 5784
    },
    {
      "epoch": 0.40522910154368824,
      "grad_norm": 3.8317768573760986,
      "learning_rate": 5.95151313485114e-05,
      "loss": 1.1,
      "num_input_tokens_seen": 93077328,
      "step": 5785
    },
    {
      "epoch": 0.40529914978941745,
      "grad_norm": 5.8504252433776855,
      "learning_rate": 5.950813309982487e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 93092016,
      "step": 5786
    },
    {
      "epoch": 0.4053691980351467,
      "grad_norm": 3.6718640327453613,
      "learning_rate": 5.9501134851138365e-05,
      "loss": 0.9614,
      "num_input_tokens_seen": 93108400,
      "step": 5787
    },
    {
      "epoch": 0.40543924628087596,
      "grad_norm": 4.16236686706543,
      "learning_rate": 5.949413660245185e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 93124784,
      "step": 5788
    },
    {
      "epoch": 0.4055092945266052,
      "grad_norm": 4.1938958168029785,
      "learning_rate": 5.948713835376533e-05,
      "loss": 1.241,
      "num_input_tokens_seen": 93141168,
      "step": 5789
    },
    {
      "epoch": 0.40557934277233443,
      "grad_norm": 3.7515947818756104,
      "learning_rate": 5.948014010507881e-05,
      "loss": 1.0706,
      "num_input_tokens_seen": 93157552,
      "step": 5790
    },
    {
      "epoch": 0.4056493910180637,
      "grad_norm": 3.710805654525757,
      "learning_rate": 5.94731418563923e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 93173936,
      "step": 5791
    },
    {
      "epoch": 0.40571943926379295,
      "grad_norm": 3.4491820335388184,
      "learning_rate": 5.9466143607705794e-05,
      "loss": 1.0282,
      "num_input_tokens_seen": 93190320,
      "step": 5792
    },
    {
      "epoch": 0.4057894875095222,
      "grad_norm": 3.8709781169891357,
      "learning_rate": 5.945914535901926e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 93206648,
      "step": 5793
    },
    {
      "epoch": 0.4058595357552514,
      "grad_norm": 5.104569435119629,
      "learning_rate": 5.945214711033276e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 93223032,
      "step": 5794
    },
    {
      "epoch": 0.40592958400098067,
      "grad_norm": 4.294493675231934,
      "learning_rate": 5.944514886164625e-05,
      "loss": 1.0673,
      "num_input_tokens_seen": 93239344,
      "step": 5795
    },
    {
      "epoch": 0.40599963224670993,
      "grad_norm": 4.654513835906982,
      "learning_rate": 5.943815061295972e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 93255104,
      "step": 5796
    },
    {
      "epoch": 0.4060696804924392,
      "grad_norm": 4.339935779571533,
      "learning_rate": 5.9431152364273204e-05,
      "loss": 0.982,
      "num_input_tokens_seen": 93270448,
      "step": 5797
    },
    {
      "epoch": 0.4061397287381684,
      "grad_norm": 3.6498191356658936,
      "learning_rate": 5.942415411558669e-05,
      "loss": 0.8815,
      "num_input_tokens_seen": 93286672,
      "step": 5798
    },
    {
      "epoch": 0.40620977698389765,
      "grad_norm": 3.9488580226898193,
      "learning_rate": 5.9417155866900174e-05,
      "loss": 1.2938,
      "num_input_tokens_seen": 93302256,
      "step": 5799
    },
    {
      "epoch": 0.4062798252296269,
      "grad_norm": 3.9446182250976562,
      "learning_rate": 5.9410157618213656e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 93318640,
      "step": 5800
    },
    {
      "epoch": 0.4062798252296269,
      "eval_loss": 1.1256848573684692,
      "eval_runtime": 0.2106,
      "eval_samples_per_second": 4.748,
      "eval_steps_per_second": 4.748,
      "num_input_tokens_seen": 93318640,
      "step": 5800
    },
    {
      "epoch": 0.40634987347535617,
      "grad_norm": 3.986703872680664,
      "learning_rate": 5.940315936952715e-05,
      "loss": 1.1783,
      "num_input_tokens_seen": 93335024,
      "step": 5801
    },
    {
      "epoch": 0.4064199217210854,
      "grad_norm": 3.56948184967041,
      "learning_rate": 5.9396161120840647e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 93351408,
      "step": 5802
    },
    {
      "epoch": 0.40648996996681463,
      "grad_norm": 3.9857194423675537,
      "learning_rate": 5.9389162872154115e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 93367728,
      "step": 5803
    },
    {
      "epoch": 0.4065600182125439,
      "grad_norm": 5.165848731994629,
      "learning_rate": 5.93821646234676e-05,
      "loss": 0.9764,
      "num_input_tokens_seen": 93384112,
      "step": 5804
    },
    {
      "epoch": 0.40663006645827315,
      "grad_norm": 3.742520809173584,
      "learning_rate": 5.9375166374781085e-05,
      "loss": 1.0802,
      "num_input_tokens_seen": 93400072,
      "step": 5805
    },
    {
      "epoch": 0.40670011470400236,
      "grad_norm": 4.13803768157959,
      "learning_rate": 5.936816812609457e-05,
      "loss": 1.1845,
      "num_input_tokens_seen": 93416016,
      "step": 5806
    },
    {
      "epoch": 0.4067701629497316,
      "grad_norm": 4.530385494232178,
      "learning_rate": 5.936116987740805e-05,
      "loss": 1.1034,
      "num_input_tokens_seen": 93432400,
      "step": 5807
    },
    {
      "epoch": 0.4068402111954609,
      "grad_norm": 4.162608623504639,
      "learning_rate": 5.935417162872156e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 93448360,
      "step": 5808
    },
    {
      "epoch": 0.40691025944119014,
      "grad_norm": 3.5075366497039795,
      "learning_rate": 5.934717338003504e-05,
      "loss": 0.8568,
      "num_input_tokens_seen": 93464744,
      "step": 5809
    },
    {
      "epoch": 0.40698030768691934,
      "grad_norm": 4.963081359863281,
      "learning_rate": 5.934017513134851e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 93480448,
      "step": 5810
    },
    {
      "epoch": 0.4070503559326486,
      "grad_norm": 3.83306622505188,
      "learning_rate": 5.9333176882662e-05,
      "loss": 1.1558,
      "num_input_tokens_seen": 93496832,
      "step": 5811
    },
    {
      "epoch": 0.40712040417837786,
      "grad_norm": 3.878345489501953,
      "learning_rate": 5.932617863397549e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 93513216,
      "step": 5812
    },
    {
      "epoch": 0.4071904524241071,
      "grad_norm": 4.1416192054748535,
      "learning_rate": 5.931918038528897e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 93529504,
      "step": 5813
    },
    {
      "epoch": 0.4072605006698363,
      "grad_norm": 5.459712028503418,
      "learning_rate": 5.9312182136602455e-05,
      "loss": 1.0816,
      "num_input_tokens_seen": 93545888,
      "step": 5814
    },
    {
      "epoch": 0.4073305489155656,
      "grad_norm": 5.8356852531433105,
      "learning_rate": 5.930518388791595e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 93562272,
      "step": 5815
    },
    {
      "epoch": 0.40740059716129484,
      "grad_norm": 6.23671817779541,
      "learning_rate": 5.929818563922943e-05,
      "loss": 1.1929,
      "num_input_tokens_seen": 93578656,
      "step": 5816
    },
    {
      "epoch": 0.4074706454070241,
      "grad_norm": 3.6252057552337646,
      "learning_rate": 5.9291187390542914e-05,
      "loss": 1.1491,
      "num_input_tokens_seen": 93594816,
      "step": 5817
    },
    {
      "epoch": 0.40754069365275336,
      "grad_norm": 4.245891571044922,
      "learning_rate": 5.9284189141856396e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 93611200,
      "step": 5818
    },
    {
      "epoch": 0.40761074189848256,
      "grad_norm": 4.052443504333496,
      "learning_rate": 5.9277190893169885e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 93627080,
      "step": 5819
    },
    {
      "epoch": 0.4076807901442118,
      "grad_norm": 4.427778720855713,
      "learning_rate": 5.927019264448336e-05,
      "loss": 0.9268,
      "num_input_tokens_seen": 93643464,
      "step": 5820
    },
    {
      "epoch": 0.4077508383899411,
      "grad_norm": 4.1961541175842285,
      "learning_rate": 5.926319439579685e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 93659088,
      "step": 5821
    },
    {
      "epoch": 0.40782088663567034,
      "grad_norm": 4.001824378967285,
      "learning_rate": 5.9256196147110344e-05,
      "loss": 1.1674,
      "num_input_tokens_seen": 93674928,
      "step": 5822
    },
    {
      "epoch": 0.40789093488139955,
      "grad_norm": 4.010315895080566,
      "learning_rate": 5.924919789842381e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 93690840,
      "step": 5823
    },
    {
      "epoch": 0.4079609831271288,
      "grad_norm": 3.771390438079834,
      "learning_rate": 5.9242199649737294e-05,
      "loss": 1.1374,
      "num_input_tokens_seen": 93707224,
      "step": 5824
    },
    {
      "epoch": 0.40803103137285807,
      "grad_norm": 3.920438051223755,
      "learning_rate": 5.92352014010508e-05,
      "loss": 0.9206,
      "num_input_tokens_seen": 93722536,
      "step": 5825
    },
    {
      "epoch": 0.4081010796185873,
      "grad_norm": 4.679770469665527,
      "learning_rate": 5.922820315236428e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 93738736,
      "step": 5826
    },
    {
      "epoch": 0.40817112786431653,
      "grad_norm": 4.082173824310303,
      "learning_rate": 5.922120490367776e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 93755120,
      "step": 5827
    },
    {
      "epoch": 0.4082411761100458,
      "grad_norm": 4.076204776763916,
      "learning_rate": 5.921420665499124e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 93771504,
      "step": 5828
    },
    {
      "epoch": 0.40831122435577505,
      "grad_norm": 4.721165180206299,
      "learning_rate": 5.920720840630474e-05,
      "loss": 1.237,
      "num_input_tokens_seen": 93787888,
      "step": 5829
    },
    {
      "epoch": 0.4083812726015043,
      "grad_norm": 4.313892841339111,
      "learning_rate": 5.9200210157618205e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 93804272,
      "step": 5830
    },
    {
      "epoch": 0.4084513208472335,
      "grad_norm": 4.506958961486816,
      "learning_rate": 5.91932119089317e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 93820416,
      "step": 5831
    },
    {
      "epoch": 0.40852136909296277,
      "grad_norm": 4.6436991691589355,
      "learning_rate": 5.9186213660245196e-05,
      "loss": 0.8856,
      "num_input_tokens_seen": 93836648,
      "step": 5832
    },
    {
      "epoch": 0.40859141733869203,
      "grad_norm": 3.9535446166992188,
      "learning_rate": 5.917921541155868e-05,
      "loss": 1.1285,
      "num_input_tokens_seen": 93853032,
      "step": 5833
    },
    {
      "epoch": 0.4086614655844213,
      "grad_norm": 6.974640846252441,
      "learning_rate": 5.917221716287216e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 93868640,
      "step": 5834
    },
    {
      "epoch": 0.4087315138301505,
      "grad_norm": 4.3199262619018555,
      "learning_rate": 5.916521891418565e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 93885024,
      "step": 5835
    },
    {
      "epoch": 0.40880156207587975,
      "grad_norm": 3.892812967300415,
      "learning_rate": 5.915822066549913e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 93901408,
      "step": 5836
    },
    {
      "epoch": 0.408871610321609,
      "grad_norm": 4.434093952178955,
      "learning_rate": 5.915122241681261e-05,
      "loss": 1.1484,
      "num_input_tokens_seen": 93917632,
      "step": 5837
    },
    {
      "epoch": 0.4089416585673383,
      "grad_norm": 3.960766553878784,
      "learning_rate": 5.914422416812611e-05,
      "loss": 0.994,
      "num_input_tokens_seen": 93934016,
      "step": 5838
    },
    {
      "epoch": 0.4090117068130675,
      "grad_norm": 6.450897693634033,
      "learning_rate": 5.913722591943959e-05,
      "loss": 1.1364,
      "num_input_tokens_seen": 93950328,
      "step": 5839
    },
    {
      "epoch": 0.40908175505879674,
      "grad_norm": 4.3362956047058105,
      "learning_rate": 5.913022767075307e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 93966712,
      "step": 5840
    },
    {
      "epoch": 0.409151803304526,
      "grad_norm": 4.250185966491699,
      "learning_rate": 5.912322942206655e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 93982840,
      "step": 5841
    },
    {
      "epoch": 0.40922185155025526,
      "grad_norm": 5.399633407592773,
      "learning_rate": 5.911623117338004e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 93998192,
      "step": 5842
    },
    {
      "epoch": 0.40929189979598446,
      "grad_norm": 3.926515579223633,
      "learning_rate": 5.910923292469352e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 94014296,
      "step": 5843
    },
    {
      "epoch": 0.4093619480417137,
      "grad_norm": 3.9847402572631836,
      "learning_rate": 5.9102234676007005e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 94030520,
      "step": 5844
    },
    {
      "epoch": 0.409431996287443,
      "grad_norm": 4.106837272644043,
      "learning_rate": 5.9095236427320486e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 94046512,
      "step": 5845
    },
    {
      "epoch": 0.40950204453317224,
      "grad_norm": 5.044075965881348,
      "learning_rate": 5.908823817863398e-05,
      "loss": 0.8541,
      "num_input_tokens_seen": 94061536,
      "step": 5846
    },
    {
      "epoch": 0.40957209277890144,
      "grad_norm": 7.444840908050537,
      "learning_rate": 5.908123992994745e-05,
      "loss": 0.9811,
      "num_input_tokens_seen": 94077920,
      "step": 5847
    },
    {
      "epoch": 0.4096421410246307,
      "grad_norm": 4.676487445831299,
      "learning_rate": 5.907424168126095e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 94094040,
      "step": 5848
    },
    {
      "epoch": 0.40971218927035996,
      "grad_norm": 3.945162057876587,
      "learning_rate": 5.9067243432574434e-05,
      "loss": 1.0101,
      "num_input_tokens_seen": 94110424,
      "step": 5849
    },
    {
      "epoch": 0.4097822375160892,
      "grad_norm": 3.9881343841552734,
      "learning_rate": 5.9060245183887916e-05,
      "loss": 1.2385,
      "num_input_tokens_seen": 94126808,
      "step": 5850
    },
    {
      "epoch": 0.4098522857618185,
      "grad_norm": 3.5962657928466797,
      "learning_rate": 5.90532469352014e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 94143192,
      "step": 5851
    },
    {
      "epoch": 0.4099223340075477,
      "grad_norm": 5.755387783050537,
      "learning_rate": 5.904624868651491e-05,
      "loss": 1.1936,
      "num_input_tokens_seen": 94158728,
      "step": 5852
    },
    {
      "epoch": 0.40999238225327694,
      "grad_norm": 4.999855995178223,
      "learning_rate": 5.9039250437828375e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 94174736,
      "step": 5853
    },
    {
      "epoch": 0.4100624304990062,
      "grad_norm": 5.452749729156494,
      "learning_rate": 5.903225218914186e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 94191120,
      "step": 5854
    },
    {
      "epoch": 0.41013247874473546,
      "grad_norm": 4.211399078369141,
      "learning_rate": 5.902525394045535e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 94207504,
      "step": 5855
    },
    {
      "epoch": 0.41020252699046467,
      "grad_norm": 3.3774921894073486,
      "learning_rate": 5.9018255691768834e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 94223888,
      "step": 5856
    },
    {
      "epoch": 0.4102725752361939,
      "grad_norm": 6.850539207458496,
      "learning_rate": 5.9011257443082316e-05,
      "loss": 1.0313,
      "num_input_tokens_seen": 94240120,
      "step": 5857
    },
    {
      "epoch": 0.4103426234819232,
      "grad_norm": 3.2469308376312256,
      "learning_rate": 5.9004259194395804e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 94256488,
      "step": 5858
    },
    {
      "epoch": 0.41041267172765244,
      "grad_norm": 5.0675201416015625,
      "learning_rate": 5.89972609457093e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 94272048,
      "step": 5859
    },
    {
      "epoch": 0.41048271997338165,
      "grad_norm": 3.8971400260925293,
      "learning_rate": 5.899026269702277e-05,
      "loss": 1.2323,
      "num_input_tokens_seen": 94287984,
      "step": 5860
    },
    {
      "epoch": 0.4105527682191109,
      "grad_norm": 3.8453164100646973,
      "learning_rate": 5.898326444833625e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 94303976,
      "step": 5861
    },
    {
      "epoch": 0.41062281646484017,
      "grad_norm": 3.7564680576324463,
      "learning_rate": 5.897626619964975e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 94319552,
      "step": 5862
    },
    {
      "epoch": 0.4106928647105694,
      "grad_norm": 4.541357517242432,
      "learning_rate": 5.896926795096323e-05,
      "loss": 1.2374,
      "num_input_tokens_seen": 94335936,
      "step": 5863
    },
    {
      "epoch": 0.41076291295629863,
      "grad_norm": 4.841330051422119,
      "learning_rate": 5.896226970227671e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 94351784,
      "step": 5864
    },
    {
      "epoch": 0.4108329612020279,
      "grad_norm": 3.9247653484344482,
      "learning_rate": 5.89552714535902e-05,
      "loss": 1.0169,
      "num_input_tokens_seen": 94368080,
      "step": 5865
    },
    {
      "epoch": 0.41090300944775715,
      "grad_norm": 4.763803958892822,
      "learning_rate": 5.894827320490368e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 94384056,
      "step": 5866
    },
    {
      "epoch": 0.4109730576934864,
      "grad_norm": 5.141749858856201,
      "learning_rate": 5.894127495621716e-05,
      "loss": 1.0001,
      "num_input_tokens_seen": 94400440,
      "step": 5867
    },
    {
      "epoch": 0.4110431059392156,
      "grad_norm": 3.360656261444092,
      "learning_rate": 5.8934276707530656e-05,
      "loss": 0.988,
      "num_input_tokens_seen": 94416824,
      "step": 5868
    },
    {
      "epoch": 0.4111131541849449,
      "grad_norm": 4.226006507873535,
      "learning_rate": 5.892727845884415e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 94433208,
      "step": 5869
    },
    {
      "epoch": 0.41118320243067413,
      "grad_norm": 6.924234390258789,
      "learning_rate": 5.892028021015762e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 94448848,
      "step": 5870
    },
    {
      "epoch": 0.4112532506764034,
      "grad_norm": 3.643950939178467,
      "learning_rate": 5.89132819614711e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 94464720,
      "step": 5871
    },
    {
      "epoch": 0.4113232989221326,
      "grad_norm": 3.3077268600463867,
      "learning_rate": 5.890628371278459e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 94481104,
      "step": 5872
    },
    {
      "epoch": 0.41139334716786186,
      "grad_norm": 3.9556264877319336,
      "learning_rate": 5.889928546409807e-05,
      "loss": 0.9313,
      "num_input_tokens_seen": 94496944,
      "step": 5873
    },
    {
      "epoch": 0.4114633954135911,
      "grad_norm": 7.0952606201171875,
      "learning_rate": 5.8892287215411554e-05,
      "loss": 1.3029,
      "num_input_tokens_seen": 94511336,
      "step": 5874
    },
    {
      "epoch": 0.4115334436593204,
      "grad_norm": 3.946803569793701,
      "learning_rate": 5.888528896672505e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 94526904,
      "step": 5875
    },
    {
      "epoch": 0.4116034919050496,
      "grad_norm": 4.165796756744385,
      "learning_rate": 5.8878290718038545e-05,
      "loss": 1.2249,
      "num_input_tokens_seen": 94543024,
      "step": 5876
    },
    {
      "epoch": 0.41167354015077884,
      "grad_norm": 3.9492764472961426,
      "learning_rate": 5.887129246935201e-05,
      "loss": 1.0304,
      "num_input_tokens_seen": 94558744,
      "step": 5877
    },
    {
      "epoch": 0.4117435883965081,
      "grad_norm": 4.144934177398682,
      "learning_rate": 5.8864294220665495e-05,
      "loss": 1.0468,
      "num_input_tokens_seen": 94574288,
      "step": 5878
    },
    {
      "epoch": 0.41181363664223736,
      "grad_norm": 3.7620224952697754,
      "learning_rate": 5.8857295971979e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 94590672,
      "step": 5879
    },
    {
      "epoch": 0.41188368488796656,
      "grad_norm": 4.788266181945801,
      "learning_rate": 5.885029772329247e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 94606792,
      "step": 5880
    },
    {
      "epoch": 0.4119537331336958,
      "grad_norm": 4.383455276489258,
      "learning_rate": 5.8843299474605954e-05,
      "loss": 0.9611,
      "num_input_tokens_seen": 94623176,
      "step": 5881
    },
    {
      "epoch": 0.4120237813794251,
      "grad_norm": 7.097373962402344,
      "learning_rate": 5.883630122591944e-05,
      "loss": 1.1559,
      "num_input_tokens_seen": 94638952,
      "step": 5882
    },
    {
      "epoch": 0.41209382962515434,
      "grad_norm": 5.4228901863098145,
      "learning_rate": 5.882930297723294e-05,
      "loss": 1.2569,
      "num_input_tokens_seen": 94655224,
      "step": 5883
    },
    {
      "epoch": 0.41216387787088354,
      "grad_norm": 3.792999267578125,
      "learning_rate": 5.882230472854642e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 94671608,
      "step": 5884
    },
    {
      "epoch": 0.4122339261166128,
      "grad_norm": 5.401544094085693,
      "learning_rate": 5.88153064798599e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 94687048,
      "step": 5885
    },
    {
      "epoch": 0.41230397436234206,
      "grad_norm": 3.2229812145233154,
      "learning_rate": 5.880830823117339e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 94703432,
      "step": 5886
    },
    {
      "epoch": 0.4123740226080713,
      "grad_norm": 4.5977277755737305,
      "learning_rate": 5.8801309982486865e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 94719152,
      "step": 5887
    },
    {
      "epoch": 0.4124440708538006,
      "grad_norm": 3.436765670776367,
      "learning_rate": 5.879431173380035e-05,
      "loss": 0.8353,
      "num_input_tokens_seen": 94735536,
      "step": 5888
    },
    {
      "epoch": 0.4125141190995298,
      "grad_norm": 3.4720351696014404,
      "learning_rate": 5.878731348511385e-05,
      "loss": 1.0329,
      "num_input_tokens_seen": 94751840,
      "step": 5889
    },
    {
      "epoch": 0.41258416734525905,
      "grad_norm": 3.621783971786499,
      "learning_rate": 5.878031523642732e-05,
      "loss": 1.0355,
      "num_input_tokens_seen": 94768224,
      "step": 5890
    },
    {
      "epoch": 0.4126542155909883,
      "grad_norm": 4.453585624694824,
      "learning_rate": 5.87733169877408e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 94783672,
      "step": 5891
    },
    {
      "epoch": 0.41272426383671756,
      "grad_norm": 4.191892147064209,
      "learning_rate": 5.8766318739054294e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 94800056,
      "step": 5892
    },
    {
      "epoch": 0.41279431208244677,
      "grad_norm": 3.5963308811187744,
      "learning_rate": 5.875932049036778e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 94816440,
      "step": 5893
    },
    {
      "epoch": 0.412864360328176,
      "grad_norm": 3.904618978500366,
      "learning_rate": 5.8752322241681265e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 94832824,
      "step": 5894
    },
    {
      "epoch": 0.4129344085739053,
      "grad_norm": 3.650961399078369,
      "learning_rate": 5.8745323992994747e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 94849208,
      "step": 5895
    },
    {
      "epoch": 0.41300445681963455,
      "grad_norm": 4.438238143920898,
      "learning_rate": 5.873832574430824e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 94865592,
      "step": 5896
    },
    {
      "epoch": 0.41307450506536375,
      "grad_norm": 3.6906070709228516,
      "learning_rate": 5.873132749562171e-05,
      "loss": 1.1228,
      "num_input_tokens_seen": 94881976,
      "step": 5897
    },
    {
      "epoch": 0.413144553311093,
      "grad_norm": 5.111722469329834,
      "learning_rate": 5.872432924693519e-05,
      "loss": 1.2678,
      "num_input_tokens_seen": 94898360,
      "step": 5898
    },
    {
      "epoch": 0.41321460155682227,
      "grad_norm": 3.7190804481506348,
      "learning_rate": 5.871733099824869e-05,
      "loss": 1.152,
      "num_input_tokens_seen": 94914512,
      "step": 5899
    },
    {
      "epoch": 0.41328464980255153,
      "grad_norm": 3.779918670654297,
      "learning_rate": 5.871033274956218e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 94930816,
      "step": 5900
    },
    {
      "epoch": 0.41335469804828073,
      "grad_norm": 3.5921852588653564,
      "learning_rate": 5.8703334500875665e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 94947200,
      "step": 5901
    },
    {
      "epoch": 0.41342474629401,
      "grad_norm": 4.024271011352539,
      "learning_rate": 5.8696336252189146e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 94963584,
      "step": 5902
    },
    {
      "epoch": 0.41349479453973925,
      "grad_norm": 4.798417091369629,
      "learning_rate": 5.8689338003502635e-05,
      "loss": 1.1399,
      "num_input_tokens_seen": 94979968,
      "step": 5903
    },
    {
      "epoch": 0.4135648427854685,
      "grad_norm": 3.5821495056152344,
      "learning_rate": 5.868233975481612e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 94996320,
      "step": 5904
    },
    {
      "epoch": 0.4136348910311977,
      "grad_norm": 3.998082399368286,
      "learning_rate": 5.867534150612959e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 95012704,
      "step": 5905
    },
    {
      "epoch": 0.413704939276927,
      "grad_norm": 3.6389498710632324,
      "learning_rate": 5.8668343257443094e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 95029088,
      "step": 5906
    },
    {
      "epoch": 0.41377498752265623,
      "grad_norm": 4.137228012084961,
      "learning_rate": 5.8661345008756576e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 95045472,
      "step": 5907
    },
    {
      "epoch": 0.4138450357683855,
      "grad_norm": 4.181145668029785,
      "learning_rate": 5.865434676007006e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 95061856,
      "step": 5908
    },
    {
      "epoch": 0.4139150840141147,
      "grad_norm": 3.759474754333496,
      "learning_rate": 5.864734851138354e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 95077904,
      "step": 5909
    },
    {
      "epoch": 0.41398513225984396,
      "grad_norm": 4.474549293518066,
      "learning_rate": 5.864035026269703e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 95094288,
      "step": 5910
    },
    {
      "epoch": 0.4140551805055732,
      "grad_norm": 7.399059295654297,
      "learning_rate": 5.863335201401051e-05,
      "loss": 1.2234,
      "num_input_tokens_seen": 95107872,
      "step": 5911
    },
    {
      "epoch": 0.4141252287513025,
      "grad_norm": 4.018132209777832,
      "learning_rate": 5.862635376532399e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 95124256,
      "step": 5912
    },
    {
      "epoch": 0.4141952769970317,
      "grad_norm": 3.825305223464966,
      "learning_rate": 5.861935551663749e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 95139784,
      "step": 5913
    },
    {
      "epoch": 0.41426532524276094,
      "grad_norm": 3.7942214012145996,
      "learning_rate": 5.861235726795097e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 95155136,
      "step": 5914
    },
    {
      "epoch": 0.4143353734884902,
      "grad_norm": 3.9006733894348145,
      "learning_rate": 5.860535901926444e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 95171520,
      "step": 5915
    },
    {
      "epoch": 0.41440542173421946,
      "grad_norm": 3.688754081726074,
      "learning_rate": 5.859836077057794e-05,
      "loss": 1.1338,
      "num_input_tokens_seen": 95187752,
      "step": 5916
    },
    {
      "epoch": 0.41447546997994866,
      "grad_norm": 3.954989433288574,
      "learning_rate": 5.859136252189142e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 95203864,
      "step": 5917
    },
    {
      "epoch": 0.4145455182256779,
      "grad_norm": 3.733185052871704,
      "learning_rate": 5.85843642732049e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 95220248,
      "step": 5918
    },
    {
      "epoch": 0.4146155664714072,
      "grad_norm": 4.39019775390625,
      "learning_rate": 5.8577366024518385e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 95236632,
      "step": 5919
    },
    {
      "epoch": 0.41468561471713644,
      "grad_norm": 3.716066837310791,
      "learning_rate": 5.857036777583188e-05,
      "loss": 0.9231,
      "num_input_tokens_seen": 95251888,
      "step": 5920
    },
    {
      "epoch": 0.4147556629628657,
      "grad_norm": 3.7525405883789062,
      "learning_rate": 5.856336952714536e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 95268272,
      "step": 5921
    },
    {
      "epoch": 0.4148257112085949,
      "grad_norm": 3.605818033218384,
      "learning_rate": 5.855637127845886e-05,
      "loss": 0.9874,
      "num_input_tokens_seen": 95283880,
      "step": 5922
    },
    {
      "epoch": 0.41489575945432416,
      "grad_norm": 3.878814458847046,
      "learning_rate": 5.854937302977234e-05,
      "loss": 1.1556,
      "num_input_tokens_seen": 95300264,
      "step": 5923
    },
    {
      "epoch": 0.4149658077000534,
      "grad_norm": 4.94001579284668,
      "learning_rate": 5.854237478108582e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 95316648,
      "step": 5924
    },
    {
      "epoch": 0.4150358559457827,
      "grad_norm": 4.213568210601807,
      "learning_rate": 5.85353765323993e-05,
      "loss": 1.2315,
      "num_input_tokens_seen": 95332880,
      "step": 5925
    },
    {
      "epoch": 0.4151059041915119,
      "grad_norm": 4.2475996017456055,
      "learning_rate": 5.8528378283712784e-05,
      "loss": 1.0717,
      "num_input_tokens_seen": 95349176,
      "step": 5926
    },
    {
      "epoch": 0.41517595243724115,
      "grad_norm": 4.237911224365234,
      "learning_rate": 5.852138003502627e-05,
      "loss": 0.93,
      "num_input_tokens_seen": 95365560,
      "step": 5927
    },
    {
      "epoch": 0.4152460006829704,
      "grad_norm": 3.695140838623047,
      "learning_rate": 5.8514381786339755e-05,
      "loss": 0.9561,
      "num_input_tokens_seen": 95381280,
      "step": 5928
    },
    {
      "epoch": 0.41531604892869967,
      "grad_norm": 4.457770347595215,
      "learning_rate": 5.850738353765325e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 95397664,
      "step": 5929
    },
    {
      "epoch": 0.41538609717442887,
      "grad_norm": 4.237982273101807,
      "learning_rate": 5.850038528896673e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 95414048,
      "step": 5930
    },
    {
      "epoch": 0.41545614542015813,
      "grad_norm": 3.5690579414367676,
      "learning_rate": 5.8493387040280214e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 95430432,
      "step": 5931
    },
    {
      "epoch": 0.4155261936658874,
      "grad_norm": 4.665618419647217,
      "learning_rate": 5.8486388791593696e-05,
      "loss": 1.0138,
      "num_input_tokens_seen": 95446816,
      "step": 5932
    },
    {
      "epoch": 0.41559624191161665,
      "grad_norm": 3.76755952835083,
      "learning_rate": 5.8479390542907184e-05,
      "loss": 0.9853,
      "num_input_tokens_seen": 95463200,
      "step": 5933
    },
    {
      "epoch": 0.41566629015734585,
      "grad_norm": 4.855432033538818,
      "learning_rate": 5.8472392294220666e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 95478704,
      "step": 5934
    },
    {
      "epoch": 0.4157363384030751,
      "grad_norm": 4.860055446624756,
      "learning_rate": 5.846539404553415e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 95495088,
      "step": 5935
    },
    {
      "epoch": 0.41580638664880437,
      "grad_norm": 4.268356800079346,
      "learning_rate": 5.845839579684763e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 95511472,
      "step": 5936
    },
    {
      "epoch": 0.41587643489453363,
      "grad_norm": 7.423181533813477,
      "learning_rate": 5.8451397548161125e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 95527008,
      "step": 5937
    },
    {
      "epoch": 0.41594648314026283,
      "grad_norm": 5.740126609802246,
      "learning_rate": 5.844439929947461e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 95542536,
      "step": 5938
    },
    {
      "epoch": 0.4160165313859921,
      "grad_norm": 5.135944366455078,
      "learning_rate": 5.8437401050788096e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 95558688,
      "step": 5939
    },
    {
      "epoch": 0.41608657963172135,
      "grad_norm": 3.879530191421509,
      "learning_rate": 5.843040280210158e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 95575064,
      "step": 5940
    },
    {
      "epoch": 0.4161566278774506,
      "grad_norm": 4.096410274505615,
      "learning_rate": 5.842340455341506e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 95591272,
      "step": 5941
    },
    {
      "epoch": 0.4162266761231798,
      "grad_norm": 4.095829486846924,
      "learning_rate": 5.841640630472854e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 95607656,
      "step": 5942
    },
    {
      "epoch": 0.4162967243689091,
      "grad_norm": 4.076023101806641,
      "learning_rate": 5.840940805604205e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 95623384,
      "step": 5943
    },
    {
      "epoch": 0.41636677261463834,
      "grad_norm": 4.08365535736084,
      "learning_rate": 5.840240980735553e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 95639768,
      "step": 5944
    },
    {
      "epoch": 0.4164368208603676,
      "grad_norm": 4.182791233062744,
      "learning_rate": 5.8395411558669e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 95655928,
      "step": 5945
    },
    {
      "epoch": 0.4165068691060968,
      "grad_norm": 4.3107428550720215,
      "learning_rate": 5.8388413309982495e-05,
      "loss": 0.8607,
      "num_input_tokens_seen": 95672312,
      "step": 5946
    },
    {
      "epoch": 0.41657691735182606,
      "grad_norm": 3.7357101440429688,
      "learning_rate": 5.838141506129598e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 95688696,
      "step": 5947
    },
    {
      "epoch": 0.4166469655975553,
      "grad_norm": 3.9959046840667725,
      "learning_rate": 5.837441681260946e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 95704304,
      "step": 5948
    },
    {
      "epoch": 0.4167170138432846,
      "grad_norm": 4.395400524139404,
      "learning_rate": 5.836741856392295e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 95720688,
      "step": 5949
    },
    {
      "epoch": 0.4167870620890138,
      "grad_norm": 4.998651027679443,
      "learning_rate": 5.836042031523644e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 95737072,
      "step": 5950
    },
    {
      "epoch": 0.41685711033474304,
      "grad_norm": 4.060539245605469,
      "learning_rate": 5.8353422066549925e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 95753456,
      "step": 5951
    },
    {
      "epoch": 0.4169271585804723,
      "grad_norm": 5.286706924438477,
      "learning_rate": 5.834642381786339e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 95769840,
      "step": 5952
    },
    {
      "epoch": 0.41699720682620156,
      "grad_norm": 4.501932144165039,
      "learning_rate": 5.833942556917689e-05,
      "loss": 1.2626,
      "num_input_tokens_seen": 95786224,
      "step": 5953
    },
    {
      "epoch": 0.41706725507193076,
      "grad_norm": 5.144174575805664,
      "learning_rate": 5.833242732049038e-05,
      "loss": 0.8148,
      "num_input_tokens_seen": 95802608,
      "step": 5954
    },
    {
      "epoch": 0.41713730331766,
      "grad_norm": 3.6604678630828857,
      "learning_rate": 5.832542907180385e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 95818992,
      "step": 5955
    },
    {
      "epoch": 0.4172073515633893,
      "grad_norm": 5.387998104095459,
      "learning_rate": 5.831843082311734e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 95835376,
      "step": 5956
    },
    {
      "epoch": 0.41727739980911854,
      "grad_norm": 4.576782703399658,
      "learning_rate": 5.831143257443082e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 95851760,
      "step": 5957
    },
    {
      "epoch": 0.4173474480548478,
      "grad_norm": 5.737542629241943,
      "learning_rate": 5.8304434325744304e-05,
      "loss": 1.039,
      "num_input_tokens_seen": 95868144,
      "step": 5958
    },
    {
      "epoch": 0.417417496300577,
      "grad_norm": 3.515028238296509,
      "learning_rate": 5.8297436077057786e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 95884528,
      "step": 5959
    },
    {
      "epoch": 0.41748754454630627,
      "grad_norm": 3.9339003562927246,
      "learning_rate": 5.8290437828371295e-05,
      "loss": 0.928,
      "num_input_tokens_seen": 95900688,
      "step": 5960
    },
    {
      "epoch": 0.4175575927920355,
      "grad_norm": 3.896474838256836,
      "learning_rate": 5.828343957968477e-05,
      "loss": 1.1249,
      "num_input_tokens_seen": 95917072,
      "step": 5961
    },
    {
      "epoch": 0.4176276410377648,
      "grad_norm": 5.101248264312744,
      "learning_rate": 5.8276441330998245e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 95931976,
      "step": 5962
    },
    {
      "epoch": 0.417697689283494,
      "grad_norm": 7.063873291015625,
      "learning_rate": 5.8269443082311734e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 95948360,
      "step": 5963
    },
    {
      "epoch": 0.41776773752922325,
      "grad_norm": 3.564887762069702,
      "learning_rate": 5.826244483362523e-05,
      "loss": 1.1375,
      "num_input_tokens_seen": 95964744,
      "step": 5964
    },
    {
      "epoch": 0.4178377857749525,
      "grad_norm": 3.821101427078247,
      "learning_rate": 5.82554465849387e-05,
      "loss": 1.252,
      "num_input_tokens_seen": 95980760,
      "step": 5965
    },
    {
      "epoch": 0.41790783402068177,
      "grad_norm": 3.609252691268921,
      "learning_rate": 5.824844833625219e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 95997144,
      "step": 5966
    },
    {
      "epoch": 0.41797788226641097,
      "grad_norm": 4.1750874519348145,
      "learning_rate": 5.824145008756569e-05,
      "loss": 1.119,
      "num_input_tokens_seen": 96012872,
      "step": 5967
    },
    {
      "epoch": 0.41804793051214023,
      "grad_norm": 3.776747465133667,
      "learning_rate": 5.823445183887917e-05,
      "loss": 0.9792,
      "num_input_tokens_seen": 96029168,
      "step": 5968
    },
    {
      "epoch": 0.4181179787578695,
      "grad_norm": 8.143741607666016,
      "learning_rate": 5.822745359019264e-05,
      "loss": 1.0686,
      "num_input_tokens_seen": 96045376,
      "step": 5969
    },
    {
      "epoch": 0.41818802700359875,
      "grad_norm": 4.336330890655518,
      "learning_rate": 5.822045534150614e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 96060904,
      "step": 5970
    },
    {
      "epoch": 0.41825807524932795,
      "grad_norm": 3.731605052947998,
      "learning_rate": 5.821345709281962e-05,
      "loss": 1.1302,
      "num_input_tokens_seen": 96076824,
      "step": 5971
    },
    {
      "epoch": 0.4183281234950572,
      "grad_norm": 3.8380699157714844,
      "learning_rate": 5.82064588441331e-05,
      "loss": 1.1711,
      "num_input_tokens_seen": 96092616,
      "step": 5972
    },
    {
      "epoch": 0.4183981717407865,
      "grad_norm": 3.9088358879089355,
      "learning_rate": 5.8199460595446586e-05,
      "loss": 1.1622,
      "num_input_tokens_seen": 96109000,
      "step": 5973
    },
    {
      "epoch": 0.41846821998651573,
      "grad_norm": 4.0047783851623535,
      "learning_rate": 5.819246234676008e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 96125344,
      "step": 5974
    },
    {
      "epoch": 0.41853826823224494,
      "grad_norm": 3.936495542526245,
      "learning_rate": 5.818546409807356e-05,
      "loss": 1.1566,
      "num_input_tokens_seen": 96141536,
      "step": 5975
    },
    {
      "epoch": 0.4186083164779742,
      "grad_norm": 3.8510451316833496,
      "learning_rate": 5.8178465849387045e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 96157920,
      "step": 5976
    },
    {
      "epoch": 0.41867836472370346,
      "grad_norm": 3.5825259685516357,
      "learning_rate": 5.817146760070053e-05,
      "loss": 1.0491,
      "num_input_tokens_seen": 96173848,
      "step": 5977
    },
    {
      "epoch": 0.4187484129694327,
      "grad_norm": 4.0553717613220215,
      "learning_rate": 5.8164469352014015e-05,
      "loss": 1.0669,
      "num_input_tokens_seen": 96190232,
      "step": 5978
    },
    {
      "epoch": 0.4188184612151619,
      "grad_norm": 4.085362434387207,
      "learning_rate": 5.815747110332749e-05,
      "loss": 1.2832,
      "num_input_tokens_seen": 96205896,
      "step": 5979
    },
    {
      "epoch": 0.4188885094608912,
      "grad_norm": 6.552733421325684,
      "learning_rate": 5.815047285464098e-05,
      "loss": 1.1183,
      "num_input_tokens_seen": 96221784,
      "step": 5980
    },
    {
      "epoch": 0.41895855770662044,
      "grad_norm": 4.052005290985107,
      "learning_rate": 5.8143474605954474e-05,
      "loss": 1.0309,
      "num_input_tokens_seen": 96237472,
      "step": 5981
    },
    {
      "epoch": 0.4190286059523497,
      "grad_norm": 3.9679994583129883,
      "learning_rate": 5.813647635726794e-05,
      "loss": 1.0298,
      "num_input_tokens_seen": 96253616,
      "step": 5982
    },
    {
      "epoch": 0.4190986541980789,
      "grad_norm": 4.879584312438965,
      "learning_rate": 5.812947810858144e-05,
      "loss": 0.9334,
      "num_input_tokens_seen": 96270000,
      "step": 5983
    },
    {
      "epoch": 0.41916870244380816,
      "grad_norm": 4.894060134887695,
      "learning_rate": 5.8122479859894926e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 96286384,
      "step": 5984
    },
    {
      "epoch": 0.4192387506895374,
      "grad_norm": 3.9925336837768555,
      "learning_rate": 5.811548161120841e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 96302200,
      "step": 5985
    },
    {
      "epoch": 0.4193087989352667,
      "grad_norm": 4.043905258178711,
      "learning_rate": 5.810848336252189e-05,
      "loss": 1.0217,
      "num_input_tokens_seen": 96318584,
      "step": 5986
    },
    {
      "epoch": 0.4193788471809959,
      "grad_norm": 4.216322422027588,
      "learning_rate": 5.8101485113835385e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 96334720,
      "step": 5987
    },
    {
      "epoch": 0.41944889542672514,
      "grad_norm": 3.772749900817871,
      "learning_rate": 5.809448686514887e-05,
      "loss": 1.272,
      "num_input_tokens_seen": 96351104,
      "step": 5988
    },
    {
      "epoch": 0.4195189436724544,
      "grad_norm": 3.6716036796569824,
      "learning_rate": 5.8087488616462335e-05,
      "loss": 1.1796,
      "num_input_tokens_seen": 96367488,
      "step": 5989
    },
    {
      "epoch": 0.41958899191818366,
      "grad_norm": 3.9748408794403076,
      "learning_rate": 5.8080490367775844e-05,
      "loss": 1.0994,
      "num_input_tokens_seen": 96383872,
      "step": 5990
    },
    {
      "epoch": 0.4196590401639129,
      "grad_norm": 5.4619269371032715,
      "learning_rate": 5.8073492119089326e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 96400256,
      "step": 5991
    },
    {
      "epoch": 0.4197290884096421,
      "grad_norm": 3.4772391319274902,
      "learning_rate": 5.806649387040281e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 96416640,
      "step": 5992
    },
    {
      "epoch": 0.4197991366553714,
      "grad_norm": 4.003359794616699,
      "learning_rate": 5.805949562171629e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 96432872,
      "step": 5993
    },
    {
      "epoch": 0.41986918490110064,
      "grad_norm": 3.433760166168213,
      "learning_rate": 5.805249737302979e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 96449256,
      "step": 5994
    },
    {
      "epoch": 0.4199392331468299,
      "grad_norm": 4.519425868988037,
      "learning_rate": 5.804549912434326e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 96465456,
      "step": 5995
    },
    {
      "epoch": 0.4200092813925591,
      "grad_norm": 3.8798038959503174,
      "learning_rate": 5.803850087565674e-05,
      "loss": 1.1426,
      "num_input_tokens_seen": 96481840,
      "step": 5996
    },
    {
      "epoch": 0.42007932963828837,
      "grad_norm": 7.4741058349609375,
      "learning_rate": 5.803150262697024e-05,
      "loss": 1.0759,
      "num_input_tokens_seen": 96497160,
      "step": 5997
    },
    {
      "epoch": 0.4201493778840176,
      "grad_norm": 3.6269989013671875,
      "learning_rate": 5.802450437828372e-05,
      "loss": 1.0676,
      "num_input_tokens_seen": 96512672,
      "step": 5998
    },
    {
      "epoch": 0.4202194261297469,
      "grad_norm": 3.6369056701660156,
      "learning_rate": 5.80175061295972e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 96529056,
      "step": 5999
    },
    {
      "epoch": 0.4202894743754761,
      "grad_norm": 6.0609564781188965,
      "learning_rate": 5.801050788091069e-05,
      "loss": 1.3902,
      "num_input_tokens_seen": 96545320,
      "step": 6000
    },
    {
      "epoch": 0.4202894743754761,
      "eval_loss": 1.1260257959365845,
      "eval_runtime": 0.1972,
      "eval_samples_per_second": 5.072,
      "eval_steps_per_second": 5.072,
      "num_input_tokens_seen": 96545320,
      "step": 6000
    },
    {
      "epoch": 0.42035952262120535,
      "grad_norm": 3.939091682434082,
      "learning_rate": 5.800350963222417e-05,
      "loss": 1.2935,
      "num_input_tokens_seen": 96561704,
      "step": 6001
    },
    {
      "epoch": 0.4204295708669346,
      "grad_norm": 4.907895565032959,
      "learning_rate": 5.799651138353765e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 96578088,
      "step": 6002
    },
    {
      "epoch": 0.42049961911266387,
      "grad_norm": 4.598423480987549,
      "learning_rate": 5.7989513134851135e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 96594392,
      "step": 6003
    },
    {
      "epoch": 0.4205696673583931,
      "grad_norm": 4.8221540451049805,
      "learning_rate": 5.7982514886164644e-05,
      "loss": 1.2558,
      "num_input_tokens_seen": 96609688,
      "step": 6004
    },
    {
      "epoch": 0.42063971560412233,
      "grad_norm": 6.331230163574219,
      "learning_rate": 5.797551663747811e-05,
      "loss": 0.943,
      "num_input_tokens_seen": 96625480,
      "step": 6005
    },
    {
      "epoch": 0.4207097638498516,
      "grad_norm": 4.262217044830322,
      "learning_rate": 5.7968518388791594e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 96641040,
      "step": 6006
    },
    {
      "epoch": 0.42077981209558085,
      "grad_norm": 4.552499294281006,
      "learning_rate": 5.796152014010508e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 96657424,
      "step": 6007
    },
    {
      "epoch": 0.42084986034131006,
      "grad_norm": 3.414970874786377,
      "learning_rate": 5.7954521891418564e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 96673792,
      "step": 6008
    },
    {
      "epoch": 0.4209199085870393,
      "grad_norm": 3.70623517036438,
      "learning_rate": 5.7947523642732046e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 96690176,
      "step": 6009
    },
    {
      "epoch": 0.4209899568327686,
      "grad_norm": 4.370288848876953,
      "learning_rate": 5.794052539404554e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 96706272,
      "step": 6010
    },
    {
      "epoch": 0.42106000507849783,
      "grad_norm": 3.4775140285491943,
      "learning_rate": 5.793352714535904e-05,
      "loss": 0.9383,
      "num_input_tokens_seen": 96722312,
      "step": 6011
    },
    {
      "epoch": 0.42113005332422704,
      "grad_norm": 3.9860763549804688,
      "learning_rate": 5.7926528896672505e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 96737040,
      "step": 6012
    },
    {
      "epoch": 0.4212001015699563,
      "grad_norm": 5.0287933349609375,
      "learning_rate": 5.791953064798599e-05,
      "loss": 0.9892,
      "num_input_tokens_seen": 96753424,
      "step": 6013
    },
    {
      "epoch": 0.42127014981568556,
      "grad_norm": 3.821143627166748,
      "learning_rate": 5.791253239929949e-05,
      "loss": 1.0388,
      "num_input_tokens_seen": 96769552,
      "step": 6014
    },
    {
      "epoch": 0.4213401980614148,
      "grad_norm": 4.180905818939209,
      "learning_rate": 5.7905534150612964e-05,
      "loss": 1.1118,
      "num_input_tokens_seen": 96785936,
      "step": 6015
    },
    {
      "epoch": 0.421410246307144,
      "grad_norm": 4.334224224090576,
      "learning_rate": 5.7898535901926446e-05,
      "loss": 1.3425,
      "num_input_tokens_seen": 96802320,
      "step": 6016
    },
    {
      "epoch": 0.4214802945528733,
      "grad_norm": 4.317337989807129,
      "learning_rate": 5.7891537653239934e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 96818360,
      "step": 6017
    },
    {
      "epoch": 0.42155034279860254,
      "grad_norm": 3.789919376373291,
      "learning_rate": 5.788453940455343e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 96833928,
      "step": 6018
    },
    {
      "epoch": 0.4216203910443318,
      "grad_norm": 4.240170001983643,
      "learning_rate": 5.78775411558669e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 96850312,
      "step": 6019
    },
    {
      "epoch": 0.421690439290061,
      "grad_norm": 5.163384914398193,
      "learning_rate": 5.7870542907180393e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 96866696,
      "step": 6020
    },
    {
      "epoch": 0.42176048753579026,
      "grad_norm": 3.653265953063965,
      "learning_rate": 5.786354465849388e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 96883080,
      "step": 6021
    },
    {
      "epoch": 0.4218305357815195,
      "grad_norm": 3.4269649982452393,
      "learning_rate": 5.785654640980736e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 96899176,
      "step": 6022
    },
    {
      "epoch": 0.4219005840272488,
      "grad_norm": 5.8838276863098145,
      "learning_rate": 5.784954816112084e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 96914576,
      "step": 6023
    },
    {
      "epoch": 0.42197063227297804,
      "grad_norm": 4.201550006866455,
      "learning_rate": 5.784254991243433e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 96930808,
      "step": 6024
    },
    {
      "epoch": 0.42204068051870725,
      "grad_norm": 3.961399793624878,
      "learning_rate": 5.783555166374781e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 96947192,
      "step": 6025
    },
    {
      "epoch": 0.4221107287644365,
      "grad_norm": 4.811456680297852,
      "learning_rate": 5.782855341506129e-05,
      "loss": 1.0576,
      "num_input_tokens_seen": 96961896,
      "step": 6026
    },
    {
      "epoch": 0.42218077701016576,
      "grad_norm": 3.6154356002807617,
      "learning_rate": 5.7821555166374787e-05,
      "loss": 0.9678,
      "num_input_tokens_seen": 96977656,
      "step": 6027
    },
    {
      "epoch": 0.422250825255895,
      "grad_norm": 3.787724256515503,
      "learning_rate": 5.7814556917688275e-05,
      "loss": 1.0813,
      "num_input_tokens_seen": 96993936,
      "step": 6028
    },
    {
      "epoch": 0.4223208735016242,
      "grad_norm": 4.215615272521973,
      "learning_rate": 5.780755866900175e-05,
      "loss": 1.2758,
      "num_input_tokens_seen": 97010320,
      "step": 6029
    },
    {
      "epoch": 0.4223909217473535,
      "grad_norm": 3.9257047176361084,
      "learning_rate": 5.780056042031524e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 97026704,
      "step": 6030
    },
    {
      "epoch": 0.42246096999308275,
      "grad_norm": 3.5415945053100586,
      "learning_rate": 5.7793562171628734e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 97043088,
      "step": 6031
    },
    {
      "epoch": 0.422531018238812,
      "grad_norm": 4.213465213775635,
      "learning_rate": 5.77865639229422e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 97059472,
      "step": 6032
    },
    {
      "epoch": 0.4226010664845412,
      "grad_norm": 3.8070178031921387,
      "learning_rate": 5.7779565674255684e-05,
      "loss": 0.881,
      "num_input_tokens_seen": 97074712,
      "step": 6033
    },
    {
      "epoch": 0.42267111473027047,
      "grad_norm": 3.8083109855651855,
      "learning_rate": 5.777256742556918e-05,
      "loss": 1.0003,
      "num_input_tokens_seen": 97091096,
      "step": 6034
    },
    {
      "epoch": 0.42274116297599973,
      "grad_norm": 3.491002082824707,
      "learning_rate": 5.7765569176882675e-05,
      "loss": 1.0276,
      "num_input_tokens_seen": 97107304,
      "step": 6035
    },
    {
      "epoch": 0.422811211221729,
      "grad_norm": 4.1060919761657715,
      "learning_rate": 5.775857092819616e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 97123688,
      "step": 6036
    },
    {
      "epoch": 0.4228812594674582,
      "grad_norm": 3.8975484371185303,
      "learning_rate": 5.775157267950964e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 97140008,
      "step": 6037
    },
    {
      "epoch": 0.42295130771318745,
      "grad_norm": 4.4457197189331055,
      "learning_rate": 5.774457443082313e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 97156392,
      "step": 6038
    },
    {
      "epoch": 0.4230213559589167,
      "grad_norm": 3.5186471939086914,
      "learning_rate": 5.77375761821366e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 97172776,
      "step": 6039
    },
    {
      "epoch": 0.42309140420464597,
      "grad_norm": 3.729041814804077,
      "learning_rate": 5.7730577933450084e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 97189160,
      "step": 6040
    },
    {
      "epoch": 0.4231614524503752,
      "grad_norm": 4.501081466674805,
      "learning_rate": 5.7723579684763586e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 97205544,
      "step": 6041
    },
    {
      "epoch": 0.42323150069610443,
      "grad_norm": 5.922353744506836,
      "learning_rate": 5.771658143607707e-05,
      "loss": 1.1519,
      "num_input_tokens_seen": 97221928,
      "step": 6042
    },
    {
      "epoch": 0.4233015489418337,
      "grad_norm": 3.649948835372925,
      "learning_rate": 5.770958318739055e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 97238048,
      "step": 6043
    },
    {
      "epoch": 0.42337159718756295,
      "grad_norm": 4.660130977630615,
      "learning_rate": 5.770258493870403e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 97254272,
      "step": 6044
    },
    {
      "epoch": 0.42344164543329216,
      "grad_norm": 4.064535140991211,
      "learning_rate": 5.769558669001752e-05,
      "loss": 1.2646,
      "num_input_tokens_seen": 97270656,
      "step": 6045
    },
    {
      "epoch": 0.4235116936790214,
      "grad_norm": 3.931034803390503,
      "learning_rate": 5.7688588441331e-05,
      "loss": 0.9511,
      "num_input_tokens_seen": 97287040,
      "step": 6046
    },
    {
      "epoch": 0.4235817419247507,
      "grad_norm": 3.920013427734375,
      "learning_rate": 5.7681590192644484e-05,
      "loss": 1.1886,
      "num_input_tokens_seen": 97302784,
      "step": 6047
    },
    {
      "epoch": 0.42365179017047994,
      "grad_norm": 3.356661319732666,
      "learning_rate": 5.767459194395798e-05,
      "loss": 0.9283,
      "num_input_tokens_seen": 97319168,
      "step": 6048
    },
    {
      "epoch": 0.42372183841620914,
      "grad_norm": 4.33698034286499,
      "learning_rate": 5.766759369527145e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 97335552,
      "step": 6049
    },
    {
      "epoch": 0.4237918866619384,
      "grad_norm": 6.201281547546387,
      "learning_rate": 5.766059544658493e-05,
      "loss": 0.9756,
      "num_input_tokens_seen": 97350720,
      "step": 6050
    },
    {
      "epoch": 0.42386193490766766,
      "grad_norm": 4.005791664123535,
      "learning_rate": 5.7653597197898425e-05,
      "loss": 0.9559,
      "num_input_tokens_seen": 97367104,
      "step": 6051
    },
    {
      "epoch": 0.4239319831533969,
      "grad_norm": 4.238742828369141,
      "learning_rate": 5.764659894921191e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 97383488,
      "step": 6052
    },
    {
      "epoch": 0.4240020313991261,
      "grad_norm": 4.139926433563232,
      "learning_rate": 5.7639600700525395e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 97398864,
      "step": 6053
    },
    {
      "epoch": 0.4240720796448554,
      "grad_norm": 3.538890838623047,
      "learning_rate": 5.763260245183888e-05,
      "loss": 0.9162,
      "num_input_tokens_seen": 97414416,
      "step": 6054
    },
    {
      "epoch": 0.42414212789058464,
      "grad_norm": 3.888108253479004,
      "learning_rate": 5.762560420315237e-05,
      "loss": 1.0937,
      "num_input_tokens_seen": 97429616,
      "step": 6055
    },
    {
      "epoch": 0.4242121761363139,
      "grad_norm": 4.287962436676025,
      "learning_rate": 5.7618605954465854e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 97444784,
      "step": 6056
    },
    {
      "epoch": 0.4242822243820431,
      "grad_norm": 3.5160460472106934,
      "learning_rate": 5.761160770577935e-05,
      "loss": 1.0405,
      "num_input_tokens_seen": 97461104,
      "step": 6057
    },
    {
      "epoch": 0.42435227262777236,
      "grad_norm": 4.076432704925537,
      "learning_rate": 5.760460945709283e-05,
      "loss": 1.1768,
      "num_input_tokens_seen": 97477488,
      "step": 6058
    },
    {
      "epoch": 0.4244223208735016,
      "grad_norm": 3.4506590366363525,
      "learning_rate": 5.759761120840631e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 97493872,
      "step": 6059
    },
    {
      "epoch": 0.4244923691192309,
      "grad_norm": 4.196661472320557,
      "learning_rate": 5.7590612959719795e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 97509088,
      "step": 6060
    },
    {
      "epoch": 0.42456241736496014,
      "grad_norm": 4.412662506103516,
      "learning_rate": 5.758361471103328e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 97525472,
      "step": 6061
    },
    {
      "epoch": 0.42463246561068935,
      "grad_norm": 3.4199881553649902,
      "learning_rate": 5.7576616462346765e-05,
      "loss": 1.0078,
      "num_input_tokens_seen": 97541856,
      "step": 6062
    },
    {
      "epoch": 0.4247025138564186,
      "grad_norm": 4.215256214141846,
      "learning_rate": 5.756961821366025e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 97558240,
      "step": 6063
    },
    {
      "epoch": 0.42477256210214787,
      "grad_norm": 4.764070510864258,
      "learning_rate": 5.756261996497374e-05,
      "loss": 1.1994,
      "num_input_tokens_seen": 97574624,
      "step": 6064
    },
    {
      "epoch": 0.4248426103478771,
      "grad_norm": 3.8896613121032715,
      "learning_rate": 5.7555621716287224e-05,
      "loss": 1.1135,
      "num_input_tokens_seen": 97591008,
      "step": 6065
    },
    {
      "epoch": 0.42491265859360633,
      "grad_norm": 5.101664066314697,
      "learning_rate": 5.7548623467600706e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 97607392,
      "step": 6066
    },
    {
      "epoch": 0.4249827068393356,
      "grad_norm": 4.464064121246338,
      "learning_rate": 5.754162521891419e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 97623776,
      "step": 6067
    },
    {
      "epoch": 0.42505275508506485,
      "grad_norm": 3.9882681369781494,
      "learning_rate": 5.7534626970227676e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 97639296,
      "step": 6068
    },
    {
      "epoch": 0.4251228033307941,
      "grad_norm": 5.5437331199646,
      "learning_rate": 5.752762872154116e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 97655592,
      "step": 6069
    },
    {
      "epoch": 0.4251928515765233,
      "grad_norm": 4.157887935638428,
      "learning_rate": 5.752063047285464e-05,
      "loss": 1.2531,
      "num_input_tokens_seen": 97671976,
      "step": 6070
    },
    {
      "epoch": 0.42526289982225257,
      "grad_norm": 4.455500602722168,
      "learning_rate": 5.751363222416812e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 97688360,
      "step": 6071
    },
    {
      "epoch": 0.42533294806798183,
      "grad_norm": 5.3056254386901855,
      "learning_rate": 5.750663397548162e-05,
      "loss": 1.2483,
      "num_input_tokens_seen": 97704008,
      "step": 6072
    },
    {
      "epoch": 0.4254029963137111,
      "grad_norm": 3.5183193683624268,
      "learning_rate": 5.74996357267951e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 97720392,
      "step": 6073
    },
    {
      "epoch": 0.4254730445594403,
      "grad_norm": 4.44768762588501,
      "learning_rate": 5.749263747810859e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 97736584,
      "step": 6074
    },
    {
      "epoch": 0.42554309280516955,
      "grad_norm": 3.6957905292510986,
      "learning_rate": 5.748563922942207e-05,
      "loss": 1.2134,
      "num_input_tokens_seen": 97752968,
      "step": 6075
    },
    {
      "epoch": 0.4256131410508988,
      "grad_norm": 3.6841094493865967,
      "learning_rate": 5.747864098073555e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 97769352,
      "step": 6076
    },
    {
      "epoch": 0.4256831892966281,
      "grad_norm": 6.541488170623779,
      "learning_rate": 5.747164273204903e-05,
      "loss": 1.1638,
      "num_input_tokens_seen": 97785736,
      "step": 6077
    },
    {
      "epoch": 0.4257532375423573,
      "grad_norm": 4.056735515594482,
      "learning_rate": 5.7464644483362515e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 97801624,
      "step": 6078
    },
    {
      "epoch": 0.42582328578808654,
      "grad_norm": 3.5294058322906494,
      "learning_rate": 5.745764623467601e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 97817544,
      "step": 6079
    },
    {
      "epoch": 0.4258933340338158,
      "grad_norm": 3.851330280303955,
      "learning_rate": 5.745064798598949e-05,
      "loss": 1.0858,
      "num_input_tokens_seen": 97833600,
      "step": 6080
    },
    {
      "epoch": 0.42596338227954506,
      "grad_norm": 3.6939046382904053,
      "learning_rate": 5.744364973730299e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 97849984,
      "step": 6081
    },
    {
      "epoch": 0.42603343052527426,
      "grad_norm": 3.7894139289855957,
      "learning_rate": 5.743665148861647e-05,
      "loss": 1.1953,
      "num_input_tokens_seen": 97866368,
      "step": 6082
    },
    {
      "epoch": 0.4261034787710035,
      "grad_norm": 3.377105712890625,
      "learning_rate": 5.742965323992995e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 97882752,
      "step": 6083
    },
    {
      "epoch": 0.4261735270167328,
      "grad_norm": 4.0349440574646,
      "learning_rate": 5.742265499124344e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 97899136,
      "step": 6084
    },
    {
      "epoch": 0.42624357526246204,
      "grad_norm": 3.9353208541870117,
      "learning_rate": 5.7415656742556935e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 97915520,
      "step": 6085
    },
    {
      "epoch": 0.42631362350819124,
      "grad_norm": 3.6593427658081055,
      "learning_rate": 5.740865849387042e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 97931904,
      "step": 6086
    },
    {
      "epoch": 0.4263836717539205,
      "grad_norm": 3.543994665145874,
      "learning_rate": 5.7401660245183885e-05,
      "loss": 0.9268,
      "num_input_tokens_seen": 97948288,
      "step": 6087
    },
    {
      "epoch": 0.42645371999964976,
      "grad_norm": 3.925420045852661,
      "learning_rate": 5.739466199649738e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 97964672,
      "step": 6088
    },
    {
      "epoch": 0.426523768245379,
      "grad_norm": 4.6036224365234375,
      "learning_rate": 5.738766374781086e-05,
      "loss": 1.1229,
      "num_input_tokens_seen": 97981056,
      "step": 6089
    },
    {
      "epoch": 0.4265938164911082,
      "grad_norm": 6.555153846740723,
      "learning_rate": 5.7380665499124344e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 97997440,
      "step": 6090
    },
    {
      "epoch": 0.4266638647368375,
      "grad_norm": 3.7414231300354004,
      "learning_rate": 5.737366725043783e-05,
      "loss": 1.0223,
      "num_input_tokens_seen": 98013264,
      "step": 6091
    },
    {
      "epoch": 0.42673391298256674,
      "grad_norm": 4.380615234375,
      "learning_rate": 5.7366669001751314e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 98029176,
      "step": 6092
    },
    {
      "epoch": 0.426803961228296,
      "grad_norm": 4.624136924743652,
      "learning_rate": 5.7359670753064796e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 98044384,
      "step": 6093
    },
    {
      "epoch": 0.42687400947402526,
      "grad_norm": 4.984564781188965,
      "learning_rate": 5.735267250437828e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 98060768,
      "step": 6094
    },
    {
      "epoch": 0.42694405771975447,
      "grad_norm": 5.481975078582764,
      "learning_rate": 5.734567425569178e-05,
      "loss": 1.167,
      "num_input_tokens_seen": 98077152,
      "step": 6095
    },
    {
      "epoch": 0.4270141059654837,
      "grad_norm": 3.3822808265686035,
      "learning_rate": 5.733867600700526e-05,
      "loss": 0.9442,
      "num_input_tokens_seen": 98093224,
      "step": 6096
    },
    {
      "epoch": 0.427084154211213,
      "grad_norm": 3.8090853691101074,
      "learning_rate": 5.733167775831874e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 98109608,
      "step": 6097
    },
    {
      "epoch": 0.42715420245694224,
      "grad_norm": 4.279370307922363,
      "learning_rate": 5.7324679509632226e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 98125992,
      "step": 6098
    },
    {
      "epoch": 0.42722425070267145,
      "grad_norm": 5.998210430145264,
      "learning_rate": 5.731768126094571e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 98142376,
      "step": 6099
    },
    {
      "epoch": 0.4272942989484007,
      "grad_norm": 4.388184070587158,
      "learning_rate": 5.731068301225919e-05,
      "loss": 1.0364,
      "num_input_tokens_seen": 98158760,
      "step": 6100
    },
    {
      "epoch": 0.42736434719412997,
      "grad_norm": 4.937825679779053,
      "learning_rate": 5.7303684763572685e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 98175144,
      "step": 6101
    },
    {
      "epoch": 0.4274343954398592,
      "grad_norm": 3.800776720046997,
      "learning_rate": 5.729668651488618e-05,
      "loss": 1.1608,
      "num_input_tokens_seen": 98191184,
      "step": 6102
    },
    {
      "epoch": 0.42750444368558843,
      "grad_norm": 3.857093334197998,
      "learning_rate": 5.728968826619966e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 98207568,
      "step": 6103
    },
    {
      "epoch": 0.4275744919313177,
      "grad_norm": 3.562218189239502,
      "learning_rate": 5.728269001751313e-05,
      "loss": 1.1002,
      "num_input_tokens_seen": 98223952,
      "step": 6104
    },
    {
      "epoch": 0.42764454017704695,
      "grad_norm": 3.826802968978882,
      "learning_rate": 5.7275691768826626e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 98239576,
      "step": 6105
    },
    {
      "epoch": 0.4277145884227762,
      "grad_norm": 4.127960205078125,
      "learning_rate": 5.7268693520140114e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 98255960,
      "step": 6106
    },
    {
      "epoch": 0.4277846366685054,
      "grad_norm": 4.270632743835449,
      "learning_rate": 5.726169527145359e-05,
      "loss": 1.2359,
      "num_input_tokens_seen": 98272080,
      "step": 6107
    },
    {
      "epoch": 0.4278546849142347,
      "grad_norm": 4.543783187866211,
      "learning_rate": 5.725469702276708e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 98288464,
      "step": 6108
    },
    {
      "epoch": 0.42792473315996393,
      "grad_norm": 3.993234634399414,
      "learning_rate": 5.724769877408057e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 98304424,
      "step": 6109
    },
    {
      "epoch": 0.4279947814056932,
      "grad_norm": 4.11693000793457,
      "learning_rate": 5.7240700525394055e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 98320808,
      "step": 6110
    },
    {
      "epoch": 0.4280648296514224,
      "grad_norm": 4.000871658325195,
      "learning_rate": 5.723370227670754e-05,
      "loss": 0.9777,
      "num_input_tokens_seen": 98337192,
      "step": 6111
    },
    {
      "epoch": 0.42813487789715166,
      "grad_norm": 3.642763614654541,
      "learning_rate": 5.7226704028021025e-05,
      "loss": 0.9108,
      "num_input_tokens_seen": 98353320,
      "step": 6112
    },
    {
      "epoch": 0.4282049261428809,
      "grad_norm": 4.22330379486084,
      "learning_rate": 5.721970577933451e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 98369704,
      "step": 6113
    },
    {
      "epoch": 0.4282749743886102,
      "grad_norm": 3.7961175441741943,
      "learning_rate": 5.721270753064798e-05,
      "loss": 0.8756,
      "num_input_tokens_seen": 98385544,
      "step": 6114
    },
    {
      "epoch": 0.4283450226343394,
      "grad_norm": 3.771034002304077,
      "learning_rate": 5.720570928196147e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 98401928,
      "step": 6115
    },
    {
      "epoch": 0.42841507088006864,
      "grad_norm": 3.8084332942962646,
      "learning_rate": 5.719871103327495e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 98418136,
      "step": 6116
    },
    {
      "epoch": 0.4284851191257979,
      "grad_norm": 3.890608549118042,
      "learning_rate": 5.7191712784588434e-05,
      "loss": 0.9865,
      "num_input_tokens_seen": 98433656,
      "step": 6117
    },
    {
      "epoch": 0.42855516737152716,
      "grad_norm": 6.781351089477539,
      "learning_rate": 5.718471453590193e-05,
      "loss": 0.8032,
      "num_input_tokens_seen": 98448776,
      "step": 6118
    },
    {
      "epoch": 0.42862521561725636,
      "grad_norm": 3.941107749938965,
      "learning_rate": 5.717771628721542e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 98465160,
      "step": 6119
    },
    {
      "epoch": 0.4286952638629856,
      "grad_norm": 4.457616329193115,
      "learning_rate": 5.71707180385289e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 98481184,
      "step": 6120
    },
    {
      "epoch": 0.4287653121087149,
      "grad_norm": 3.889111042022705,
      "learning_rate": 5.7163719789842375e-05,
      "loss": 1.0685,
      "num_input_tokens_seen": 98497568,
      "step": 6121
    },
    {
      "epoch": 0.42883536035444414,
      "grad_norm": 3.7574422359466553,
      "learning_rate": 5.715672154115588e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 98513920,
      "step": 6122
    },
    {
      "epoch": 0.42890540860017334,
      "grad_norm": 3.578437089920044,
      "learning_rate": 5.714972329246936e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 98529664,
      "step": 6123
    },
    {
      "epoch": 0.4289754568459026,
      "grad_norm": 5.0676398277282715,
      "learning_rate": 5.714272504378283e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 98544936,
      "step": 6124
    },
    {
      "epoch": 0.42904550509163186,
      "grad_norm": 4.475335121154785,
      "learning_rate": 5.713572679509632e-05,
      "loss": 0.9347,
      "num_input_tokens_seen": 98560520,
      "step": 6125
    },
    {
      "epoch": 0.4291155533373611,
      "grad_norm": 6.345788955688477,
      "learning_rate": 5.712872854640982e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 98576320,
      "step": 6126
    },
    {
      "epoch": 0.4291856015830903,
      "grad_norm": 3.775374174118042,
      "learning_rate": 5.71217302977233e-05,
      "loss": 0.9803,
      "num_input_tokens_seen": 98592704,
      "step": 6127
    },
    {
      "epoch": 0.4292556498288196,
      "grad_norm": 4.224292278289795,
      "learning_rate": 5.711473204903678e-05,
      "loss": 1.2253,
      "num_input_tokens_seen": 98607664,
      "step": 6128
    },
    {
      "epoch": 0.42932569807454884,
      "grad_norm": 4.470034122467041,
      "learning_rate": 5.710773380035027e-05,
      "loss": 0.9915,
      "num_input_tokens_seen": 98624048,
      "step": 6129
    },
    {
      "epoch": 0.4293957463202781,
      "grad_norm": 6.22687292098999,
      "learning_rate": 5.710073555166375e-05,
      "loss": 1.2048,
      "num_input_tokens_seen": 98640432,
      "step": 6130
    },
    {
      "epoch": 0.42946579456600736,
      "grad_norm": 3.9434430599212646,
      "learning_rate": 5.709373730297722e-05,
      "loss": 1.0306,
      "num_input_tokens_seen": 98656672,
      "step": 6131
    },
    {
      "epoch": 0.42953584281173657,
      "grad_norm": 3.7640228271484375,
      "learning_rate": 5.7086739054290716e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 98673056,
      "step": 6132
    },
    {
      "epoch": 0.4296058910574658,
      "grad_norm": 5.742674827575684,
      "learning_rate": 5.707974080560421e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 98688400,
      "step": 6133
    },
    {
      "epoch": 0.4296759393031951,
      "grad_norm": 4.938521862030029,
      "learning_rate": 5.707274255691769e-05,
      "loss": 0.9877,
      "num_input_tokens_seen": 98703304,
      "step": 6134
    },
    {
      "epoch": 0.42974598754892435,
      "grad_norm": 3.7322773933410645,
      "learning_rate": 5.7065744308231175e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 98717536,
      "step": 6135
    },
    {
      "epoch": 0.42981603579465355,
      "grad_norm": 3.741265296936035,
      "learning_rate": 5.7058746059544663e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 98733632,
      "step": 6136
    },
    {
      "epoch": 0.4298860840403828,
      "grad_norm": 3.9021074771881104,
      "learning_rate": 5.7051747810858145e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 98749088,
      "step": 6137
    },
    {
      "epoch": 0.42995613228611207,
      "grad_norm": 4.327329635620117,
      "learning_rate": 5.704474956217163e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 98765328,
      "step": 6138
    },
    {
      "epoch": 0.43002618053184133,
      "grad_norm": 4.335643768310547,
      "learning_rate": 5.703775131348512e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 98780744,
      "step": 6139
    },
    {
      "epoch": 0.43009622877757053,
      "grad_norm": 4.66419792175293,
      "learning_rate": 5.7030753064798604e-05,
      "loss": 1.0696,
      "num_input_tokens_seen": 98797128,
      "step": 6140
    },
    {
      "epoch": 0.4301662770232998,
      "grad_norm": 4.208861351013184,
      "learning_rate": 5.702375481611207e-05,
      "loss": 0.9743,
      "num_input_tokens_seen": 98812776,
      "step": 6141
    },
    {
      "epoch": 0.43023632526902905,
      "grad_norm": 4.90700626373291,
      "learning_rate": 5.701675656742557e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 98829160,
      "step": 6142
    },
    {
      "epoch": 0.4303063735147583,
      "grad_norm": 3.942166805267334,
      "learning_rate": 5.7009758318739056e-05,
      "loss": 1.0032,
      "num_input_tokens_seen": 98845544,
      "step": 6143
    },
    {
      "epoch": 0.4303764217604875,
      "grad_norm": 4.919578552246094,
      "learning_rate": 5.700276007005254e-05,
      "loss": 1.0218,
      "num_input_tokens_seen": 98861928,
      "step": 6144
    },
    {
      "epoch": 0.4304464700062168,
      "grad_norm": 3.6429073810577393,
      "learning_rate": 5.699576182136602e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 98878288,
      "step": 6145
    },
    {
      "epoch": 0.43051651825194603,
      "grad_norm": 4.227152347564697,
      "learning_rate": 5.6988763572679515e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 98893816,
      "step": 6146
    },
    {
      "epoch": 0.4305865664976753,
      "grad_norm": 4.016188144683838,
      "learning_rate": 5.6981765323993e-05,
      "loss": 0.9898,
      "num_input_tokens_seen": 98909968,
      "step": 6147
    },
    {
      "epoch": 0.4306566147434045,
      "grad_norm": 4.0402069091796875,
      "learning_rate": 5.697476707530648e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 98926352,
      "step": 6148
    },
    {
      "epoch": 0.43072666298913376,
      "grad_norm": 5.771969318389893,
      "learning_rate": 5.6967768826619974e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 98941512,
      "step": 6149
    },
    {
      "epoch": 0.430796711234863,
      "grad_norm": 4.444697856903076,
      "learning_rate": 5.6960770577933456e-05,
      "loss": 1.1114,
      "num_input_tokens_seen": 98957624,
      "step": 6150
    },
    {
      "epoch": 0.4308667594805923,
      "grad_norm": 3.386268377304077,
      "learning_rate": 5.695377232924694e-05,
      "loss": 0.9552,
      "num_input_tokens_seen": 98974008,
      "step": 6151
    },
    {
      "epoch": 0.4309368077263215,
      "grad_norm": 3.950138807296753,
      "learning_rate": 5.694677408056042e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 98990392,
      "step": 6152
    },
    {
      "epoch": 0.43100685597205074,
      "grad_norm": 3.7290585041046143,
      "learning_rate": 5.693977583187392e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 99006776,
      "step": 6153
    },
    {
      "epoch": 0.43107690421778,
      "grad_norm": 3.3678364753723145,
      "learning_rate": 5.693277758318739e-05,
      "loss": 0.7396,
      "num_input_tokens_seen": 99022912,
      "step": 6154
    },
    {
      "epoch": 0.43114695246350926,
      "grad_norm": 5.882314682006836,
      "learning_rate": 5.6925779334500886e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 99038208,
      "step": 6155
    },
    {
      "epoch": 0.43121700070923846,
      "grad_norm": 4.231525421142578,
      "learning_rate": 5.691878108581437e-05,
      "loss": 1.0437,
      "num_input_tokens_seen": 99053496,
      "step": 6156
    },
    {
      "epoch": 0.4312870489549677,
      "grad_norm": 4.864506721496582,
      "learning_rate": 5.691178283712785e-05,
      "loss": 1.0978,
      "num_input_tokens_seen": 99069600,
      "step": 6157
    },
    {
      "epoch": 0.431357097200697,
      "grad_norm": 6.483276844024658,
      "learning_rate": 5.690478458844133e-05,
      "loss": 0.9262,
      "num_input_tokens_seen": 99085456,
      "step": 6158
    },
    {
      "epoch": 0.43142714544642624,
      "grad_norm": 3.830292224884033,
      "learning_rate": 5.689778633975482e-05,
      "loss": 1.1837,
      "num_input_tokens_seen": 99101840,
      "step": 6159
    },
    {
      "epoch": 0.43149719369215545,
      "grad_norm": 4.078514099121094,
      "learning_rate": 5.68907880910683e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 99118224,
      "step": 6160
    },
    {
      "epoch": 0.4315672419378847,
      "grad_norm": 4.1833648681640625,
      "learning_rate": 5.688378984238178e-05,
      "loss": 1.2243,
      "num_input_tokens_seen": 99134608,
      "step": 6161
    },
    {
      "epoch": 0.43163729018361396,
      "grad_norm": 4.761826515197754,
      "learning_rate": 5.6876791593695265e-05,
      "loss": 1.1017,
      "num_input_tokens_seen": 99150992,
      "step": 6162
    },
    {
      "epoch": 0.4317073384293432,
      "grad_norm": 4.992908954620361,
      "learning_rate": 5.686979334500877e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 99167320,
      "step": 6163
    },
    {
      "epoch": 0.4317773866750725,
      "grad_norm": 3.8283936977386475,
      "learning_rate": 5.686279509632224e-05,
      "loss": 1.0521,
      "num_input_tokens_seen": 99183512,
      "step": 6164
    },
    {
      "epoch": 0.4318474349208017,
      "grad_norm": 3.4508893489837646,
      "learning_rate": 5.685579684763573e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 99199896,
      "step": 6165
    },
    {
      "epoch": 0.43191748316653095,
      "grad_norm": 4.4798431396484375,
      "learning_rate": 5.684879859894921e-05,
      "loss": 0.8848,
      "num_input_tokens_seen": 99216280,
      "step": 6166
    },
    {
      "epoch": 0.4319875314122602,
      "grad_norm": 5.227555751800537,
      "learning_rate": 5.6841800350262694e-05,
      "loss": 0.9621,
      "num_input_tokens_seen": 99230656,
      "step": 6167
    },
    {
      "epoch": 0.43205757965798947,
      "grad_norm": 5.358756065368652,
      "learning_rate": 5.6834802101576176e-05,
      "loss": 1.0361,
      "num_input_tokens_seen": 99246864,
      "step": 6168
    },
    {
      "epoch": 0.43212762790371867,
      "grad_norm": 4.224287986755371,
      "learning_rate": 5.682780385288967e-05,
      "loss": 1.0804,
      "num_input_tokens_seen": 99263248,
      "step": 6169
    },
    {
      "epoch": 0.43219767614944793,
      "grad_norm": 5.748126983642578,
      "learning_rate": 5.682080560420317e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 99279632,
      "step": 6170
    },
    {
      "epoch": 0.4322677243951772,
      "grad_norm": 4.036735534667969,
      "learning_rate": 5.6813807355516635e-05,
      "loss": 1.0776,
      "num_input_tokens_seen": 99296016,
      "step": 6171
    },
    {
      "epoch": 0.43233777264090645,
      "grad_norm": 4.133121013641357,
      "learning_rate": 5.680680910683013e-05,
      "loss": 1.0796,
      "num_input_tokens_seen": 99312400,
      "step": 6172
    },
    {
      "epoch": 0.43240782088663565,
      "grad_norm": 4.70187520980835,
      "learning_rate": 5.679981085814362e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 99328504,
      "step": 6173
    },
    {
      "epoch": 0.4324778691323649,
      "grad_norm": 3.515967845916748,
      "learning_rate": 5.6792812609457094e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 99344424,
      "step": 6174
    },
    {
      "epoch": 0.43254791737809417,
      "grad_norm": 5.408679962158203,
      "learning_rate": 5.6785814360770576e-05,
      "loss": 1.0638,
      "num_input_tokens_seen": 99360056,
      "step": 6175
    },
    {
      "epoch": 0.43261796562382343,
      "grad_norm": 3.7438695430755615,
      "learning_rate": 5.677881611208408e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 99375920,
      "step": 6176
    },
    {
      "epoch": 0.43268801386955263,
      "grad_norm": 4.337923526763916,
      "learning_rate": 5.677181786339756e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 99392040,
      "step": 6177
    },
    {
      "epoch": 0.4327580621152819,
      "grad_norm": 6.04982852935791,
      "learning_rate": 5.676481961471103e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 99407592,
      "step": 6178
    },
    {
      "epoch": 0.43282811036101115,
      "grad_norm": 3.9268686771392822,
      "learning_rate": 5.6757821366024524e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 99423976,
      "step": 6179
    },
    {
      "epoch": 0.4328981586067404,
      "grad_norm": 4.466431140899658,
      "learning_rate": 5.675082311733801e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 99440360,
      "step": 6180
    },
    {
      "epoch": 0.4329682068524696,
      "grad_norm": 4.166913032531738,
      "learning_rate": 5.674382486865149e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 99456744,
      "step": 6181
    },
    {
      "epoch": 0.4330382550981989,
      "grad_norm": 3.525611400604248,
      "learning_rate": 5.6736826619964976e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 99473080,
      "step": 6182
    },
    {
      "epoch": 0.43310830334392814,
      "grad_norm": 6.099409103393555,
      "learning_rate": 5.672982837127847e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 99489088,
      "step": 6183
    },
    {
      "epoch": 0.4331783515896574,
      "grad_norm": 3.5886685848236084,
      "learning_rate": 5.672283012259194e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 99505248,
      "step": 6184
    },
    {
      "epoch": 0.4332483998353866,
      "grad_norm": 3.5279197692871094,
      "learning_rate": 5.671583187390542e-05,
      "loss": 0.9724,
      "num_input_tokens_seen": 99521632,
      "step": 6185
    },
    {
      "epoch": 0.43331844808111586,
      "grad_norm": 4.606603622436523,
      "learning_rate": 5.670883362521892e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 99537336,
      "step": 6186
    },
    {
      "epoch": 0.4333884963268451,
      "grad_norm": 3.5966908931732178,
      "learning_rate": 5.6701835376532405e-05,
      "loss": 0.9549,
      "num_input_tokens_seen": 99553720,
      "step": 6187
    },
    {
      "epoch": 0.4334585445725744,
      "grad_norm": 3.981893301010132,
      "learning_rate": 5.669483712784588e-05,
      "loss": 0.923,
      "num_input_tokens_seen": 99569384,
      "step": 6188
    },
    {
      "epoch": 0.4335285928183036,
      "grad_norm": 4.06168270111084,
      "learning_rate": 5.668783887915937e-05,
      "loss": 1.12,
      "num_input_tokens_seen": 99585768,
      "step": 6189
    },
    {
      "epoch": 0.43359864106403284,
      "grad_norm": 4.476738929748535,
      "learning_rate": 5.6680840630472864e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 99602152,
      "step": 6190
    },
    {
      "epoch": 0.4336686893097621,
      "grad_norm": 7.592894554138184,
      "learning_rate": 5.667384238178633e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 99618536,
      "step": 6191
    },
    {
      "epoch": 0.43373873755549136,
      "grad_norm": 3.4367337226867676,
      "learning_rate": 5.666684413309983e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 99634920,
      "step": 6192
    },
    {
      "epoch": 0.43380878580122056,
      "grad_norm": 3.8736166954040527,
      "learning_rate": 5.665984588441332e-05,
      "loss": 1.1062,
      "num_input_tokens_seen": 99651304,
      "step": 6193
    },
    {
      "epoch": 0.4338788340469498,
      "grad_norm": 4.0018463134765625,
      "learning_rate": 5.6652847635726805e-05,
      "loss": 1.15,
      "num_input_tokens_seen": 99667688,
      "step": 6194
    },
    {
      "epoch": 0.4339488822926791,
      "grad_norm": 4.243009090423584,
      "learning_rate": 5.664584938704029e-05,
      "loss": 1.0738,
      "num_input_tokens_seen": 99684072,
      "step": 6195
    },
    {
      "epoch": 0.43401893053840834,
      "grad_norm": 5.533624172210693,
      "learning_rate": 5.663885113835377e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 99698440,
      "step": 6196
    },
    {
      "epoch": 0.4340889787841376,
      "grad_norm": 3.9158618450164795,
      "learning_rate": 5.663185288966726e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 99714824,
      "step": 6197
    },
    {
      "epoch": 0.4341590270298668,
      "grad_norm": 6.666274070739746,
      "learning_rate": 5.662485464098073e-05,
      "loss": 1.3152,
      "num_input_tokens_seen": 99731208,
      "step": 6198
    },
    {
      "epoch": 0.43422907527559607,
      "grad_norm": 4.02492618560791,
      "learning_rate": 5.6617856392294235e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 99747592,
      "step": 6199
    },
    {
      "epoch": 0.4342991235213253,
      "grad_norm": 3.4257941246032715,
      "learning_rate": 5.6610858143607716e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 99763656,
      "step": 6200
    },
    {
      "epoch": 0.4342991235213253,
      "eval_loss": 1.126607060432434,
      "eval_runtime": 0.1953,
      "eval_samples_per_second": 5.119,
      "eval_steps_per_second": 5.119,
      "num_input_tokens_seen": 99763656,
      "step": 6200
    },
    {
      "epoch": 0.4343691717670546,
      "grad_norm": 3.7283267974853516,
      "learning_rate": 5.66038598949212e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 99779416,
      "step": 6201
    },
    {
      "epoch": 0.4344392200127838,
      "grad_norm": 5.657198429107666,
      "learning_rate": 5.659686164623468e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 99795376,
      "step": 6202
    },
    {
      "epoch": 0.43450926825851305,
      "grad_norm": 4.102888107299805,
      "learning_rate": 5.658986339754817e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 99811760,
      "step": 6203
    },
    {
      "epoch": 0.4345793165042423,
      "grad_norm": 5.449219226837158,
      "learning_rate": 5.658286514886165e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 99827264,
      "step": 6204
    },
    {
      "epoch": 0.43464936474997157,
      "grad_norm": 4.1982197761535645,
      "learning_rate": 5.657586690017513e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 99843648,
      "step": 6205
    },
    {
      "epoch": 0.43471941299570077,
      "grad_norm": 4.246870517730713,
      "learning_rate": 5.6568868651488614e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 99859800,
      "step": 6206
    },
    {
      "epoch": 0.43478946124143003,
      "grad_norm": 7.575351238250732,
      "learning_rate": 5.656187040280211e-05,
      "loss": 1.231,
      "num_input_tokens_seen": 99875240,
      "step": 6207
    },
    {
      "epoch": 0.4348595094871593,
      "grad_norm": 4.253138065338135,
      "learning_rate": 5.655487215411558e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 99891624,
      "step": 6208
    },
    {
      "epoch": 0.43492955773288855,
      "grad_norm": 3.5073490142822266,
      "learning_rate": 5.654787390542908e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 99908008,
      "step": 6209
    },
    {
      "epoch": 0.43499960597861775,
      "grad_norm": 3.669361114501953,
      "learning_rate": 5.654087565674256e-05,
      "loss": 1.0233,
      "num_input_tokens_seen": 99924392,
      "step": 6210
    },
    {
      "epoch": 0.435069654224347,
      "grad_norm": 4.25203800201416,
      "learning_rate": 5.6533877408056043e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 99940776,
      "step": 6211
    },
    {
      "epoch": 0.4351397024700763,
      "grad_norm": 3.7570602893829346,
      "learning_rate": 5.6526879159369525e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 99956488,
      "step": 6212
    },
    {
      "epoch": 0.43520975071580553,
      "grad_norm": 3.47245454788208,
      "learning_rate": 5.651988091068301e-05,
      "loss": 0.8526,
      "num_input_tokens_seen": 99972792,
      "step": 6213
    },
    {
      "epoch": 0.43527979896153474,
      "grad_norm": 4.902298927307129,
      "learning_rate": 5.65128826619965e-05,
      "loss": 1.1583,
      "num_input_tokens_seen": 99988144,
      "step": 6214
    },
    {
      "epoch": 0.435349847207264,
      "grad_norm": 3.796644926071167,
      "learning_rate": 5.6505884413309984e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 100004528,
      "step": 6215
    },
    {
      "epoch": 0.43541989545299326,
      "grad_norm": 4.4678425788879395,
      "learning_rate": 5.649888616462348e-05,
      "loss": 1.0843,
      "num_input_tokens_seen": 100020448,
      "step": 6216
    },
    {
      "epoch": 0.4354899436987225,
      "grad_norm": 5.181003570556641,
      "learning_rate": 5.649188791593696e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 100036832,
      "step": 6217
    },
    {
      "epoch": 0.4355599919444517,
      "grad_norm": 3.4864094257354736,
      "learning_rate": 5.648488966725044e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 100052656,
      "step": 6218
    },
    {
      "epoch": 0.435630040190181,
      "grad_norm": 3.6500463485717773,
      "learning_rate": 5.647789141856393e-05,
      "loss": 0.9406,
      "num_input_tokens_seen": 100069040,
      "step": 6219
    },
    {
      "epoch": 0.43570008843591024,
      "grad_norm": 4.463146686553955,
      "learning_rate": 5.647089316987743e-05,
      "loss": 1.0597,
      "num_input_tokens_seen": 100085248,
      "step": 6220
    },
    {
      "epoch": 0.4357701366816395,
      "grad_norm": 4.013953685760498,
      "learning_rate": 5.6463894921190895e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 100100600,
      "step": 6221
    },
    {
      "epoch": 0.4358401849273687,
      "grad_norm": 4.549919128417969,
      "learning_rate": 5.645689667250438e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 100116368,
      "step": 6222
    },
    {
      "epoch": 0.43591023317309796,
      "grad_norm": 4.227685451507568,
      "learning_rate": 5.644989842381787e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 100132752,
      "step": 6223
    },
    {
      "epoch": 0.4359802814188272,
      "grad_norm": 3.699406147003174,
      "learning_rate": 5.6442900175131354e-05,
      "loss": 0.9597,
      "num_input_tokens_seen": 100149048,
      "step": 6224
    },
    {
      "epoch": 0.4360503296645565,
      "grad_norm": 4.6700944900512695,
      "learning_rate": 5.6435901926444836e-05,
      "loss": 1.1303,
      "num_input_tokens_seen": 100165432,
      "step": 6225
    },
    {
      "epoch": 0.4361203779102857,
      "grad_norm": 3.524517774581909,
      "learning_rate": 5.6428903677758325e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 100181672,
      "step": 6226
    },
    {
      "epoch": 0.43619042615601494,
      "grad_norm": 3.6368181705474854,
      "learning_rate": 5.642190542907181e-05,
      "loss": 0.9204,
      "num_input_tokens_seen": 100198056,
      "step": 6227
    },
    {
      "epoch": 0.4362604744017442,
      "grad_norm": 4.629672050476074,
      "learning_rate": 5.641490718038529e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 100214440,
      "step": 6228
    },
    {
      "epoch": 0.43633052264747346,
      "grad_norm": 4.120620250701904,
      "learning_rate": 5.640790893169877e-05,
      "loss": 0.9761,
      "num_input_tokens_seen": 100230824,
      "step": 6229
    },
    {
      "epoch": 0.43640057089320267,
      "grad_norm": 3.8496460914611816,
      "learning_rate": 5.640091068301227e-05,
      "loss": 1.0973,
      "num_input_tokens_seen": 100245920,
      "step": 6230
    },
    {
      "epoch": 0.4364706191389319,
      "grad_norm": 3.8419101238250732,
      "learning_rate": 5.639391243432575e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 100262184,
      "step": 6231
    },
    {
      "epoch": 0.4365406673846612,
      "grad_norm": 4.283138751983643,
      "learning_rate": 5.638691418563923e-05,
      "loss": 0.999,
      "num_input_tokens_seen": 100278568,
      "step": 6232
    },
    {
      "epoch": 0.43661071563039044,
      "grad_norm": 3.7390506267547607,
      "learning_rate": 5.637991593695272e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 100294952,
      "step": 6233
    },
    {
      "epoch": 0.4366807638761197,
      "grad_norm": 3.894780158996582,
      "learning_rate": 5.63729176882662e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 100310536,
      "step": 6234
    },
    {
      "epoch": 0.4367508121218489,
      "grad_norm": 5.446288108825684,
      "learning_rate": 5.636591943957968e-05,
      "loss": 0.973,
      "num_input_tokens_seen": 100326336,
      "step": 6235
    },
    {
      "epoch": 0.43682086036757817,
      "grad_norm": 5.487906455993652,
      "learning_rate": 5.635892119089318e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 100342720,
      "step": 6236
    },
    {
      "epoch": 0.4368909086133074,
      "grad_norm": 4.296425819396973,
      "learning_rate": 5.635192294220667e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 100358912,
      "step": 6237
    },
    {
      "epoch": 0.4369609568590367,
      "grad_norm": 4.339141845703125,
      "learning_rate": 5.634492469352014e-05,
      "loss": 1.1232,
      "num_input_tokens_seen": 100374952,
      "step": 6238
    },
    {
      "epoch": 0.4370310051047659,
      "grad_norm": 4.520789623260498,
      "learning_rate": 5.633792644483362e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 100390712,
      "step": 6239
    },
    {
      "epoch": 0.43710105335049515,
      "grad_norm": 3.7790653705596924,
      "learning_rate": 5.633092819614712e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 100406936,
      "step": 6240
    },
    {
      "epoch": 0.4371711015962244,
      "grad_norm": 3.7649457454681396,
      "learning_rate": 5.63239299474606e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 100423320,
      "step": 6241
    },
    {
      "epoch": 0.43724114984195367,
      "grad_norm": 5.27927827835083,
      "learning_rate": 5.631693169877408e-05,
      "loss": 1.1157,
      "num_input_tokens_seen": 100439704,
      "step": 6242
    },
    {
      "epoch": 0.4373111980876829,
      "grad_norm": 4.266254901885986,
      "learning_rate": 5.630993345008757e-05,
      "loss": 1.1607,
      "num_input_tokens_seen": 100455008,
      "step": 6243
    },
    {
      "epoch": 0.43738124633341213,
      "grad_norm": 4.195004940032959,
      "learning_rate": 5.6302935201401065e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 100471392,
      "step": 6244
    },
    {
      "epoch": 0.4374512945791414,
      "grad_norm": 4.85727596282959,
      "learning_rate": 5.629593695271455e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 100486832,
      "step": 6245
    },
    {
      "epoch": 0.43752134282487065,
      "grad_norm": 3.865466594696045,
      "learning_rate": 5.628893870402803e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 100502496,
      "step": 6246
    },
    {
      "epoch": 0.43759139107059986,
      "grad_norm": 4.107895851135254,
      "learning_rate": 5.628194045534152e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 100518816,
      "step": 6247
    },
    {
      "epoch": 0.4376614393163291,
      "grad_norm": 5.402096271514893,
      "learning_rate": 5.627494220665499e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 100535200,
      "step": 6248
    },
    {
      "epoch": 0.4377314875620584,
      "grad_norm": 4.255467414855957,
      "learning_rate": 5.6267943957968474e-05,
      "loss": 1.268,
      "num_input_tokens_seen": 100551456,
      "step": 6249
    },
    {
      "epoch": 0.43780153580778763,
      "grad_norm": 3.4338836669921875,
      "learning_rate": 5.626094570928196e-05,
      "loss": 1.0149,
      "num_input_tokens_seen": 100567840,
      "step": 6250
    },
    {
      "epoch": 0.43787158405351684,
      "grad_norm": 4.445374488830566,
      "learning_rate": 5.6253947460595445e-05,
      "loss": 1.2527,
      "num_input_tokens_seen": 100583848,
      "step": 6251
    },
    {
      "epoch": 0.4379416322992461,
      "grad_norm": 4.0756072998046875,
      "learning_rate": 5.6246949211908927e-05,
      "loss": 0.8848,
      "num_input_tokens_seen": 100600000,
      "step": 6252
    },
    {
      "epoch": 0.43801168054497536,
      "grad_norm": 3.910945177078247,
      "learning_rate": 5.623995096322242e-05,
      "loss": 0.9661,
      "num_input_tokens_seen": 100615424,
      "step": 6253
    },
    {
      "epoch": 0.4380817287907046,
      "grad_norm": 3.878586769104004,
      "learning_rate": 5.623295271453591e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 100631808,
      "step": 6254
    },
    {
      "epoch": 0.4381517770364338,
      "grad_norm": 4.295658111572266,
      "learning_rate": 5.622595446584939e-05,
      "loss": 1.2368,
      "num_input_tokens_seen": 100647344,
      "step": 6255
    },
    {
      "epoch": 0.4382218252821631,
      "grad_norm": 3.88688063621521,
      "learning_rate": 5.621895621716287e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 100663464,
      "step": 6256
    },
    {
      "epoch": 0.43829187352789234,
      "grad_norm": 3.6060731410980225,
      "learning_rate": 5.621195796847637e-05,
      "loss": 1.1189,
      "num_input_tokens_seen": 100679848,
      "step": 6257
    },
    {
      "epoch": 0.4383619217736216,
      "grad_norm": 4.274289608001709,
      "learning_rate": 5.620495971978984e-05,
      "loss": 0.9809,
      "num_input_tokens_seen": 100695760,
      "step": 6258
    },
    {
      "epoch": 0.4384319700193508,
      "grad_norm": 4.854022979736328,
      "learning_rate": 5.619796147110332e-05,
      "loss": 0.8043,
      "num_input_tokens_seen": 100711144,
      "step": 6259
    },
    {
      "epoch": 0.43850201826508006,
      "grad_norm": 3.9589812755584717,
      "learning_rate": 5.6190963222416815e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 100727088,
      "step": 6260
    },
    {
      "epoch": 0.4385720665108093,
      "grad_norm": 5.07575798034668,
      "learning_rate": 5.618396497373031e-05,
      "loss": 1.0708,
      "num_input_tokens_seen": 100743232,
      "step": 6261
    },
    {
      "epoch": 0.4386421147565386,
      "grad_norm": 3.557736396789551,
      "learning_rate": 5.617696672504379e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 100759616,
      "step": 6262
    },
    {
      "epoch": 0.4387121630022678,
      "grad_norm": 4.200889587402344,
      "learning_rate": 5.6169968476357274e-05,
      "loss": 0.9893,
      "num_input_tokens_seen": 100775176,
      "step": 6263
    },
    {
      "epoch": 0.43878221124799704,
      "grad_norm": 4.214064121246338,
      "learning_rate": 5.616297022767076e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 100791232,
      "step": 6264
    },
    {
      "epoch": 0.4388522594937263,
      "grad_norm": 5.372243881225586,
      "learning_rate": 5.6155971978984244e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 100807496,
      "step": 6265
    },
    {
      "epoch": 0.43892230773945556,
      "grad_norm": 4.754215717315674,
      "learning_rate": 5.614897373029771e-05,
      "loss": 1.1895,
      "num_input_tokens_seen": 100823880,
      "step": 6266
    },
    {
      "epoch": 0.4389923559851848,
      "grad_norm": 3.3892760276794434,
      "learning_rate": 5.614197548161121e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 100839928,
      "step": 6267
    },
    {
      "epoch": 0.439062404230914,
      "grad_norm": 4.554326057434082,
      "learning_rate": 5.61349772329247e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 100856312,
      "step": 6268
    },
    {
      "epoch": 0.4391324524766433,
      "grad_norm": 4.118383407592773,
      "learning_rate": 5.6127978984238185e-05,
      "loss": 0.7906,
      "num_input_tokens_seen": 100872696,
      "step": 6269
    },
    {
      "epoch": 0.43920250072237255,
      "grad_norm": 4.403461456298828,
      "learning_rate": 5.612098073555167e-05,
      "loss": 1.1391,
      "num_input_tokens_seen": 100888808,
      "step": 6270
    },
    {
      "epoch": 0.4392725489681018,
      "grad_norm": 3.841547966003418,
      "learning_rate": 5.6113982486865156e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 100905192,
      "step": 6271
    },
    {
      "epoch": 0.439342597213831,
      "grad_norm": 4.147423267364502,
      "learning_rate": 5.610698423817864e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 100920552,
      "step": 6272
    },
    {
      "epoch": 0.43941264545956027,
      "grad_norm": 4.717578887939453,
      "learning_rate": 5.609998598949212e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 100936656,
      "step": 6273
    },
    {
      "epoch": 0.43948269370528953,
      "grad_norm": 5.503146171569824,
      "learning_rate": 5.6092987740805615e-05,
      "loss": 1.0768,
      "num_input_tokens_seen": 100952952,
      "step": 6274
    },
    {
      "epoch": 0.4395527419510188,
      "grad_norm": 3.6871585845947266,
      "learning_rate": 5.6085989492119096e-05,
      "loss": 0.9386,
      "num_input_tokens_seen": 100969152,
      "step": 6275
    },
    {
      "epoch": 0.439622790196748,
      "grad_norm": 3.751429796218872,
      "learning_rate": 5.6078991243432565e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 100985520,
      "step": 6276
    },
    {
      "epoch": 0.43969283844247725,
      "grad_norm": 4.053867340087891,
      "learning_rate": 5.607199299474606e-05,
      "loss": 1.1341,
      "num_input_tokens_seen": 101001336,
      "step": 6277
    },
    {
      "epoch": 0.4397628866882065,
      "grad_norm": 3.786154270172119,
      "learning_rate": 5.606499474605955e-05,
      "loss": 1.077,
      "num_input_tokens_seen": 101017240,
      "step": 6278
    },
    {
      "epoch": 0.43983293493393577,
      "grad_norm": 3.516772747039795,
      "learning_rate": 5.605799649737303e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 101033264,
      "step": 6279
    },
    {
      "epoch": 0.439902983179665,
      "grad_norm": 4.568872928619385,
      "learning_rate": 5.605099824868651e-05,
      "loss": 1.1444,
      "num_input_tokens_seen": 101049648,
      "step": 6280
    },
    {
      "epoch": 0.43997303142539423,
      "grad_norm": 4.430622577667236,
      "learning_rate": 5.604400000000001e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 101066032,
      "step": 6281
    },
    {
      "epoch": 0.4400430796711235,
      "grad_norm": 5.061071872711182,
      "learning_rate": 5.603700175131349e-05,
      "loss": 1.0891,
      "num_input_tokens_seen": 101082416,
      "step": 6282
    },
    {
      "epoch": 0.44011312791685275,
      "grad_norm": 3.696657180786133,
      "learning_rate": 5.603000350262696e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 101098800,
      "step": 6283
    },
    {
      "epoch": 0.44018317616258196,
      "grad_norm": 4.1430840492248535,
      "learning_rate": 5.6023005253940467e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 101114672,
      "step": 6284
    },
    {
      "epoch": 0.4402532244083112,
      "grad_norm": 4.862906455993652,
      "learning_rate": 5.601600700525395e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 101130632,
      "step": 6285
    },
    {
      "epoch": 0.4403232726540405,
      "grad_norm": 4.017249584197998,
      "learning_rate": 5.600900875656743e-05,
      "loss": 0.9911,
      "num_input_tokens_seen": 101145920,
      "step": 6286
    },
    {
      "epoch": 0.44039332089976974,
      "grad_norm": 4.797904014587402,
      "learning_rate": 5.600201050788091e-05,
      "loss": 1.106,
      "num_input_tokens_seen": 101162296,
      "step": 6287
    },
    {
      "epoch": 0.44046336914549894,
      "grad_norm": 3.685084342956543,
      "learning_rate": 5.59950122591944e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 101178048,
      "step": 6288
    },
    {
      "epoch": 0.4405334173912282,
      "grad_norm": 4.259701728820801,
      "learning_rate": 5.598801401050788e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 101194080,
      "step": 6289
    },
    {
      "epoch": 0.44060346563695746,
      "grad_norm": 3.961292266845703,
      "learning_rate": 5.598101576182138e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 101209000,
      "step": 6290
    },
    {
      "epoch": 0.4406735138826867,
      "grad_norm": 3.863640308380127,
      "learning_rate": 5.597401751313486e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 101224184,
      "step": 6291
    },
    {
      "epoch": 0.4407435621284159,
      "grad_norm": 6.002960681915283,
      "learning_rate": 5.596701926444834e-05,
      "loss": 1.3855,
      "num_input_tokens_seen": 101240568,
      "step": 6292
    },
    {
      "epoch": 0.4408136103741452,
      "grad_norm": 3.870892286300659,
      "learning_rate": 5.596002101576182e-05,
      "loss": 1.0754,
      "num_input_tokens_seen": 101256952,
      "step": 6293
    },
    {
      "epoch": 0.44088365861987444,
      "grad_norm": 3.654907703399658,
      "learning_rate": 5.5953022767075305e-05,
      "loss": 1.1244,
      "num_input_tokens_seen": 101273192,
      "step": 6294
    },
    {
      "epoch": 0.4409537068656037,
      "grad_norm": 3.2243661880493164,
      "learning_rate": 5.5946024518388794e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 101289488,
      "step": 6295
    },
    {
      "epoch": 0.4410237551113329,
      "grad_norm": 3.947880983352661,
      "learning_rate": 5.5939026269702275e-05,
      "loss": 1.1628,
      "num_input_tokens_seen": 101305064,
      "step": 6296
    },
    {
      "epoch": 0.44109380335706216,
      "grad_norm": 3.546065092086792,
      "learning_rate": 5.593202802101576e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 101321448,
      "step": 6297
    },
    {
      "epoch": 0.4411638516027914,
      "grad_norm": 4.489794731140137,
      "learning_rate": 5.592502977232925e-05,
      "loss": 1.1958,
      "num_input_tokens_seen": 101337832,
      "step": 6298
    },
    {
      "epoch": 0.4412338998485207,
      "grad_norm": 3.9517438411712646,
      "learning_rate": 5.5918031523642734e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 101354216,
      "step": 6299
    },
    {
      "epoch": 0.4413039480942499,
      "grad_norm": 4.599244594573975,
      "learning_rate": 5.591103327495622e-05,
      "loss": 0.9712,
      "num_input_tokens_seen": 101370088,
      "step": 6300
    },
    {
      "epoch": 0.44137399633997915,
      "grad_norm": 3.753528356552124,
      "learning_rate": 5.5904035026269705e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 101385432,
      "step": 6301
    },
    {
      "epoch": 0.4414440445857084,
      "grad_norm": 4.569333553314209,
      "learning_rate": 5.589703677758319e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 101401816,
      "step": 6302
    },
    {
      "epoch": 0.44151409283143767,
      "grad_norm": 4.010447978973389,
      "learning_rate": 5.589003852889667e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 101417272,
      "step": 6303
    },
    {
      "epoch": 0.4415841410771669,
      "grad_norm": 5.169422626495361,
      "learning_rate": 5.588304028021015e-05,
      "loss": 1.3555,
      "num_input_tokens_seen": 101433656,
      "step": 6304
    },
    {
      "epoch": 0.44165418932289613,
      "grad_norm": 4.6301069259643555,
      "learning_rate": 5.587604203152366e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 101450040,
      "step": 6305
    },
    {
      "epoch": 0.4417242375686254,
      "grad_norm": 4.814012050628662,
      "learning_rate": 5.586904378283713e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 101466320,
      "step": 6306
    },
    {
      "epoch": 0.44179428581435465,
      "grad_norm": 4.340104579925537,
      "learning_rate": 5.586204553415062e-05,
      "loss": 1.1179,
      "num_input_tokens_seen": 101482592,
      "step": 6307
    },
    {
      "epoch": 0.4418643340600839,
      "grad_norm": 3.807495355606079,
      "learning_rate": 5.5855047285464105e-05,
      "loss": 1.1407,
      "num_input_tokens_seen": 101498920,
      "step": 6308
    },
    {
      "epoch": 0.4419343823058131,
      "grad_norm": 3.97273325920105,
      "learning_rate": 5.5848049036777586e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 101515304,
      "step": 6309
    },
    {
      "epoch": 0.44200443055154237,
      "grad_norm": 6.926362037658691,
      "learning_rate": 5.584105078809107e-05,
      "loss": 0.9045,
      "num_input_tokens_seen": 101531688,
      "step": 6310
    },
    {
      "epoch": 0.44207447879727163,
      "grad_norm": 4.482272624969482,
      "learning_rate": 5.583405253940457e-05,
      "loss": 1.1431,
      "num_input_tokens_seen": 101547912,
      "step": 6311
    },
    {
      "epoch": 0.4421445270430009,
      "grad_norm": 3.726999044418335,
      "learning_rate": 5.582705429071805e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 101563640,
      "step": 6312
    },
    {
      "epoch": 0.4422145752887301,
      "grad_norm": 4.305807113647461,
      "learning_rate": 5.582005604203152e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 101580024,
      "step": 6313
    },
    {
      "epoch": 0.44228462353445935,
      "grad_norm": 5.402091979980469,
      "learning_rate": 5.5813057793345016e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 101596408,
      "step": 6314
    },
    {
      "epoch": 0.4423546717801886,
      "grad_norm": 3.658170700073242,
      "learning_rate": 5.5806059544658504e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 101612792,
      "step": 6315
    },
    {
      "epoch": 0.4424247200259179,
      "grad_norm": 3.91109561920166,
      "learning_rate": 5.579906129597198e-05,
      "loss": 0.9991,
      "num_input_tokens_seen": 101628408,
      "step": 6316
    },
    {
      "epoch": 0.4424947682716471,
      "grad_norm": 3.9523725509643555,
      "learning_rate": 5.579206304728547e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 101644616,
      "step": 6317
    },
    {
      "epoch": 0.44256481651737634,
      "grad_norm": 4.591569423675537,
      "learning_rate": 5.578506479859895e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 101660536,
      "step": 6318
    },
    {
      "epoch": 0.4426348647631056,
      "grad_norm": 3.7487003803253174,
      "learning_rate": 5.577806654991243e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 101676920,
      "step": 6319
    },
    {
      "epoch": 0.44270491300883485,
      "grad_norm": 4.111825942993164,
      "learning_rate": 5.5771068301225913e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 101692904,
      "step": 6320
    },
    {
      "epoch": 0.44277496125456406,
      "grad_norm": 3.7022197246551514,
      "learning_rate": 5.576407005253941e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 101709288,
      "step": 6321
    },
    {
      "epoch": 0.4428450095002933,
      "grad_norm": 5.004938125610352,
      "learning_rate": 5.57570718038529e-05,
      "loss": 1.1042,
      "num_input_tokens_seen": 101725176,
      "step": 6322
    },
    {
      "epoch": 0.4429150577460226,
      "grad_norm": 3.728410005569458,
      "learning_rate": 5.575007355516637e-05,
      "loss": 0.9879,
      "num_input_tokens_seen": 101741160,
      "step": 6323
    },
    {
      "epoch": 0.44298510599175184,
      "grad_norm": 4.526604175567627,
      "learning_rate": 5.574307530647986e-05,
      "loss": 1.1465,
      "num_input_tokens_seen": 101756848,
      "step": 6324
    },
    {
      "epoch": 0.44305515423748104,
      "grad_norm": 3.4281585216522217,
      "learning_rate": 5.5736077057793356e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 101773232,
      "step": 6325
    },
    {
      "epoch": 0.4431252024832103,
      "grad_norm": 5.678319931030273,
      "learning_rate": 5.5729078809106825e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 101789064,
      "step": 6326
    },
    {
      "epoch": 0.44319525072893956,
      "grad_norm": 4.231290817260742,
      "learning_rate": 5.572208056042032e-05,
      "loss": 1.0336,
      "num_input_tokens_seen": 101805360,
      "step": 6327
    },
    {
      "epoch": 0.4432652989746688,
      "grad_norm": 3.9336435794830322,
      "learning_rate": 5.5715082311733815e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 101821744,
      "step": 6328
    },
    {
      "epoch": 0.443335347220398,
      "grad_norm": 3.6775193214416504,
      "learning_rate": 5.57080840630473e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 101838128,
      "step": 6329
    },
    {
      "epoch": 0.4434053954661273,
      "grad_norm": 3.7788445949554443,
      "learning_rate": 5.5701085814360766e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 101854216,
      "step": 6330
    },
    {
      "epoch": 0.44347544371185654,
      "grad_norm": 5.49542236328125,
      "learning_rate": 5.569408756567426e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 101870600,
      "step": 6331
    },
    {
      "epoch": 0.4435454919575858,
      "grad_norm": 4.486842155456543,
      "learning_rate": 5.568708931698775e-05,
      "loss": 1.2571,
      "num_input_tokens_seen": 101886984,
      "step": 6332
    },
    {
      "epoch": 0.443615540203315,
      "grad_norm": 4.949841499328613,
      "learning_rate": 5.5680091068301225e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 101903368,
      "step": 6333
    },
    {
      "epoch": 0.44368558844904427,
      "grad_norm": 3.6375255584716797,
      "learning_rate": 5.567309281961471e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 101919568,
      "step": 6334
    },
    {
      "epoch": 0.4437556366947735,
      "grad_norm": 4.649466037750244,
      "learning_rate": 5.566609457092821e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 101935952,
      "step": 6335
    },
    {
      "epoch": 0.4438256849405028,
      "grad_norm": 3.971482276916504,
      "learning_rate": 5.565909632224169e-05,
      "loss": 1.2222,
      "num_input_tokens_seen": 101952336,
      "step": 6336
    },
    {
      "epoch": 0.44389573318623204,
      "grad_norm": 4.605628967285156,
      "learning_rate": 5.565209807355517e-05,
      "loss": 1.1634,
      "num_input_tokens_seen": 101967752,
      "step": 6337
    },
    {
      "epoch": 0.44396578143196125,
      "grad_norm": 6.878963947296143,
      "learning_rate": 5.564509982486866e-05,
      "loss": 0.831,
      "num_input_tokens_seen": 101982648,
      "step": 6338
    },
    {
      "epoch": 0.4440358296776905,
      "grad_norm": 4.339694976806641,
      "learning_rate": 5.563810157618214e-05,
      "loss": 1.206,
      "num_input_tokens_seen": 101998912,
      "step": 6339
    },
    {
      "epoch": 0.44410587792341977,
      "grad_norm": 3.5509302616119385,
      "learning_rate": 5.563110332749562e-05,
      "loss": 0.969,
      "num_input_tokens_seen": 102015296,
      "step": 6340
    },
    {
      "epoch": 0.444175926169149,
      "grad_norm": 6.927268981933594,
      "learning_rate": 5.5624105078809106e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 102031136,
      "step": 6341
    },
    {
      "epoch": 0.44424597441487823,
      "grad_norm": 4.596194267272949,
      "learning_rate": 5.56171068301226e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 102047016,
      "step": 6342
    },
    {
      "epoch": 0.4443160226606075,
      "grad_norm": 3.8641550540924072,
      "learning_rate": 5.561010858143607e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 102062344,
      "step": 6343
    },
    {
      "epoch": 0.44438607090633675,
      "grad_norm": 5.471240997314453,
      "learning_rate": 5.5603110332749565e-05,
      "loss": 0.9372,
      "num_input_tokens_seen": 102078392,
      "step": 6344
    },
    {
      "epoch": 0.444456119152066,
      "grad_norm": 3.971010208129883,
      "learning_rate": 5.5596112084063054e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 102092872,
      "step": 6345
    },
    {
      "epoch": 0.4445261673977952,
      "grad_norm": 4.037472724914551,
      "learning_rate": 5.5589113835376536e-05,
      "loss": 0.959,
      "num_input_tokens_seen": 102109256,
      "step": 6346
    },
    {
      "epoch": 0.4445962156435245,
      "grad_norm": 3.834984302520752,
      "learning_rate": 5.558211558669002e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 102125424,
      "step": 6347
    },
    {
      "epoch": 0.44466626388925373,
      "grad_norm": 5.131717205047607,
      "learning_rate": 5.55751173380035e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 102141808,
      "step": 6348
    },
    {
      "epoch": 0.444736312134983,
      "grad_norm": 4.5308837890625,
      "learning_rate": 5.5568119089316995e-05,
      "loss": 1.1752,
      "num_input_tokens_seen": 102158192,
      "step": 6349
    },
    {
      "epoch": 0.4448063603807122,
      "grad_norm": 5.088570594787598,
      "learning_rate": 5.556112084063046e-05,
      "loss": 1.0215,
      "num_input_tokens_seen": 102174576,
      "step": 6350
    },
    {
      "epoch": 0.44487640862644146,
      "grad_norm": 3.588543176651001,
      "learning_rate": 5.555412259194397e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 102190928,
      "step": 6351
    },
    {
      "epoch": 0.4449464568721707,
      "grad_norm": 3.5698747634887695,
      "learning_rate": 5.5547124343257454e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 102207072,
      "step": 6352
    },
    {
      "epoch": 0.4450165051179,
      "grad_norm": 3.4842190742492676,
      "learning_rate": 5.5540126094570935e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 102222664,
      "step": 6353
    },
    {
      "epoch": 0.4450865533636292,
      "grad_norm": 3.742471218109131,
      "learning_rate": 5.553312784588442e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 102239048,
      "step": 6354
    },
    {
      "epoch": 0.44515660160935844,
      "grad_norm": 4.108808994293213,
      "learning_rate": 5.552612959719792e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 102255432,
      "step": 6355
    },
    {
      "epoch": 0.4452266498550877,
      "grad_norm": 5.498636722564697,
      "learning_rate": 5.551913134851139e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 102270760,
      "step": 6356
    },
    {
      "epoch": 0.44529669810081696,
      "grad_norm": 3.639392137527466,
      "learning_rate": 5.551213309982487e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 102285560,
      "step": 6357
    },
    {
      "epoch": 0.44536674634654616,
      "grad_norm": 4.534916400909424,
      "learning_rate": 5.5505134851138365e-05,
      "loss": 1.0912,
      "num_input_tokens_seen": 102301456,
      "step": 6358
    },
    {
      "epoch": 0.4454367945922754,
      "grad_norm": 3.961845636367798,
      "learning_rate": 5.5498136602451847e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 102317840,
      "step": 6359
    },
    {
      "epoch": 0.4455068428380047,
      "grad_norm": 3.712111473083496,
      "learning_rate": 5.549113835376533e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 102334104,
      "step": 6360
    },
    {
      "epoch": 0.44557689108373394,
      "grad_norm": 4.441688537597656,
      "learning_rate": 5.548414010507882e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 102350488,
      "step": 6361
    },
    {
      "epoch": 0.44564693932946314,
      "grad_norm": 5.881339073181152,
      "learning_rate": 5.54771418563923e-05,
      "loss": 0.8988,
      "num_input_tokens_seen": 102366872,
      "step": 6362
    },
    {
      "epoch": 0.4457169875751924,
      "grad_norm": 3.9736666679382324,
      "learning_rate": 5.547014360770578e-05,
      "loss": 1.1402,
      "num_input_tokens_seen": 102383256,
      "step": 6363
    },
    {
      "epoch": 0.44578703582092166,
      "grad_norm": 4.064074516296387,
      "learning_rate": 5.546314535901926e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 102399024,
      "step": 6364
    },
    {
      "epoch": 0.4458570840666509,
      "grad_norm": 4.238128662109375,
      "learning_rate": 5.5456147110332765e-05,
      "loss": 0.8919,
      "num_input_tokens_seen": 102415408,
      "step": 6365
    },
    {
      "epoch": 0.4459271323123801,
      "grad_norm": 3.8058905601501465,
      "learning_rate": 5.544914886164624e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 102430616,
      "step": 6366
    },
    {
      "epoch": 0.4459971805581094,
      "grad_norm": 5.049830436706543,
      "learning_rate": 5.544215061295972e-05,
      "loss": 1.0986,
      "num_input_tokens_seen": 102446448,
      "step": 6367
    },
    {
      "epoch": 0.44606722880383864,
      "grad_norm": 4.0549116134643555,
      "learning_rate": 5.543515236427321e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 102462832,
      "step": 6368
    },
    {
      "epoch": 0.4461372770495679,
      "grad_norm": 4.005105495452881,
      "learning_rate": 5.542815411558669e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 102478688,
      "step": 6369
    },
    {
      "epoch": 0.44620732529529716,
      "grad_norm": 4.298024654388428,
      "learning_rate": 5.5421155866900174e-05,
      "loss": 1.0249,
      "num_input_tokens_seen": 102495072,
      "step": 6370
    },
    {
      "epoch": 0.44627737354102637,
      "grad_norm": 4.816470623016357,
      "learning_rate": 5.541415761821367e-05,
      "loss": 1.0492,
      "num_input_tokens_seen": 102511456,
      "step": 6371
    },
    {
      "epoch": 0.4463474217867556,
      "grad_norm": 3.89819598197937,
      "learning_rate": 5.5407159369527164e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 102527840,
      "step": 6372
    },
    {
      "epoch": 0.4464174700324849,
      "grad_norm": 3.883650541305542,
      "learning_rate": 5.540016112084063e-05,
      "loss": 0.9475,
      "num_input_tokens_seen": 102543008,
      "step": 6373
    },
    {
      "epoch": 0.44648751827821415,
      "grad_norm": 3.8635551929473877,
      "learning_rate": 5.5393162872154114e-05,
      "loss": 1.2026,
      "num_input_tokens_seen": 102558600,
      "step": 6374
    },
    {
      "epoch": 0.44655756652394335,
      "grad_norm": 4.465150356292725,
      "learning_rate": 5.538616462346761e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 102574984,
      "step": 6375
    },
    {
      "epoch": 0.4466276147696726,
      "grad_norm": 3.8807246685028076,
      "learning_rate": 5.537916637478109e-05,
      "loss": 1.2355,
      "num_input_tokens_seen": 102591368,
      "step": 6376
    },
    {
      "epoch": 0.44669766301540187,
      "grad_norm": 4.526896953582764,
      "learning_rate": 5.5372168126094573e-05,
      "loss": 1.4136,
      "num_input_tokens_seen": 102607520,
      "step": 6377
    },
    {
      "epoch": 0.44676771126113113,
      "grad_norm": 4.905179023742676,
      "learning_rate": 5.536516987740806e-05,
      "loss": 1.1221,
      "num_input_tokens_seen": 102622928,
      "step": 6378
    },
    {
      "epoch": 0.44683775950686033,
      "grad_norm": 3.658268690109253,
      "learning_rate": 5.535817162872156e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 102639312,
      "step": 6379
    },
    {
      "epoch": 0.4469078077525896,
      "grad_norm": 4.125054359436035,
      "learning_rate": 5.5351173380035026e-05,
      "loss": 1.0569,
      "num_input_tokens_seen": 102655104,
      "step": 6380
    },
    {
      "epoch": 0.44697785599831885,
      "grad_norm": 3.6429343223571777,
      "learning_rate": 5.534417513134852e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 102671488,
      "step": 6381
    },
    {
      "epoch": 0.4470479042440481,
      "grad_norm": 4.249630928039551,
      "learning_rate": 5.533717688266201e-05,
      "loss": 1.248,
      "num_input_tokens_seen": 102687872,
      "step": 6382
    },
    {
      "epoch": 0.4471179524897773,
      "grad_norm": 4.6079277992248535,
      "learning_rate": 5.5330178633975485e-05,
      "loss": 1.1774,
      "num_input_tokens_seen": 102704256,
      "step": 6383
    },
    {
      "epoch": 0.4471880007355066,
      "grad_norm": 4.406512260437012,
      "learning_rate": 5.5323180385288966e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 102720640,
      "step": 6384
    },
    {
      "epoch": 0.44725804898123583,
      "grad_norm": 5.140946865081787,
      "learning_rate": 5.5316182136602455e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 102735608,
      "step": 6385
    },
    {
      "epoch": 0.4473280972269651,
      "grad_norm": 5.1392903327941895,
      "learning_rate": 5.530918388791594e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 102751992,
      "step": 6386
    },
    {
      "epoch": 0.4473981454726943,
      "grad_norm": 5.466961860656738,
      "learning_rate": 5.530218563922942e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 102768376,
      "step": 6387
    },
    {
      "epoch": 0.44746819371842356,
      "grad_norm": 3.7714486122131348,
      "learning_rate": 5.5295187390542914e-05,
      "loss": 0.9847,
      "num_input_tokens_seen": 102784328,
      "step": 6388
    },
    {
      "epoch": 0.4475382419641528,
      "grad_norm": 3.665609836578369,
      "learning_rate": 5.52881891418564e-05,
      "loss": 1.0716,
      "num_input_tokens_seen": 102800032,
      "step": 6389
    },
    {
      "epoch": 0.4476082902098821,
      "grad_norm": 6.100143909454346,
      "learning_rate": 5.528119089316988e-05,
      "loss": 0.8844,
      "num_input_tokens_seen": 102816416,
      "step": 6390
    },
    {
      "epoch": 0.4476783384556113,
      "grad_norm": 3.8393003940582275,
      "learning_rate": 5.527419264448336e-05,
      "loss": 1.1134,
      "num_input_tokens_seen": 102832800,
      "step": 6391
    },
    {
      "epoch": 0.44774838670134054,
      "grad_norm": 3.862710952758789,
      "learning_rate": 5.526719439579686e-05,
      "loss": 1.0571,
      "num_input_tokens_seen": 102849128,
      "step": 6392
    },
    {
      "epoch": 0.4478184349470698,
      "grad_norm": 4.032309055328369,
      "learning_rate": 5.526019614711033e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 102865512,
      "step": 6393
    },
    {
      "epoch": 0.44788848319279906,
      "grad_norm": 4.250918388366699,
      "learning_rate": 5.525319789842381e-05,
      "loss": 0.8542,
      "num_input_tokens_seen": 102881896,
      "step": 6394
    },
    {
      "epoch": 0.44795853143852826,
      "grad_norm": 3.8701565265655518,
      "learning_rate": 5.524619964973731e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 102898192,
      "step": 6395
    },
    {
      "epoch": 0.4480285796842575,
      "grad_norm": 4.2415571212768555,
      "learning_rate": 5.52392014010508e-05,
      "loss": 1.0586,
      "num_input_tokens_seen": 102914192,
      "step": 6396
    },
    {
      "epoch": 0.4480986279299868,
      "grad_norm": 3.937345027923584,
      "learning_rate": 5.5232203152364284e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 102929976,
      "step": 6397
    },
    {
      "epoch": 0.44816867617571604,
      "grad_norm": 4.339337348937988,
      "learning_rate": 5.5225204903677766e-05,
      "loss": 1.2494,
      "num_input_tokens_seen": 102946360,
      "step": 6398
    },
    {
      "epoch": 0.44823872442144524,
      "grad_norm": 4.744752883911133,
      "learning_rate": 5.5218206654991255e-05,
      "loss": 1.1717,
      "num_input_tokens_seen": 102962744,
      "step": 6399
    },
    {
      "epoch": 0.4483087726671745,
      "grad_norm": 3.6590077877044678,
      "learning_rate": 5.521120840630473e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 102978456,
      "step": 6400
    },
    {
      "epoch": 0.4483087726671745,
      "eval_loss": 1.1259907484054565,
      "eval_runtime": 0.159,
      "eval_samples_per_second": 6.289,
      "eval_steps_per_second": 6.289,
      "num_input_tokens_seen": 102978456,
      "step": 6400
    },
    {
      "epoch": 0.44837882091290376,
      "grad_norm": 4.073358535766602,
      "learning_rate": 5.5204210157618205e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 102994280,
      "step": 6401
    },
    {
      "epoch": 0.448448869158633,
      "grad_norm": 6.3949480056762695,
      "learning_rate": 5.51972119089317e-05,
      "loss": 1.0338,
      "num_input_tokens_seen": 103010664,
      "step": 6402
    },
    {
      "epoch": 0.4485189174043622,
      "grad_norm": 4.09867525100708,
      "learning_rate": 5.5190213660245195e-05,
      "loss": 1.175,
      "num_input_tokens_seen": 103027048,
      "step": 6403
    },
    {
      "epoch": 0.4485889656500915,
      "grad_norm": 3.672407865524292,
      "learning_rate": 5.518321541155868e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 103043432,
      "step": 6404
    },
    {
      "epoch": 0.44865901389582075,
      "grad_norm": 3.5733370780944824,
      "learning_rate": 5.517621716287216e-05,
      "loss": 0.9414,
      "num_input_tokens_seen": 103059816,
      "step": 6405
    },
    {
      "epoch": 0.44872906214155,
      "grad_norm": 5.21142578125,
      "learning_rate": 5.516921891418565e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 103076200,
      "step": 6406
    },
    {
      "epoch": 0.44879911038727927,
      "grad_norm": 3.4936230182647705,
      "learning_rate": 5.516222066549913e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 103092040,
      "step": 6407
    },
    {
      "epoch": 0.44886915863300847,
      "grad_norm": 4.558346271514893,
      "learning_rate": 5.515522241681261e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 103108424,
      "step": 6408
    },
    {
      "epoch": 0.44893920687873773,
      "grad_norm": 5.485194206237793,
      "learning_rate": 5.514822416812611e-05,
      "loss": 1.0012,
      "num_input_tokens_seen": 103123544,
      "step": 6409
    },
    {
      "epoch": 0.449009255124467,
      "grad_norm": 4.365593433380127,
      "learning_rate": 5.5141225919439575e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 103139928,
      "step": 6410
    },
    {
      "epoch": 0.44907930337019625,
      "grad_norm": 6.034286022186279,
      "learning_rate": 5.513422767075306e-05,
      "loss": 1.1408,
      "num_input_tokens_seen": 103154960,
      "step": 6411
    },
    {
      "epoch": 0.44914935161592545,
      "grad_norm": 3.88476300239563,
      "learning_rate": 5.512722942206655e-05,
      "loss": 0.8513,
      "num_input_tokens_seen": 103169984,
      "step": 6412
    },
    {
      "epoch": 0.4492193998616547,
      "grad_norm": 3.760528087615967,
      "learning_rate": 5.512023117338004e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 103186296,
      "step": 6413
    },
    {
      "epoch": 0.44928944810738397,
      "grad_norm": 3.779690980911255,
      "learning_rate": 5.511323292469352e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 103202680,
      "step": 6414
    },
    {
      "epoch": 0.44935949635311323,
      "grad_norm": 3.6536929607391357,
      "learning_rate": 5.5106234676007004e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 103218656,
      "step": 6415
    },
    {
      "epoch": 0.44942954459884243,
      "grad_norm": 3.9258713722229004,
      "learning_rate": 5.50992364273205e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 103234872,
      "step": 6416
    },
    {
      "epoch": 0.4494995928445717,
      "grad_norm": 4.860123634338379,
      "learning_rate": 5.509223817863397e-05,
      "loss": 1.1094,
      "num_input_tokens_seen": 103251000,
      "step": 6417
    },
    {
      "epoch": 0.44956964109030095,
      "grad_norm": 4.924446105957031,
      "learning_rate": 5.508523992994745e-05,
      "loss": 1.1265,
      "num_input_tokens_seen": 103267296,
      "step": 6418
    },
    {
      "epoch": 0.4496396893360302,
      "grad_norm": 4.334608554840088,
      "learning_rate": 5.507824168126096e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 103283440,
      "step": 6419
    },
    {
      "epoch": 0.4497097375817594,
      "grad_norm": 4.686522483825684,
      "learning_rate": 5.507124343257444e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 103299760,
      "step": 6420
    },
    {
      "epoch": 0.4497797858274887,
      "grad_norm": 4.797657012939453,
      "learning_rate": 5.506424518388792e-05,
      "loss": 1.1277,
      "num_input_tokens_seen": 103316144,
      "step": 6421
    },
    {
      "epoch": 0.44984983407321794,
      "grad_norm": 3.443018674850464,
      "learning_rate": 5.5057246935201404e-05,
      "loss": 0.943,
      "num_input_tokens_seen": 103331712,
      "step": 6422
    },
    {
      "epoch": 0.4499198823189472,
      "grad_norm": 4.118574619293213,
      "learning_rate": 5.505024868651489e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 103348096,
      "step": 6423
    },
    {
      "epoch": 0.4499899305646764,
      "grad_norm": 4.0539937019348145,
      "learning_rate": 5.5043250437828375e-05,
      "loss": 0.8437,
      "num_input_tokens_seen": 103364480,
      "step": 6424
    },
    {
      "epoch": 0.45005997881040566,
      "grad_norm": 4.269721031188965,
      "learning_rate": 5.503625218914187e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 103380120,
      "step": 6425
    },
    {
      "epoch": 0.4501300270561349,
      "grad_norm": 4.6834516525268555,
      "learning_rate": 5.502925394045535e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 103396504,
      "step": 6426
    },
    {
      "epoch": 0.4502000753018642,
      "grad_norm": 4.42267370223999,
      "learning_rate": 5.5022255691768834e-05,
      "loss": 1.0416,
      "num_input_tokens_seen": 103412632,
      "step": 6427
    },
    {
      "epoch": 0.4502701235475934,
      "grad_norm": 4.8119797706604,
      "learning_rate": 5.5015257443082315e-05,
      "loss": 1.2585,
      "num_input_tokens_seen": 103428128,
      "step": 6428
    },
    {
      "epoch": 0.45034017179332264,
      "grad_norm": 4.170595169067383,
      "learning_rate": 5.50082591943958e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 103444088,
      "step": 6429
    },
    {
      "epoch": 0.4504102200390519,
      "grad_norm": 3.7060906887054443,
      "learning_rate": 5.5001260945709286e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 103460456,
      "step": 6430
    },
    {
      "epoch": 0.45048026828478116,
      "grad_norm": 4.4231977462768555,
      "learning_rate": 5.499426269702277e-05,
      "loss": 1.043,
      "num_input_tokens_seen": 103476840,
      "step": 6431
    },
    {
      "epoch": 0.45055031653051036,
      "grad_norm": 4.086833477020264,
      "learning_rate": 5.498726444833625e-05,
      "loss": 1.2797,
      "num_input_tokens_seen": 103492808,
      "step": 6432
    },
    {
      "epoch": 0.4506203647762396,
      "grad_norm": 3.912932872772217,
      "learning_rate": 5.4980266199649745e-05,
      "loss": 0.9846,
      "num_input_tokens_seen": 103508672,
      "step": 6433
    },
    {
      "epoch": 0.4506904130219689,
      "grad_norm": 3.6088106632232666,
      "learning_rate": 5.4973267950963227e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 103525056,
      "step": 6434
    },
    {
      "epoch": 0.45076046126769814,
      "grad_norm": 4.725728511810303,
      "learning_rate": 5.4966269702276715e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 103541440,
      "step": 6435
    },
    {
      "epoch": 0.45083050951342735,
      "grad_norm": 6.745354175567627,
      "learning_rate": 5.49592714535902e-05,
      "loss": 0.9549,
      "num_input_tokens_seen": 103556264,
      "step": 6436
    },
    {
      "epoch": 0.4509005577591566,
      "grad_norm": 4.462937355041504,
      "learning_rate": 5.495227320490368e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 103571576,
      "step": 6437
    },
    {
      "epoch": 0.45097060600488587,
      "grad_norm": 4.77189826965332,
      "learning_rate": 5.494527495621716e-05,
      "loss": 1.2534,
      "num_input_tokens_seen": 103587360,
      "step": 6438
    },
    {
      "epoch": 0.4510406542506151,
      "grad_norm": 5.734838962554932,
      "learning_rate": 5.493827670753064e-05,
      "loss": 1.0698,
      "num_input_tokens_seen": 103603744,
      "step": 6439
    },
    {
      "epoch": 0.4511107024963444,
      "grad_norm": 4.221588134765625,
      "learning_rate": 5.493127845884414e-05,
      "loss": 1.1712,
      "num_input_tokens_seen": 103620128,
      "step": 6440
    },
    {
      "epoch": 0.4511807507420736,
      "grad_norm": 3.894184112548828,
      "learning_rate": 5.492428021015762e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 103636512,
      "step": 6441
    },
    {
      "epoch": 0.45125079898780285,
      "grad_norm": 4.208652496337891,
      "learning_rate": 5.4917281961471115e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 103652824,
      "step": 6442
    },
    {
      "epoch": 0.4513208472335321,
      "grad_norm": 3.4426159858703613,
      "learning_rate": 5.49102837127846e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 103669000,
      "step": 6443
    },
    {
      "epoch": 0.45139089547926137,
      "grad_norm": 3.628229856491089,
      "learning_rate": 5.490328546409808e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 103684792,
      "step": 6444
    },
    {
      "epoch": 0.45146094372499057,
      "grad_norm": 6.008549690246582,
      "learning_rate": 5.489628721541156e-05,
      "loss": 1.2477,
      "num_input_tokens_seen": 103701176,
      "step": 6445
    },
    {
      "epoch": 0.45153099197071983,
      "grad_norm": 4.023336887359619,
      "learning_rate": 5.488928896672506e-05,
      "loss": 1.2522,
      "num_input_tokens_seen": 103716488,
      "step": 6446
    },
    {
      "epoch": 0.4516010402164491,
      "grad_norm": 3.6931705474853516,
      "learning_rate": 5.488229071803853e-05,
      "loss": 0.9123,
      "num_input_tokens_seen": 103732560,
      "step": 6447
    },
    {
      "epoch": 0.45167108846217835,
      "grad_norm": 3.7356324195861816,
      "learning_rate": 5.487529246935201e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 103748512,
      "step": 6448
    },
    {
      "epoch": 0.45174113670790755,
      "grad_norm": 5.176403045654297,
      "learning_rate": 5.486829422066551e-05,
      "loss": 1.0758,
      "num_input_tokens_seen": 103764896,
      "step": 6449
    },
    {
      "epoch": 0.4518111849536368,
      "grad_norm": 4.492616176605225,
      "learning_rate": 5.486129597197899e-05,
      "loss": 1.2558,
      "num_input_tokens_seen": 103781280,
      "step": 6450
    },
    {
      "epoch": 0.4518812331993661,
      "grad_norm": 4.058090686798096,
      "learning_rate": 5.485429772329247e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 103797664,
      "step": 6451
    },
    {
      "epoch": 0.45195128144509533,
      "grad_norm": 3.6828136444091797,
      "learning_rate": 5.484729947460596e-05,
      "loss": 1.0972,
      "num_input_tokens_seen": 103813912,
      "step": 6452
    },
    {
      "epoch": 0.45202132969082454,
      "grad_norm": 3.4555649757385254,
      "learning_rate": 5.484030122591944e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 103830296,
      "step": 6453
    },
    {
      "epoch": 0.4520913779365538,
      "grad_norm": 4.12479305267334,
      "learning_rate": 5.4833302977232924e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 103846272,
      "step": 6454
    },
    {
      "epoch": 0.45216142618228305,
      "grad_norm": 4.1249237060546875,
      "learning_rate": 5.4826304728546406e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 103862408,
      "step": 6455
    },
    {
      "epoch": 0.4522314744280123,
      "grad_norm": 4.026651859283447,
      "learning_rate": 5.481930647985989e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 103878592,
      "step": 6456
    },
    {
      "epoch": 0.4523015226737415,
      "grad_norm": 4.157918453216553,
      "learning_rate": 5.481230823117338e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 103894512,
      "step": 6457
    },
    {
      "epoch": 0.4523715709194708,
      "grad_norm": 3.4476771354675293,
      "learning_rate": 5.4805309982486865e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 103910728,
      "step": 6458
    },
    {
      "epoch": 0.45244161916520004,
      "grad_norm": 5.755035877227783,
      "learning_rate": 5.479831173380035e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 103927112,
      "step": 6459
    },
    {
      "epoch": 0.4525116674109293,
      "grad_norm": 3.774343967437744,
      "learning_rate": 5.4791313485113835e-05,
      "loss": 1.1775,
      "num_input_tokens_seen": 103943496,
      "step": 6460
    },
    {
      "epoch": 0.4525817156566585,
      "grad_norm": 3.8584420680999756,
      "learning_rate": 5.478431523642732e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 103959880,
      "step": 6461
    },
    {
      "epoch": 0.45265176390238776,
      "grad_norm": 3.545832872390747,
      "learning_rate": 5.477731698774081e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 103976264,
      "step": 6462
    },
    {
      "epoch": 0.452721812148117,
      "grad_norm": 4.018779277801514,
      "learning_rate": 5.477031873905431e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 103991720,
      "step": 6463
    },
    {
      "epoch": 0.4527918603938463,
      "grad_norm": 3.966514825820923,
      "learning_rate": 5.476332049036779e-05,
      "loss": 1.1632,
      "num_input_tokens_seen": 104007488,
      "step": 6464
    },
    {
      "epoch": 0.4528619086395755,
      "grad_norm": 3.8280792236328125,
      "learning_rate": 5.475632224168126e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 104023096,
      "step": 6465
    },
    {
      "epoch": 0.45293195688530474,
      "grad_norm": 6.540561676025391,
      "learning_rate": 5.474932399299475e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 104038808,
      "step": 6466
    },
    {
      "epoch": 0.453002005131034,
      "grad_norm": 4.703604221343994,
      "learning_rate": 5.4742325744308235e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 104053944,
      "step": 6467
    },
    {
      "epoch": 0.45307205337676326,
      "grad_norm": 3.950582504272461,
      "learning_rate": 5.473532749562171e-05,
      "loss": 0.9822,
      "num_input_tokens_seen": 104070304,
      "step": 6468
    },
    {
      "epoch": 0.45314210162249247,
      "grad_norm": 5.277374744415283,
      "learning_rate": 5.4728329246935205e-05,
      "loss": 1.1024,
      "num_input_tokens_seen": 104086088,
      "step": 6469
    },
    {
      "epoch": 0.4532121498682217,
      "grad_norm": 4.449152946472168,
      "learning_rate": 5.47213309982487e-05,
      "loss": 1.2031,
      "num_input_tokens_seen": 104102472,
      "step": 6470
    },
    {
      "epoch": 0.453282198113951,
      "grad_norm": 3.780017852783203,
      "learning_rate": 5.471433274956218e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 104117552,
      "step": 6471
    },
    {
      "epoch": 0.45335224635968024,
      "grad_norm": 3.502319574356079,
      "learning_rate": 5.470733450087565e-05,
      "loss": 1.0531,
      "num_input_tokens_seen": 104133936,
      "step": 6472
    },
    {
      "epoch": 0.45342229460540945,
      "grad_norm": 4.8112311363220215,
      "learning_rate": 5.470033625218915e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 104150320,
      "step": 6473
    },
    {
      "epoch": 0.4534923428511387,
      "grad_norm": 3.640571355819702,
      "learning_rate": 5.4693338003502635e-05,
      "loss": 0.8591,
      "num_input_tokens_seen": 104165544,
      "step": 6474
    },
    {
      "epoch": 0.45356239109686797,
      "grad_norm": 3.796278953552246,
      "learning_rate": 5.468633975481611e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 104181928,
      "step": 6475
    },
    {
      "epoch": 0.4536324393425972,
      "grad_norm": 4.749582767486572,
      "learning_rate": 5.46793415061296e-05,
      "loss": 1.0821,
      "num_input_tokens_seen": 104197168,
      "step": 6476
    },
    {
      "epoch": 0.4537024875883265,
      "grad_norm": 4.0797271728515625,
      "learning_rate": 5.467234325744308e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 104213000,
      "step": 6477
    },
    {
      "epoch": 0.4537725358340557,
      "grad_norm": 4.250664710998535,
      "learning_rate": 5.466534500875656e-05,
      "loss": 1.1786,
      "num_input_tokens_seen": 104229136,
      "step": 6478
    },
    {
      "epoch": 0.45384258407978495,
      "grad_norm": 3.8380579948425293,
      "learning_rate": 5.465834676007006e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 104245520,
      "step": 6479
    },
    {
      "epoch": 0.4539126323255142,
      "grad_norm": 3.7329459190368652,
      "learning_rate": 5.4651348511383546e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 104261904,
      "step": 6480
    },
    {
      "epoch": 0.45398268057124347,
      "grad_norm": 4.495264530181885,
      "learning_rate": 5.464435026269703e-05,
      "loss": 1.1216,
      "num_input_tokens_seen": 104278288,
      "step": 6481
    },
    {
      "epoch": 0.4540527288169727,
      "grad_norm": 3.7195420265197754,
      "learning_rate": 5.46373520140105e-05,
      "loss": 1.1697,
      "num_input_tokens_seen": 104294192,
      "step": 6482
    },
    {
      "epoch": 0.45412277706270193,
      "grad_norm": 5.255592346191406,
      "learning_rate": 5.463035376532399e-05,
      "loss": 1.2848,
      "num_input_tokens_seen": 104308704,
      "step": 6483
    },
    {
      "epoch": 0.4541928253084312,
      "grad_norm": 4.61810302734375,
      "learning_rate": 5.462335551663749e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 104325088,
      "step": 6484
    },
    {
      "epoch": 0.45426287355416045,
      "grad_norm": 7.175589561462402,
      "learning_rate": 5.4616357267950955e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 104341224,
      "step": 6485
    },
    {
      "epoch": 0.45433292179988966,
      "grad_norm": 3.756762742996216,
      "learning_rate": 5.460935901926445e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 104356392,
      "step": 6486
    },
    {
      "epoch": 0.4544029700456189,
      "grad_norm": 3.979435920715332,
      "learning_rate": 5.4602360770577946e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 104372776,
      "step": 6487
    },
    {
      "epoch": 0.4544730182913482,
      "grad_norm": 3.761296272277832,
      "learning_rate": 5.459536252189143e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 104389160,
      "step": 6488
    },
    {
      "epoch": 0.45454306653707743,
      "grad_norm": 5.02775239944458,
      "learning_rate": 5.458836427320491e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 104404768,
      "step": 6489
    },
    {
      "epoch": 0.45461311478280664,
      "grad_norm": 4.161303997039795,
      "learning_rate": 5.45813660245184e-05,
      "loss": 0.9746,
      "num_input_tokens_seen": 104420152,
      "step": 6490
    },
    {
      "epoch": 0.4546831630285359,
      "grad_norm": 3.7053780555725098,
      "learning_rate": 5.457436777583188e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 104435512,
      "step": 6491
    },
    {
      "epoch": 0.45475321127426516,
      "grad_norm": 4.103651523590088,
      "learning_rate": 5.456736952714535e-05,
      "loss": 1.0379,
      "num_input_tokens_seen": 104451896,
      "step": 6492
    },
    {
      "epoch": 0.4548232595199944,
      "grad_norm": 4.212504863739014,
      "learning_rate": 5.456037127845884e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 104468280,
      "step": 6493
    },
    {
      "epoch": 0.4548933077657236,
      "grad_norm": 6.549145221710205,
      "learning_rate": 5.455337302977234e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 104484520,
      "step": 6494
    },
    {
      "epoch": 0.4549633560114529,
      "grad_norm": 3.777740478515625,
      "learning_rate": 5.454637478108582e-05,
      "loss": 1.2379,
      "num_input_tokens_seen": 104500536,
      "step": 6495
    },
    {
      "epoch": 0.45503340425718214,
      "grad_norm": 3.827119827270508,
      "learning_rate": 5.45393765323993e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 104516712,
      "step": 6496
    },
    {
      "epoch": 0.4551034525029114,
      "grad_norm": 4.231398105621338,
      "learning_rate": 5.453237828371279e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 104532792,
      "step": 6497
    },
    {
      "epoch": 0.4551735007486406,
      "grad_norm": 5.237041473388672,
      "learning_rate": 5.452538003502627e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 104549136,
      "step": 6498
    },
    {
      "epoch": 0.45524354899436986,
      "grad_norm": 4.457448482513428,
      "learning_rate": 5.451838178633975e-05,
      "loss": 1.2198,
      "num_input_tokens_seen": 104565320,
      "step": 6499
    },
    {
      "epoch": 0.4553135972400991,
      "grad_norm": 3.7427215576171875,
      "learning_rate": 5.451138353765325e-05,
      "loss": 1.1668,
      "num_input_tokens_seen": 104580744,
      "step": 6500
    },
    {
      "epoch": 0.4553836454858284,
      "grad_norm": 4.094877243041992,
      "learning_rate": 5.450438528896673e-05,
      "loss": 1.1735,
      "num_input_tokens_seen": 104596576,
      "step": 6501
    },
    {
      "epoch": 0.4554536937315576,
      "grad_norm": 4.290172576904297,
      "learning_rate": 5.4497387040280214e-05,
      "loss": 1.2692,
      "num_input_tokens_seen": 104612960,
      "step": 6502
    },
    {
      "epoch": 0.45552374197728684,
      "grad_norm": 3.5206210613250732,
      "learning_rate": 5.4490388791593695e-05,
      "loss": 0.9754,
      "num_input_tokens_seen": 104627664,
      "step": 6503
    },
    {
      "epoch": 0.4555937902230161,
      "grad_norm": 3.7847232818603516,
      "learning_rate": 5.4483390542907184e-05,
      "loss": 1.1653,
      "num_input_tokens_seen": 104644048,
      "step": 6504
    },
    {
      "epoch": 0.45566383846874536,
      "grad_norm": 4.193985939025879,
      "learning_rate": 5.4476392294220666e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 104659664,
      "step": 6505
    },
    {
      "epoch": 0.45573388671447457,
      "grad_norm": 3.819429874420166,
      "learning_rate": 5.446939404553415e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 104676048,
      "step": 6506
    },
    {
      "epoch": 0.4558039349602038,
      "grad_norm": 3.7949306964874268,
      "learning_rate": 5.446239579684764e-05,
      "loss": 1.016,
      "num_input_tokens_seen": 104692432,
      "step": 6507
    },
    {
      "epoch": 0.4558739832059331,
      "grad_norm": 3.880740165710449,
      "learning_rate": 5.4455397548161125e-05,
      "loss": 1.2764,
      "num_input_tokens_seen": 104708616,
      "step": 6508
    },
    {
      "epoch": 0.45594403145166235,
      "grad_norm": 5.389898777008057,
      "learning_rate": 5.444839929947459e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 104724984,
      "step": 6509
    },
    {
      "epoch": 0.4560140796973916,
      "grad_norm": 5.053036689758301,
      "learning_rate": 5.44414010507881e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 104741368,
      "step": 6510
    },
    {
      "epoch": 0.4560841279431208,
      "grad_norm": 4.7108330726623535,
      "learning_rate": 5.4434402802101584e-05,
      "loss": 1.1051,
      "num_input_tokens_seen": 104757752,
      "step": 6511
    },
    {
      "epoch": 0.45615417618885007,
      "grad_norm": 3.8108251094818115,
      "learning_rate": 5.4427404553415066e-05,
      "loss": 0.9524,
      "num_input_tokens_seen": 104774136,
      "step": 6512
    },
    {
      "epoch": 0.45622422443457933,
      "grad_norm": 3.8631815910339355,
      "learning_rate": 5.442040630472854e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 104790288,
      "step": 6513
    },
    {
      "epoch": 0.4562942726803086,
      "grad_norm": 3.745565176010132,
      "learning_rate": 5.441340805604205e-05,
      "loss": 1.086,
      "num_input_tokens_seen": 104806672,
      "step": 6514
    },
    {
      "epoch": 0.4563643209260378,
      "grad_norm": 3.6682205200195312,
      "learning_rate": 5.440640980735552e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 104822824,
      "step": 6515
    },
    {
      "epoch": 0.45643436917176705,
      "grad_norm": 4.422383785247803,
      "learning_rate": 5.439941155866901e-05,
      "loss": 1.059,
      "num_input_tokens_seen": 104839208,
      "step": 6516
    },
    {
      "epoch": 0.4565044174174963,
      "grad_norm": 5.5291242599487305,
      "learning_rate": 5.4392413309982495e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 104855296,
      "step": 6517
    },
    {
      "epoch": 0.45657446566322557,
      "grad_norm": 7.490913391113281,
      "learning_rate": 5.438541506129598e-05,
      "loss": 1.2157,
      "num_input_tokens_seen": 104871440,
      "step": 6518
    },
    {
      "epoch": 0.4566445139089548,
      "grad_norm": 5.1885528564453125,
      "learning_rate": 5.437841681260946e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 104887824,
      "step": 6519
    },
    {
      "epoch": 0.45671456215468403,
      "grad_norm": 4.4618096351623535,
      "learning_rate": 5.437141856392295e-05,
      "loss": 1.34,
      "num_input_tokens_seen": 104904208,
      "step": 6520
    },
    {
      "epoch": 0.4567846104004133,
      "grad_norm": 3.8809101581573486,
      "learning_rate": 5.436442031523643e-05,
      "loss": 1.0538,
      "num_input_tokens_seen": 104920592,
      "step": 6521
    },
    {
      "epoch": 0.45685465864614255,
      "grad_norm": 3.429588794708252,
      "learning_rate": 5.435742206654991e-05,
      "loss": 0.975,
      "num_input_tokens_seen": 104936976,
      "step": 6522
    },
    {
      "epoch": 0.45692470689187176,
      "grad_norm": 3.714005947113037,
      "learning_rate": 5.435042381786339e-05,
      "loss": 0.9339,
      "num_input_tokens_seen": 104953136,
      "step": 6523
    },
    {
      "epoch": 0.456994755137601,
      "grad_norm": 4.082497596740723,
      "learning_rate": 5.4343425569176895e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 104969520,
      "step": 6524
    },
    {
      "epoch": 0.4570648033833303,
      "grad_norm": 5.983520030975342,
      "learning_rate": 5.433642732049037e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 104984096,
      "step": 6525
    },
    {
      "epoch": 0.45713485162905954,
      "grad_norm": 4.282527446746826,
      "learning_rate": 5.432942907180385e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 105000120,
      "step": 6526
    },
    {
      "epoch": 0.45720489987478874,
      "grad_norm": 4.0138726234436035,
      "learning_rate": 5.432243082311734e-05,
      "loss": 1.212,
      "num_input_tokens_seen": 105016504,
      "step": 6527
    },
    {
      "epoch": 0.457274948120518,
      "grad_norm": 7.047135829925537,
      "learning_rate": 5.431543257443082e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 105031360,
      "step": 6528
    },
    {
      "epoch": 0.45734499636624726,
      "grad_norm": 4.966803550720215,
      "learning_rate": 5.4308434325744304e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 105047744,
      "step": 6529
    },
    {
      "epoch": 0.4574150446119765,
      "grad_norm": 3.5288639068603516,
      "learning_rate": 5.43014360770578e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 105064128,
      "step": 6530
    },
    {
      "epoch": 0.4574850928577057,
      "grad_norm": 5.45017147064209,
      "learning_rate": 5.4294437828371295e-05,
      "loss": 1.0761,
      "num_input_tokens_seen": 105080512,
      "step": 6531
    },
    {
      "epoch": 0.457555141103435,
      "grad_norm": 4.392576694488525,
      "learning_rate": 5.428743957968476e-05,
      "loss": 1.1192,
      "num_input_tokens_seen": 105096896,
      "step": 6532
    },
    {
      "epoch": 0.45762518934916424,
      "grad_norm": 3.4714255332946777,
      "learning_rate": 5.428044133099826e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 105113280,
      "step": 6533
    },
    {
      "epoch": 0.4576952375948935,
      "grad_norm": 4.999673366546631,
      "learning_rate": 5.427344308231175e-05,
      "loss": 0.8589,
      "num_input_tokens_seen": 105128640,
      "step": 6534
    },
    {
      "epoch": 0.4577652858406227,
      "grad_norm": 3.9281249046325684,
      "learning_rate": 5.426644483362522e-05,
      "loss": 1.2802,
      "num_input_tokens_seen": 105144176,
      "step": 6535
    },
    {
      "epoch": 0.45783533408635196,
      "grad_norm": 4.223507881164551,
      "learning_rate": 5.42594465849387e-05,
      "loss": 0.9767,
      "num_input_tokens_seen": 105160560,
      "step": 6536
    },
    {
      "epoch": 0.4579053823320812,
      "grad_norm": 3.7774858474731445,
      "learning_rate": 5.425244833625219e-05,
      "loss": 0.8401,
      "num_input_tokens_seen": 105176696,
      "step": 6537
    },
    {
      "epoch": 0.4579754305778105,
      "grad_norm": 3.666398048400879,
      "learning_rate": 5.424545008756569e-05,
      "loss": 1.0199,
      "num_input_tokens_seen": 105192376,
      "step": 6538
    },
    {
      "epoch": 0.4580454788235397,
      "grad_norm": 4.442626476287842,
      "learning_rate": 5.4238451838879156e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 105208192,
      "step": 6539
    },
    {
      "epoch": 0.45811552706926895,
      "grad_norm": 3.8626255989074707,
      "learning_rate": 5.423145359019265e-05,
      "loss": 1.2679,
      "num_input_tokens_seen": 105224576,
      "step": 6540
    },
    {
      "epoch": 0.4581855753149982,
      "grad_norm": 3.713498830795288,
      "learning_rate": 5.422445534150614e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 105240960,
      "step": 6541
    },
    {
      "epoch": 0.45825562356072747,
      "grad_norm": 5.06941032409668,
      "learning_rate": 5.4217457092819615e-05,
      "loss": 1.1006,
      "num_input_tokens_seen": 105257344,
      "step": 6542
    },
    {
      "epoch": 0.4583256718064567,
      "grad_norm": 3.6487746238708496,
      "learning_rate": 5.4210458844133103e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 105273336,
      "step": 6543
    },
    {
      "epoch": 0.45839572005218593,
      "grad_norm": 3.70211124420166,
      "learning_rate": 5.42034605954466e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 105289720,
      "step": 6544
    },
    {
      "epoch": 0.4584657682979152,
      "grad_norm": 5.169928073883057,
      "learning_rate": 5.419646234676007e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 105306104,
      "step": 6545
    },
    {
      "epoch": 0.45853581654364445,
      "grad_norm": 4.101007461547852,
      "learning_rate": 5.418946409807355e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 105322488,
      "step": 6546
    },
    {
      "epoch": 0.4586058647893737,
      "grad_norm": 4.077839374542236,
      "learning_rate": 5.4182465849387044e-05,
      "loss": 0.9885,
      "num_input_tokens_seen": 105338872,
      "step": 6547
    },
    {
      "epoch": 0.4586759130351029,
      "grad_norm": 4.46600341796875,
      "learning_rate": 5.417546760070053e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 105355256,
      "step": 6548
    },
    {
      "epoch": 0.45874596128083217,
      "grad_norm": 3.765453577041626,
      "learning_rate": 5.416846935201401e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 105370928,
      "step": 6549
    },
    {
      "epoch": 0.45881600952656143,
      "grad_norm": 3.913649559020996,
      "learning_rate": 5.4161471103327496e-05,
      "loss": 1.0764,
      "num_input_tokens_seen": 105387312,
      "step": 6550
    },
    {
      "epoch": 0.4588860577722907,
      "grad_norm": 5.323554992675781,
      "learning_rate": 5.415447285464099e-05,
      "loss": 1.0202,
      "num_input_tokens_seen": 105403408,
      "step": 6551
    },
    {
      "epoch": 0.4589561060180199,
      "grad_norm": 3.8482306003570557,
      "learning_rate": 5.414747460595446e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 105419792,
      "step": 6552
    },
    {
      "epoch": 0.45902615426374915,
      "grad_norm": 4.369050025939941,
      "learning_rate": 5.414047635726794e-05,
      "loss": 0.9605,
      "num_input_tokens_seen": 105435240,
      "step": 6553
    },
    {
      "epoch": 0.4590962025094784,
      "grad_norm": 3.4255287647247314,
      "learning_rate": 5.413347810858145e-05,
      "loss": 0.9871,
      "num_input_tokens_seen": 105451568,
      "step": 6554
    },
    {
      "epoch": 0.45916625075520767,
      "grad_norm": 4.246303081512451,
      "learning_rate": 5.412647985989493e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 105467952,
      "step": 6555
    },
    {
      "epoch": 0.4592362990009369,
      "grad_norm": 3.785661220550537,
      "learning_rate": 5.4119481611208414e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 105484336,
      "step": 6556
    },
    {
      "epoch": 0.45930634724666614,
      "grad_norm": 3.661653757095337,
      "learning_rate": 5.4112483362521896e-05,
      "loss": 0.9834,
      "num_input_tokens_seen": 105500720,
      "step": 6557
    },
    {
      "epoch": 0.4593763954923954,
      "grad_norm": 4.362829685211182,
      "learning_rate": 5.4105485113835385e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 105516488,
      "step": 6558
    },
    {
      "epoch": 0.45944644373812465,
      "grad_norm": 3.867062568664551,
      "learning_rate": 5.409848686514885e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 105532736,
      "step": 6559
    },
    {
      "epoch": 0.45951649198385386,
      "grad_norm": 3.8351078033447266,
      "learning_rate": 5.409148861646236e-05,
      "loss": 1.0162,
      "num_input_tokens_seen": 105549120,
      "step": 6560
    },
    {
      "epoch": 0.4595865402295831,
      "grad_norm": 4.525234699249268,
      "learning_rate": 5.4084490367775844e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 105564776,
      "step": 6561
    },
    {
      "epoch": 0.4596565884753124,
      "grad_norm": 3.8182532787323,
      "learning_rate": 5.4077492119089326e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 105581080,
      "step": 6562
    },
    {
      "epoch": 0.45972663672104164,
      "grad_norm": 3.82145619392395,
      "learning_rate": 5.407049387040281e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 105596496,
      "step": 6563
    },
    {
      "epoch": 0.45979668496677084,
      "grad_norm": 4.378223419189453,
      "learning_rate": 5.406349562171629e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 105612200,
      "step": 6564
    },
    {
      "epoch": 0.4598667332125001,
      "grad_norm": 4.628854274749756,
      "learning_rate": 5.405649737302978e-05,
      "loss": 1.1581,
      "num_input_tokens_seen": 105628584,
      "step": 6565
    },
    {
      "epoch": 0.45993678145822936,
      "grad_norm": 5.091843128204346,
      "learning_rate": 5.404949912434326e-05,
      "loss": 1.203,
      "num_input_tokens_seen": 105644968,
      "step": 6566
    },
    {
      "epoch": 0.4600068297039586,
      "grad_norm": 4.4174580574035645,
      "learning_rate": 5.404250087565674e-05,
      "loss": 1.1291,
      "num_input_tokens_seen": 105661352,
      "step": 6567
    },
    {
      "epoch": 0.4600768779496878,
      "grad_norm": 4.136083602905273,
      "learning_rate": 5.403550262697024e-05,
      "loss": 1.183,
      "num_input_tokens_seen": 105677200,
      "step": 6568
    },
    {
      "epoch": 0.4601469261954171,
      "grad_norm": 4.388592720031738,
      "learning_rate": 5.4028504378283705e-05,
      "loss": 1.2882,
      "num_input_tokens_seen": 105693584,
      "step": 6569
    },
    {
      "epoch": 0.46021697444114634,
      "grad_norm": 5.642048358917236,
      "learning_rate": 5.402150612959721e-05,
      "loss": 1.1693,
      "num_input_tokens_seen": 105709688,
      "step": 6570
    },
    {
      "epoch": 0.4602870226868756,
      "grad_norm": 4.107602596282959,
      "learning_rate": 5.401450788091069e-05,
      "loss": 1.2681,
      "num_input_tokens_seen": 105725608,
      "step": 6571
    },
    {
      "epoch": 0.4603570709326048,
      "grad_norm": 6.477549076080322,
      "learning_rate": 5.400750963222417e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 105741992,
      "step": 6572
    },
    {
      "epoch": 0.46042711917833407,
      "grad_norm": 4.7183380126953125,
      "learning_rate": 5.400051138353765e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 105758376,
      "step": 6573
    },
    {
      "epoch": 0.4604971674240633,
      "grad_norm": 4.523158073425293,
      "learning_rate": 5.3993513134851135e-05,
      "loss": 1.0595,
      "num_input_tokens_seen": 105774760,
      "step": 6574
    },
    {
      "epoch": 0.4605672156697926,
      "grad_norm": 3.631554126739502,
      "learning_rate": 5.398651488616463e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 105791144,
      "step": 6575
    },
    {
      "epoch": 0.4606372639155218,
      "grad_norm": 8.034467697143555,
      "learning_rate": 5.397951663747811e-05,
      "loss": 1.0565,
      "num_input_tokens_seen": 105807528,
      "step": 6576
    },
    {
      "epoch": 0.46070731216125105,
      "grad_norm": 3.780055522918701,
      "learning_rate": 5.397251838879161e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 105823592,
      "step": 6577
    },
    {
      "epoch": 0.4607773604069803,
      "grad_norm": 3.975475549697876,
      "learning_rate": 5.396552014010509e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 105839976,
      "step": 6578
    },
    {
      "epoch": 0.46084740865270957,
      "grad_norm": 3.4668362140655518,
      "learning_rate": 5.395852189141857e-05,
      "loss": 0.9884,
      "num_input_tokens_seen": 105856360,
      "step": 6579
    },
    {
      "epoch": 0.4609174568984388,
      "grad_norm": 3.7928245067596436,
      "learning_rate": 5.395152364273205e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 105872744,
      "step": 6580
    },
    {
      "epoch": 0.46098750514416803,
      "grad_norm": 3.8289833068847656,
      "learning_rate": 5.3944525394045555e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 105888528,
      "step": 6581
    },
    {
      "epoch": 0.4610575533898973,
      "grad_norm": 6.435444355010986,
      "learning_rate": 5.393752714535902e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 105903592,
      "step": 6582
    },
    {
      "epoch": 0.46112760163562655,
      "grad_norm": 4.274429798126221,
      "learning_rate": 5.3930528896672505e-05,
      "loss": 1.3321,
      "num_input_tokens_seen": 105919904,
      "step": 6583
    },
    {
      "epoch": 0.4611976498813558,
      "grad_norm": 3.619840145111084,
      "learning_rate": 5.3923530647986e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 105936288,
      "step": 6584
    },
    {
      "epoch": 0.461267698127085,
      "grad_norm": 3.643489122390747,
      "learning_rate": 5.391653239929948e-05,
      "loss": 1.2037,
      "num_input_tokens_seen": 105952624,
      "step": 6585
    },
    {
      "epoch": 0.4613377463728143,
      "grad_norm": 3.5494256019592285,
      "learning_rate": 5.3909534150612964e-05,
      "loss": 1.0352,
      "num_input_tokens_seen": 105968568,
      "step": 6586
    },
    {
      "epoch": 0.46140779461854353,
      "grad_norm": 5.754514694213867,
      "learning_rate": 5.390253590192645e-05,
      "loss": 0.8431,
      "num_input_tokens_seen": 105984952,
      "step": 6587
    },
    {
      "epoch": 0.4614778428642728,
      "grad_norm": 3.9911015033721924,
      "learning_rate": 5.3895537653239934e-05,
      "loss": 0.9876,
      "num_input_tokens_seen": 106001336,
      "step": 6588
    },
    {
      "epoch": 0.461547891110002,
      "grad_norm": 4.1558756828308105,
      "learning_rate": 5.3888539404553416e-05,
      "loss": 1.0537,
      "num_input_tokens_seen": 106016736,
      "step": 6589
    },
    {
      "epoch": 0.46161793935573125,
      "grad_norm": 4.300850868225098,
      "learning_rate": 5.38815411558669e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 106033120,
      "step": 6590
    },
    {
      "epoch": 0.4616879876014605,
      "grad_norm": 6.03284215927124,
      "learning_rate": 5.387454290718038e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 106049504,
      "step": 6591
    },
    {
      "epoch": 0.4617580358471898,
      "grad_norm": 4.091002941131592,
      "learning_rate": 5.3867544658493875e-05,
      "loss": 1.205,
      "num_input_tokens_seen": 106065632,
      "step": 6592
    },
    {
      "epoch": 0.461828084092919,
      "grad_norm": 3.7395520210266113,
      "learning_rate": 5.386054640980736e-05,
      "loss": 0.9516,
      "num_input_tokens_seen": 106081632,
      "step": 6593
    },
    {
      "epoch": 0.46189813233864824,
      "grad_norm": 4.021444797515869,
      "learning_rate": 5.3853548161120845e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 106097376,
      "step": 6594
    },
    {
      "epoch": 0.4619681805843775,
      "grad_norm": 5.202040672302246,
      "learning_rate": 5.384654991243433e-05,
      "loss": 0.935,
      "num_input_tokens_seen": 106113096,
      "step": 6595
    },
    {
      "epoch": 0.46203822883010676,
      "grad_norm": 8.020401000976562,
      "learning_rate": 5.383955166374781e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 106129480,
      "step": 6596
    },
    {
      "epoch": 0.46210827707583596,
      "grad_norm": 4.892960548400879,
      "learning_rate": 5.3832553415061304e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 106145864,
      "step": 6597
    },
    {
      "epoch": 0.4621783253215652,
      "grad_norm": 3.963135004043579,
      "learning_rate": 5.38255551663748e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 106162248,
      "step": 6598
    },
    {
      "epoch": 0.4622483735672945,
      "grad_norm": 5.362968444824219,
      "learning_rate": 5.381855691768827e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 106177616,
      "step": 6599
    },
    {
      "epoch": 0.46231842181302374,
      "grad_norm": 5.272266864776611,
      "learning_rate": 5.381155866900175e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 106194000,
      "step": 6600
    },
    {
      "epoch": 0.46231842181302374,
      "eval_loss": 1.1238571405410767,
      "eval_runtime": 0.1703,
      "eval_samples_per_second": 5.872,
      "eval_steps_per_second": 5.872,
      "num_input_tokens_seen": 106194000,
      "step": 6600
    },
    {
      "epoch": 0.46238847005875294,
      "grad_norm": 6.129757881164551,
      "learning_rate": 5.3804560420315245e-05,
      "loss": 0.9134,
      "num_input_tokens_seen": 106210384,
      "step": 6601
    },
    {
      "epoch": 0.4624585183044822,
      "grad_norm": 4.237639904022217,
      "learning_rate": 5.379756217162873e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 106226240,
      "step": 6602
    },
    {
      "epoch": 0.46252856655021146,
      "grad_norm": 3.4758036136627197,
      "learning_rate": 5.37905639229422e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 106242624,
      "step": 6603
    },
    {
      "epoch": 0.4625986147959407,
      "grad_norm": 4.031625270843506,
      "learning_rate": 5.37835656742557e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 106259008,
      "step": 6604
    },
    {
      "epoch": 0.4626686630416699,
      "grad_norm": 3.9605302810668945,
      "learning_rate": 5.377656742556919e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 106274960,
      "step": 6605
    },
    {
      "epoch": 0.4627387112873992,
      "grad_norm": 3.5777320861816406,
      "learning_rate": 5.3769569176882675e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 106291344,
      "step": 6606
    },
    {
      "epoch": 0.46280875953312844,
      "grad_norm": 3.553462505340576,
      "learning_rate": 5.376257092819614e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 106306184,
      "step": 6607
    },
    {
      "epoch": 0.4628788077788577,
      "grad_norm": 3.745340347290039,
      "learning_rate": 5.3755572679509645e-05,
      "loss": 0.8471,
      "num_input_tokens_seen": 106320360,
      "step": 6608
    },
    {
      "epoch": 0.4629488560245869,
      "grad_norm": 3.7483649253845215,
      "learning_rate": 5.374857443082312e-05,
      "loss": 1.0482,
      "num_input_tokens_seen": 106336664,
      "step": 6609
    },
    {
      "epoch": 0.46301890427031617,
      "grad_norm": 3.675184726715088,
      "learning_rate": 5.37415761821366e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 106353048,
      "step": 6610
    },
    {
      "epoch": 0.4630889525160454,
      "grad_norm": 4.733851432800293,
      "learning_rate": 5.373457793345009e-05,
      "loss": 1.0269,
      "num_input_tokens_seen": 106369432,
      "step": 6611
    },
    {
      "epoch": 0.4631590007617747,
      "grad_norm": 3.9618589878082275,
      "learning_rate": 5.372757968476357e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 106385592,
      "step": 6612
    },
    {
      "epoch": 0.46322904900750395,
      "grad_norm": 3.95268177986145,
      "learning_rate": 5.3720581436077054e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 106401976,
      "step": 6613
    },
    {
      "epoch": 0.46329909725323315,
      "grad_norm": 4.600008010864258,
      "learning_rate": 5.371358318739055e-05,
      "loss": 1.0004,
      "num_input_tokens_seen": 106418360,
      "step": 6614
    },
    {
      "epoch": 0.4633691454989624,
      "grad_norm": 3.6651558876037598,
      "learning_rate": 5.370658493870404e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 106434488,
      "step": 6615
    },
    {
      "epoch": 0.46343919374469167,
      "grad_norm": 4.116913318634033,
      "learning_rate": 5.369958669001752e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 106450872,
      "step": 6616
    },
    {
      "epoch": 0.46350924199042093,
      "grad_norm": 4.44846773147583,
      "learning_rate": 5.3692588441330995e-05,
      "loss": 0.8553,
      "num_input_tokens_seen": 106467256,
      "step": 6617
    },
    {
      "epoch": 0.46357929023615013,
      "grad_norm": 5.590776443481445,
      "learning_rate": 5.3685590192644483e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 106483640,
      "step": 6618
    },
    {
      "epoch": 0.4636493384818794,
      "grad_norm": 5.505274772644043,
      "learning_rate": 5.3678591943957965e-05,
      "loss": 1.2003,
      "num_input_tokens_seen": 106500024,
      "step": 6619
    },
    {
      "epoch": 0.46371938672760865,
      "grad_norm": 3.8726046085357666,
      "learning_rate": 5.367159369527145e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 106516408,
      "step": 6620
    },
    {
      "epoch": 0.4637894349733379,
      "grad_norm": 3.9251434803009033,
      "learning_rate": 5.366459544658494e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 106531504,
      "step": 6621
    },
    {
      "epoch": 0.4638594832190671,
      "grad_norm": 3.9294116497039795,
      "learning_rate": 5.365759719789844e-05,
      "loss": 0.9266,
      "num_input_tokens_seen": 106547888,
      "step": 6622
    },
    {
      "epoch": 0.4639295314647964,
      "grad_norm": 4.324211120605469,
      "learning_rate": 5.365059894921192e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 106564272,
      "step": 6623
    },
    {
      "epoch": 0.46399957971052563,
      "grad_norm": 3.5331010818481445,
      "learning_rate": 5.364360070052539e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 106580608,
      "step": 6624
    },
    {
      "epoch": 0.4640696279562549,
      "grad_norm": 3.642073392868042,
      "learning_rate": 5.363660245183889e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 106596760,
      "step": 6625
    },
    {
      "epoch": 0.4641396762019841,
      "grad_norm": 4.356872081756592,
      "learning_rate": 5.362960420315237e-05,
      "loss": 1.0147,
      "num_input_tokens_seen": 106613144,
      "step": 6626
    },
    {
      "epoch": 0.46420972444771336,
      "grad_norm": 3.66884446144104,
      "learning_rate": 5.362260595446584e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 106629528,
      "step": 6627
    },
    {
      "epoch": 0.4642797726934426,
      "grad_norm": 4.089823246002197,
      "learning_rate": 5.3615607705779335e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 106645864,
      "step": 6628
    },
    {
      "epoch": 0.4643498209391719,
      "grad_norm": 4.411832332611084,
      "learning_rate": 5.360860945709283e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 106662248,
      "step": 6629
    },
    {
      "epoch": 0.4644198691849011,
      "grad_norm": 3.7563977241516113,
      "learning_rate": 5.360161120840631e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 106678632,
      "step": 6630
    },
    {
      "epoch": 0.46448991743063034,
      "grad_norm": 3.6493430137634277,
      "learning_rate": 5.3594612959719794e-05,
      "loss": 0.966,
      "num_input_tokens_seen": 106694552,
      "step": 6631
    },
    {
      "epoch": 0.4645599656763596,
      "grad_norm": 3.6459546089172363,
      "learning_rate": 5.358761471103328e-05,
      "loss": 1.0288,
      "num_input_tokens_seen": 106710544,
      "step": 6632
    },
    {
      "epoch": 0.46463001392208886,
      "grad_norm": 4.07296085357666,
      "learning_rate": 5.3580616462346765e-05,
      "loss": 1.0409,
      "num_input_tokens_seen": 106726928,
      "step": 6633
    },
    {
      "epoch": 0.46470006216781806,
      "grad_norm": 3.623961925506592,
      "learning_rate": 5.357361821366024e-05,
      "loss": 0.8566,
      "num_input_tokens_seen": 106742048,
      "step": 6634
    },
    {
      "epoch": 0.4647701104135473,
      "grad_norm": 3.8658370971679688,
      "learning_rate": 5.356661996497374e-05,
      "loss": 1.1116,
      "num_input_tokens_seen": 106758432,
      "step": 6635
    },
    {
      "epoch": 0.4648401586592766,
      "grad_norm": 7.479616641998291,
      "learning_rate": 5.3559621716287224e-05,
      "loss": 1.0146,
      "num_input_tokens_seen": 106774816,
      "step": 6636
    },
    {
      "epoch": 0.46491020690500584,
      "grad_norm": 5.282004356384277,
      "learning_rate": 5.3552623467600706e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 106791200,
      "step": 6637
    },
    {
      "epoch": 0.46498025515073504,
      "grad_norm": 5.323639392852783,
      "learning_rate": 5.354562521891419e-05,
      "loss": 1.159,
      "num_input_tokens_seen": 106806400,
      "step": 6638
    },
    {
      "epoch": 0.4650503033964643,
      "grad_norm": 3.709852933883667,
      "learning_rate": 5.3538626970227676e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 106822224,
      "step": 6639
    },
    {
      "epoch": 0.46512035164219356,
      "grad_norm": 3.583138942718506,
      "learning_rate": 5.353162872154116e-05,
      "loss": 1.1467,
      "num_input_tokens_seen": 106838608,
      "step": 6640
    },
    {
      "epoch": 0.4651903998879228,
      "grad_norm": 4.027291297912598,
      "learning_rate": 5.352463047285464e-05,
      "loss": 1.1139,
      "num_input_tokens_seen": 106854992,
      "step": 6641
    },
    {
      "epoch": 0.465260448133652,
      "grad_norm": 3.9708850383758545,
      "learning_rate": 5.3517632224168135e-05,
      "loss": 1.0602,
      "num_input_tokens_seen": 106871376,
      "step": 6642
    },
    {
      "epoch": 0.4653304963793813,
      "grad_norm": 5.148803234100342,
      "learning_rate": 5.351063397548162e-05,
      "loss": 1.0852,
      "num_input_tokens_seen": 106887760,
      "step": 6643
    },
    {
      "epoch": 0.46540054462511055,
      "grad_norm": 4.076368808746338,
      "learning_rate": 5.3503635726795085e-05,
      "loss": 1.1273,
      "num_input_tokens_seen": 106904112,
      "step": 6644
    },
    {
      "epoch": 0.4654705928708398,
      "grad_norm": 4.920746803283691,
      "learning_rate": 5.349663747810858e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 106919960,
      "step": 6645
    },
    {
      "epoch": 0.465540641116569,
      "grad_norm": 3.8127434253692627,
      "learning_rate": 5.3489639229422076e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 106935928,
      "step": 6646
    },
    {
      "epoch": 0.46561068936229827,
      "grad_norm": 3.9216270446777344,
      "learning_rate": 5.348264098073556e-05,
      "loss": 1.0585,
      "num_input_tokens_seen": 106952168,
      "step": 6647
    },
    {
      "epoch": 0.46568073760802753,
      "grad_norm": 3.5133566856384277,
      "learning_rate": 5.347564273204903e-05,
      "loss": 0.8579,
      "num_input_tokens_seen": 106968080,
      "step": 6648
    },
    {
      "epoch": 0.4657507858537568,
      "grad_norm": 3.634164333343506,
      "learning_rate": 5.346864448336253e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 106984464,
      "step": 6649
    },
    {
      "epoch": 0.46582083409948605,
      "grad_norm": 3.7191765308380127,
      "learning_rate": 5.346164623467601e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 107000848,
      "step": 6650
    },
    {
      "epoch": 0.46589088234521525,
      "grad_norm": 3.767498254776001,
      "learning_rate": 5.345464798598948e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 107017232,
      "step": 6651
    },
    {
      "epoch": 0.4659609305909445,
      "grad_norm": 3.8340818881988525,
      "learning_rate": 5.344764973730299e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 107033616,
      "step": 6652
    },
    {
      "epoch": 0.46603097883667377,
      "grad_norm": 9.20552921295166,
      "learning_rate": 5.344065148861647e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 107049120,
      "step": 6653
    },
    {
      "epoch": 0.46610102708240303,
      "grad_norm": 4.367069721221924,
      "learning_rate": 5.343365323992995e-05,
      "loss": 1.2988,
      "num_input_tokens_seen": 107063432,
      "step": 6654
    },
    {
      "epoch": 0.46617107532813223,
      "grad_norm": 3.6735596656799316,
      "learning_rate": 5.342665499124343e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 107079712,
      "step": 6655
    },
    {
      "epoch": 0.4662411235738615,
      "grad_norm": 4.066924095153809,
      "learning_rate": 5.341965674255692e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 107096096,
      "step": 6656
    },
    {
      "epoch": 0.46631117181959075,
      "grad_norm": 3.7454941272735596,
      "learning_rate": 5.34126584938704e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 107112480,
      "step": 6657
    },
    {
      "epoch": 0.46638122006532,
      "grad_norm": 3.738274574279785,
      "learning_rate": 5.3405660245183885e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 107128864,
      "step": 6658
    },
    {
      "epoch": 0.4664512683110492,
      "grad_norm": 8.665736198425293,
      "learning_rate": 5.339866199649738e-05,
      "loss": 1.1532,
      "num_input_tokens_seen": 107144536,
      "step": 6659
    },
    {
      "epoch": 0.4665213165567785,
      "grad_norm": 3.8733510971069336,
      "learning_rate": 5.339166374781086e-05,
      "loss": 0.7974,
      "num_input_tokens_seen": 107160664,
      "step": 6660
    },
    {
      "epoch": 0.46659136480250774,
      "grad_norm": 4.000319957733154,
      "learning_rate": 5.3384665499124344e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 107177048,
      "step": 6661
    },
    {
      "epoch": 0.466661413048237,
      "grad_norm": 3.8049557209014893,
      "learning_rate": 5.337766725043783e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 107192496,
      "step": 6662
    },
    {
      "epoch": 0.4667314612939662,
      "grad_norm": 4.009215354919434,
      "learning_rate": 5.3370669001751314e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 107207912,
      "step": 6663
    },
    {
      "epoch": 0.46680150953969546,
      "grad_norm": 6.3007378578186035,
      "learning_rate": 5.3363670753064796e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 107223504,
      "step": 6664
    },
    {
      "epoch": 0.4668715577854247,
      "grad_norm": 3.5798394680023193,
      "learning_rate": 5.335667250437828e-05,
      "loss": 1.0625,
      "num_input_tokens_seen": 107239824,
      "step": 6665
    },
    {
      "epoch": 0.466941606031154,
      "grad_norm": 4.701604843139648,
      "learning_rate": 5.334967425569179e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 107256208,
      "step": 6666
    },
    {
      "epoch": 0.4670116542768832,
      "grad_norm": 3.462380886077881,
      "learning_rate": 5.3342676007005255e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 107272592,
      "step": 6667
    },
    {
      "epoch": 0.46708170252261244,
      "grad_norm": 5.546586513519287,
      "learning_rate": 5.333567775831875e-05,
      "loss": 1.3386,
      "num_input_tokens_seen": 107287520,
      "step": 6668
    },
    {
      "epoch": 0.4671517507683417,
      "grad_norm": 4.677948474884033,
      "learning_rate": 5.332867950963223e-05,
      "loss": 1.1263,
      "num_input_tokens_seen": 107303608,
      "step": 6669
    },
    {
      "epoch": 0.46722179901407096,
      "grad_norm": 3.95694899559021,
      "learning_rate": 5.3321681260945714e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 107319992,
      "step": 6670
    },
    {
      "epoch": 0.46729184725980016,
      "grad_norm": 4.037060737609863,
      "learning_rate": 5.331468301225919e-05,
      "loss": 1.0444,
      "num_input_tokens_seen": 107336376,
      "step": 6671
    },
    {
      "epoch": 0.4673618955055294,
      "grad_norm": 3.5486528873443604,
      "learning_rate": 5.3307684763572684e-05,
      "loss": 0.9887,
      "num_input_tokens_seen": 107352760,
      "step": 6672
    },
    {
      "epoch": 0.4674319437512587,
      "grad_norm": 3.868568181991577,
      "learning_rate": 5.330068651488618e-05,
      "loss": 0.8593,
      "num_input_tokens_seen": 107369144,
      "step": 6673
    },
    {
      "epoch": 0.46750199199698794,
      "grad_norm": 7.702548980712891,
      "learning_rate": 5.329368826619965e-05,
      "loss": 1.188,
      "num_input_tokens_seen": 107385528,
      "step": 6674
    },
    {
      "epoch": 0.46757204024271715,
      "grad_norm": 4.390200614929199,
      "learning_rate": 5.328669001751314e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 107401304,
      "step": 6675
    },
    {
      "epoch": 0.4676420884884464,
      "grad_norm": 3.7440412044525146,
      "learning_rate": 5.327969176882663e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 107417688,
      "step": 6676
    },
    {
      "epoch": 0.46771213673417567,
      "grad_norm": 4.894672870635986,
      "learning_rate": 5.327269352014011e-05,
      "loss": 1.0542,
      "num_input_tokens_seen": 107433816,
      "step": 6677
    },
    {
      "epoch": 0.4677821849799049,
      "grad_norm": 4.762908458709717,
      "learning_rate": 5.326569527145359e-05,
      "loss": 1.1926,
      "num_input_tokens_seen": 107450200,
      "step": 6678
    },
    {
      "epoch": 0.46785223322563413,
      "grad_norm": 4.3587870597839355,
      "learning_rate": 5.325869702276708e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 107466584,
      "step": 6679
    },
    {
      "epoch": 0.4679222814713634,
      "grad_norm": 4.166892051696777,
      "learning_rate": 5.325169877408056e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 107482968,
      "step": 6680
    },
    {
      "epoch": 0.46799232971709265,
      "grad_norm": 4.266642093658447,
      "learning_rate": 5.324470052539404e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 107498192,
      "step": 6681
    },
    {
      "epoch": 0.4680623779628219,
      "grad_norm": 3.3419625759124756,
      "learning_rate": 5.3237702276707536e-05,
      "loss": 1.0262,
      "num_input_tokens_seen": 107514576,
      "step": 6682
    },
    {
      "epoch": 0.46813242620855117,
      "grad_norm": 3.903163433074951,
      "learning_rate": 5.3230704028021025e-05,
      "loss": 1.0785,
      "num_input_tokens_seen": 107530536,
      "step": 6683
    },
    {
      "epoch": 0.46820247445428037,
      "grad_norm": 5.467947959899902,
      "learning_rate": 5.32237057793345e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 107546248,
      "step": 6684
    },
    {
      "epoch": 0.46827252270000963,
      "grad_norm": 3.9213547706604004,
      "learning_rate": 5.321670753064799e-05,
      "loss": 1.1432,
      "num_input_tokens_seen": 107561992,
      "step": 6685
    },
    {
      "epoch": 0.4683425709457389,
      "grad_norm": 5.265954971313477,
      "learning_rate": 5.320970928196147e-05,
      "loss": 0.9934,
      "num_input_tokens_seen": 107578376,
      "step": 6686
    },
    {
      "epoch": 0.46841261919146815,
      "grad_norm": 3.9765655994415283,
      "learning_rate": 5.320271103327495e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 107594680,
      "step": 6687
    },
    {
      "epoch": 0.46848266743719735,
      "grad_norm": 4.261830806732178,
      "learning_rate": 5.3195712784588434e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 107611064,
      "step": 6688
    },
    {
      "epoch": 0.4685527156829266,
      "grad_norm": 7.026014804840088,
      "learning_rate": 5.318871453590194e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 107626112,
      "step": 6689
    },
    {
      "epoch": 0.46862276392865587,
      "grad_norm": 4.726694107055664,
      "learning_rate": 5.3181716287215425e-05,
      "loss": 1.0744,
      "num_input_tokens_seen": 107642496,
      "step": 6690
    },
    {
      "epoch": 0.46869281217438513,
      "grad_norm": 3.6380646228790283,
      "learning_rate": 5.317471803852889e-05,
      "loss": 1.1708,
      "num_input_tokens_seen": 107658880,
      "step": 6691
    },
    {
      "epoch": 0.46876286042011434,
      "grad_norm": 3.5807487964630127,
      "learning_rate": 5.316771978984239e-05,
      "loss": 1.1403,
      "num_input_tokens_seen": 107675256,
      "step": 6692
    },
    {
      "epoch": 0.4688329086658436,
      "grad_norm": 3.9915847778320312,
      "learning_rate": 5.316072154115588e-05,
      "loss": 1.0826,
      "num_input_tokens_seen": 107691016,
      "step": 6693
    },
    {
      "epoch": 0.46890295691157285,
      "grad_norm": 4.012253284454346,
      "learning_rate": 5.3153723292469345e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 107707064,
      "step": 6694
    },
    {
      "epoch": 0.4689730051573021,
      "grad_norm": 3.9562582969665527,
      "learning_rate": 5.314672504378284e-05,
      "loss": 1.0017,
      "num_input_tokens_seen": 107723152,
      "step": 6695
    },
    {
      "epoch": 0.4690430534030313,
      "grad_norm": 4.575549125671387,
      "learning_rate": 5.3139726795096336e-05,
      "loss": 1.0722,
      "num_input_tokens_seen": 107739536,
      "step": 6696
    },
    {
      "epoch": 0.4691131016487606,
      "grad_norm": 3.8225462436676025,
      "learning_rate": 5.313272854640982e-05,
      "loss": 1.0149,
      "num_input_tokens_seen": 107755920,
      "step": 6697
    },
    {
      "epoch": 0.46918314989448984,
      "grad_norm": 3.951275587081909,
      "learning_rate": 5.31257302977233e-05,
      "loss": 0.9675,
      "num_input_tokens_seen": 107772296,
      "step": 6698
    },
    {
      "epoch": 0.4692531981402191,
      "grad_norm": 3.5939061641693115,
      "learning_rate": 5.311873204903678e-05,
      "loss": 0.9912,
      "num_input_tokens_seen": 107788480,
      "step": 6699
    },
    {
      "epoch": 0.4693232463859483,
      "grad_norm": 7.109866619110107,
      "learning_rate": 5.311173380035027e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 107804768,
      "step": 6700
    },
    {
      "epoch": 0.46939329463167756,
      "grad_norm": 3.6135330200195312,
      "learning_rate": 5.3104735551663745e-05,
      "loss": 1.1388,
      "num_input_tokens_seen": 107820632,
      "step": 6701
    },
    {
      "epoch": 0.4694633428774068,
      "grad_norm": 4.7758331298828125,
      "learning_rate": 5.3097737302977234e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 107837016,
      "step": 6702
    },
    {
      "epoch": 0.4695333911231361,
      "grad_norm": 3.7631545066833496,
      "learning_rate": 5.309073905429073e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 107853400,
      "step": 6703
    },
    {
      "epoch": 0.4696034393688653,
      "grad_norm": 5.737015247344971,
      "learning_rate": 5.30837408056042e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 107869784,
      "step": 6704
    },
    {
      "epoch": 0.46967348761459454,
      "grad_norm": 3.845569610595703,
      "learning_rate": 5.307674255691769e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 107885760,
      "step": 6705
    },
    {
      "epoch": 0.4697435358603238,
      "grad_norm": 7.402350902557373,
      "learning_rate": 5.306974430823118e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 107902144,
      "step": 6706
    },
    {
      "epoch": 0.46981358410605306,
      "grad_norm": 4.036012649536133,
      "learning_rate": 5.306274605954466e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 107918528,
      "step": 6707
    },
    {
      "epoch": 0.46988363235178227,
      "grad_norm": 5.720461845397949,
      "learning_rate": 5.3055747810858145e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 107934912,
      "step": 6708
    },
    {
      "epoch": 0.4699536805975115,
      "grad_norm": 4.842574119567871,
      "learning_rate": 5.304874956217163e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 107950800,
      "step": 6709
    },
    {
      "epoch": 0.4700237288432408,
      "grad_norm": 3.787020444869995,
      "learning_rate": 5.304175131348512e-05,
      "loss": 0.8436,
      "num_input_tokens_seen": 107967184,
      "step": 6710
    },
    {
      "epoch": 0.47009377708897004,
      "grad_norm": 4.2691192626953125,
      "learning_rate": 5.303475306479859e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 107983568,
      "step": 6711
    },
    {
      "epoch": 0.47016382533469925,
      "grad_norm": 5.233339786529541,
      "learning_rate": 5.3027754816112086e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 107999952,
      "step": 6712
    },
    {
      "epoch": 0.4702338735804285,
      "grad_norm": 3.421193838119507,
      "learning_rate": 5.302075656742558e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 108016336,
      "step": 6713
    },
    {
      "epoch": 0.47030392182615777,
      "grad_norm": 4.561410427093506,
      "learning_rate": 5.301375831873906e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 108032720,
      "step": 6714
    },
    {
      "epoch": 0.470373970071887,
      "grad_norm": 4.749919891357422,
      "learning_rate": 5.3006760070052545e-05,
      "loss": 1.05,
      "num_input_tokens_seen": 108049104,
      "step": 6715
    },
    {
      "epoch": 0.47044401831761623,
      "grad_norm": 4.774212837219238,
      "learning_rate": 5.299976182136603e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 108065488,
      "step": 6716
    },
    {
      "epoch": 0.4705140665633455,
      "grad_norm": 3.6954824924468994,
      "learning_rate": 5.2992763572679515e-05,
      "loss": 0.9831,
      "num_input_tokens_seen": 108081224,
      "step": 6717
    },
    {
      "epoch": 0.47058411480907475,
      "grad_norm": 5.202620983123779,
      "learning_rate": 5.2985765323993e-05,
      "loss": 1.0583,
      "num_input_tokens_seen": 108097608,
      "step": 6718
    },
    {
      "epoch": 0.470654163054804,
      "grad_norm": 3.7043261528015137,
      "learning_rate": 5.297876707530649e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 108113992,
      "step": 6719
    },
    {
      "epoch": 0.47072421130053327,
      "grad_norm": 4.06228494644165,
      "learning_rate": 5.2971768826619974e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 108130376,
      "step": 6720
    },
    {
      "epoch": 0.4707942595462625,
      "grad_norm": 3.4427239894866943,
      "learning_rate": 5.2964770577933456e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 108146760,
      "step": 6721
    },
    {
      "epoch": 0.47086430779199173,
      "grad_norm": 6.617749214172363,
      "learning_rate": 5.295777232924694e-05,
      "loss": 1.1472,
      "num_input_tokens_seen": 108163144,
      "step": 6722
    },
    {
      "epoch": 0.470934356037721,
      "grad_norm": 3.744797706604004,
      "learning_rate": 5.2950774080560426e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 108179528,
      "step": 6723
    },
    {
      "epoch": 0.47100440428345025,
      "grad_norm": 5.034976005554199,
      "learning_rate": 5.294377583187391e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 108195248,
      "step": 6724
    },
    {
      "epoch": 0.47107445252917945,
      "grad_norm": 3.9690632820129395,
      "learning_rate": 5.293677758318739e-05,
      "loss": 1.2634,
      "num_input_tokens_seen": 108210920,
      "step": 6725
    },
    {
      "epoch": 0.4711445007749087,
      "grad_norm": 3.351450204849243,
      "learning_rate": 5.292977933450087e-05,
      "loss": 0.876,
      "num_input_tokens_seen": 108227304,
      "step": 6726
    },
    {
      "epoch": 0.471214549020638,
      "grad_norm": 3.7437920570373535,
      "learning_rate": 5.292278108581437e-05,
      "loss": 0.9549,
      "num_input_tokens_seen": 108243688,
      "step": 6727
    },
    {
      "epoch": 0.47128459726636723,
      "grad_norm": 6.022392272949219,
      "learning_rate": 5.291578283712785e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 108260072,
      "step": 6728
    },
    {
      "epoch": 0.47135464551209644,
      "grad_norm": 4.407289505004883,
      "learning_rate": 5.290878458844134e-05,
      "loss": 1.0511,
      "num_input_tokens_seen": 108276456,
      "step": 6729
    },
    {
      "epoch": 0.4714246937578257,
      "grad_norm": 3.9509878158569336,
      "learning_rate": 5.290178633975482e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 108291632,
      "step": 6730
    },
    {
      "epoch": 0.47149474200355496,
      "grad_norm": 4.2412285804748535,
      "learning_rate": 5.28947880910683e-05,
      "loss": 1.1903,
      "num_input_tokens_seen": 108308016,
      "step": 6731
    },
    {
      "epoch": 0.4715647902492842,
      "grad_norm": 4.234686374664307,
      "learning_rate": 5.288778984238178e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 108323984,
      "step": 6732
    },
    {
      "epoch": 0.4716348384950134,
      "grad_norm": 4.565019130706787,
      "learning_rate": 5.288079159369529e-05,
      "loss": 1.096,
      "num_input_tokens_seen": 108340368,
      "step": 6733
    },
    {
      "epoch": 0.4717048867407427,
      "grad_norm": 4.805628299713135,
      "learning_rate": 5.287379334500876e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 108356752,
      "step": 6734
    },
    {
      "epoch": 0.47177493498647194,
      "grad_norm": 3.9647700786590576,
      "learning_rate": 5.286679509632224e-05,
      "loss": 1.1555,
      "num_input_tokens_seen": 108372216,
      "step": 6735
    },
    {
      "epoch": 0.4718449832322012,
      "grad_norm": 3.811239004135132,
      "learning_rate": 5.285979684763574e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 108387696,
      "step": 6736
    },
    {
      "epoch": 0.4719150314779304,
      "grad_norm": 4.559319496154785,
      "learning_rate": 5.285279859894922e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 108403944,
      "step": 6737
    },
    {
      "epoch": 0.47198507972365966,
      "grad_norm": 4.727875232696533,
      "learning_rate": 5.2845800350262694e-05,
      "loss": 1.1424,
      "num_input_tokens_seen": 108420328,
      "step": 6738
    },
    {
      "epoch": 0.4720551279693889,
      "grad_norm": 3.8609120845794678,
      "learning_rate": 5.283880210157619e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 108436712,
      "step": 6739
    },
    {
      "epoch": 0.4721251762151182,
      "grad_norm": 3.804370164871216,
      "learning_rate": 5.2831803852889685e-05,
      "loss": 1.0733,
      "num_input_tokens_seen": 108453040,
      "step": 6740
    },
    {
      "epoch": 0.4721952244608474,
      "grad_norm": 3.939620018005371,
      "learning_rate": 5.282480560420315e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 108468880,
      "step": 6741
    },
    {
      "epoch": 0.47226527270657664,
      "grad_norm": 4.376893043518066,
      "learning_rate": 5.2817807355516635e-05,
      "loss": 1.131,
      "num_input_tokens_seen": 108485248,
      "step": 6742
    },
    {
      "epoch": 0.4723353209523059,
      "grad_norm": 5.025060653686523,
      "learning_rate": 5.281080910683014e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 108501040,
      "step": 6743
    },
    {
      "epoch": 0.47240536919803516,
      "grad_norm": 3.524656057357788,
      "learning_rate": 5.280381085814361e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 108516624,
      "step": 6744
    },
    {
      "epoch": 0.47247541744376437,
      "grad_norm": 3.8542211055755615,
      "learning_rate": 5.2796812609457094e-05,
      "loss": 1.0025,
      "num_input_tokens_seen": 108533008,
      "step": 6745
    },
    {
      "epoch": 0.4725454656894936,
      "grad_norm": 3.8751041889190674,
      "learning_rate": 5.278981436077058e-05,
      "loss": 1.1803,
      "num_input_tokens_seen": 108549112,
      "step": 6746
    },
    {
      "epoch": 0.4726155139352229,
      "grad_norm": 4.343238353729248,
      "learning_rate": 5.2782816112084064e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 108564328,
      "step": 6747
    },
    {
      "epoch": 0.47268556218095215,
      "grad_norm": 3.695493698120117,
      "learning_rate": 5.2775817863397546e-05,
      "loss": 0.8834,
      "num_input_tokens_seen": 108580112,
      "step": 6748
    },
    {
      "epoch": 0.47275561042668135,
      "grad_norm": 3.8947877883911133,
      "learning_rate": 5.276881961471104e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 108596136,
      "step": 6749
    },
    {
      "epoch": 0.4728256586724106,
      "grad_norm": 4.2317633628845215,
      "learning_rate": 5.276182136602453e-05,
      "loss": 0.9472,
      "num_input_tokens_seen": 108612520,
      "step": 6750
    },
    {
      "epoch": 0.47289570691813987,
      "grad_norm": 3.608283281326294,
      "learning_rate": 5.2754823117338005e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 108628904,
      "step": 6751
    },
    {
      "epoch": 0.47296575516386913,
      "grad_norm": 4.512143611907959,
      "learning_rate": 5.274782486865149e-05,
      "loss": 1.0156,
      "num_input_tokens_seen": 108644248,
      "step": 6752
    },
    {
      "epoch": 0.4730358034095984,
      "grad_norm": 3.81160044670105,
      "learning_rate": 5.2740826619964976e-05,
      "loss": 1.0496,
      "num_input_tokens_seen": 108660488,
      "step": 6753
    },
    {
      "epoch": 0.4731058516553276,
      "grad_norm": 3.760336399078369,
      "learning_rate": 5.273382837127846e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 108676872,
      "step": 6754
    },
    {
      "epoch": 0.47317589990105685,
      "grad_norm": 3.969651222229004,
      "learning_rate": 5.272683012259194e-05,
      "loss": 1.2213,
      "num_input_tokens_seen": 108693256,
      "step": 6755
    },
    {
      "epoch": 0.4732459481467861,
      "grad_norm": 4.55695915222168,
      "learning_rate": 5.2719831873905435e-05,
      "loss": 0.9125,
      "num_input_tokens_seen": 108709576,
      "step": 6756
    },
    {
      "epoch": 0.47331599639251537,
      "grad_norm": 4.36952018737793,
      "learning_rate": 5.271283362521893e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 108725520,
      "step": 6757
    },
    {
      "epoch": 0.4733860446382446,
      "grad_norm": 4.689207553863525,
      "learning_rate": 5.27058353765324e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 108741744,
      "step": 6758
    },
    {
      "epoch": 0.47345609288397383,
      "grad_norm": 3.5912058353424072,
      "learning_rate": 5.269883712784588e-05,
      "loss": 1.1125,
      "num_input_tokens_seen": 108757952,
      "step": 6759
    },
    {
      "epoch": 0.4735261411297031,
      "grad_norm": 4.725868225097656,
      "learning_rate": 5.269183887915938e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 108774088,
      "step": 6760
    },
    {
      "epoch": 0.47359618937543235,
      "grad_norm": 4.213376045227051,
      "learning_rate": 5.268484063047285e-05,
      "loss": 1.0752,
      "num_input_tokens_seen": 108790472,
      "step": 6761
    },
    {
      "epoch": 0.47366623762116156,
      "grad_norm": 4.116434574127197,
      "learning_rate": 5.267784238178633e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 108806776,
      "step": 6762
    },
    {
      "epoch": 0.4737362858668908,
      "grad_norm": 3.8367996215820312,
      "learning_rate": 5.267084413309983e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 108822416,
      "step": 6763
    },
    {
      "epoch": 0.4738063341126201,
      "grad_norm": 3.609545946121216,
      "learning_rate": 5.266384588441332e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 108838208,
      "step": 6764
    },
    {
      "epoch": 0.47387638235834934,
      "grad_norm": 4.108180522918701,
      "learning_rate": 5.2656847635726805e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 108854592,
      "step": 6765
    },
    {
      "epoch": 0.47394643060407854,
      "grad_norm": 4.884720325469971,
      "learning_rate": 5.2649849387040287e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 108870976,
      "step": 6766
    },
    {
      "epoch": 0.4740164788498078,
      "grad_norm": 4.856875896453857,
      "learning_rate": 5.2642851138353775e-05,
      "loss": 1.2403,
      "num_input_tokens_seen": 108885688,
      "step": 6767
    },
    {
      "epoch": 0.47408652709553706,
      "grad_norm": 3.5622432231903076,
      "learning_rate": 5.263585288966725e-05,
      "loss": 0.9572,
      "num_input_tokens_seen": 108902072,
      "step": 6768
    },
    {
      "epoch": 0.4741565753412663,
      "grad_norm": 5.305510997772217,
      "learning_rate": 5.262885464098073e-05,
      "loss": 1.229,
      "num_input_tokens_seen": 108917848,
      "step": 6769
    },
    {
      "epoch": 0.4742266235869955,
      "grad_norm": 3.729074478149414,
      "learning_rate": 5.2621856392294234e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 108934232,
      "step": 6770
    },
    {
      "epoch": 0.4742966718327248,
      "grad_norm": 4.5915937423706055,
      "learning_rate": 5.26148581436077e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 108949696,
      "step": 6771
    },
    {
      "epoch": 0.47436672007845404,
      "grad_norm": 3.977216958999634,
      "learning_rate": 5.26078598949212e-05,
      "loss": 1.2395,
      "num_input_tokens_seen": 108965848,
      "step": 6772
    },
    {
      "epoch": 0.4744367683241833,
      "grad_norm": 4.012653827667236,
      "learning_rate": 5.260086164623468e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 108982232,
      "step": 6773
    },
    {
      "epoch": 0.4745068165699125,
      "grad_norm": 4.10910701751709,
      "learning_rate": 5.259386339754817e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 108997800,
      "step": 6774
    },
    {
      "epoch": 0.47457686481564176,
      "grad_norm": 5.1765336990356445,
      "learning_rate": 5.258686514886165e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 109013664,
      "step": 6775
    },
    {
      "epoch": 0.474646913061371,
      "grad_norm": 5.6664228439331055,
      "learning_rate": 5.257986690017513e-05,
      "loss": 1.345,
      "num_input_tokens_seen": 109029208,
      "step": 6776
    },
    {
      "epoch": 0.4747169613071003,
      "grad_norm": 6.2354817390441895,
      "learning_rate": 5.257286865148863e-05,
      "loss": 1.1819,
      "num_input_tokens_seen": 109044528,
      "step": 6777
    },
    {
      "epoch": 0.4747870095528295,
      "grad_norm": 3.8308510780334473,
      "learning_rate": 5.2565870402802095e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 109060912,
      "step": 6778
    },
    {
      "epoch": 0.47485705779855875,
      "grad_norm": 4.019093990325928,
      "learning_rate": 5.255887215411558e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 109077296,
      "step": 6779
    },
    {
      "epoch": 0.474927106044288,
      "grad_norm": 6.938348293304443,
      "learning_rate": 5.255187390542907e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 109093680,
      "step": 6780
    },
    {
      "epoch": 0.47499715429001726,
      "grad_norm": 4.200627326965332,
      "learning_rate": 5.254487565674257e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 109110008,
      "step": 6781
    },
    {
      "epoch": 0.47506720253574647,
      "grad_norm": 4.06279993057251,
      "learning_rate": 5.253787740805605e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 109126392,
      "step": 6782
    },
    {
      "epoch": 0.47513725078147573,
      "grad_norm": 4.124377250671387,
      "learning_rate": 5.2530879159369525e-05,
      "loss": 1.2065,
      "num_input_tokens_seen": 109142680,
      "step": 6783
    },
    {
      "epoch": 0.475207299027205,
      "grad_norm": 4.182784557342529,
      "learning_rate": 5.252388091068302e-05,
      "loss": 1.073,
      "num_input_tokens_seen": 109158768,
      "step": 6784
    },
    {
      "epoch": 0.47527734727293425,
      "grad_norm": 4.513407230377197,
      "learning_rate": 5.25168826619965e-05,
      "loss": 1.0392,
      "num_input_tokens_seen": 109175152,
      "step": 6785
    },
    {
      "epoch": 0.4753473955186635,
      "grad_norm": 3.251490354537964,
      "learning_rate": 5.250988441330997e-05,
      "loss": 0.8611,
      "num_input_tokens_seen": 109191056,
      "step": 6786
    },
    {
      "epoch": 0.4754174437643927,
      "grad_norm": 8.621055603027344,
      "learning_rate": 5.250288616462348e-05,
      "loss": 1.0699,
      "num_input_tokens_seen": 109206888,
      "step": 6787
    },
    {
      "epoch": 0.47548749201012197,
      "grad_norm": 4.264245986938477,
      "learning_rate": 5.249588791593696e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 109223272,
      "step": 6788
    },
    {
      "epoch": 0.47555754025585123,
      "grad_norm": 3.6648037433624268,
      "learning_rate": 5.248888966725044e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 109239544,
      "step": 6789
    },
    {
      "epoch": 0.4756275885015805,
      "grad_norm": 4.528952598571777,
      "learning_rate": 5.2481891418563925e-05,
      "loss": 1.1251,
      "num_input_tokens_seen": 109255816,
      "step": 6790
    },
    {
      "epoch": 0.4756976367473097,
      "grad_norm": 4.45644998550415,
      "learning_rate": 5.247489316987741e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 109272200,
      "step": 6791
    },
    {
      "epoch": 0.47576768499303895,
      "grad_norm": 3.8969879150390625,
      "learning_rate": 5.2467894921190895e-05,
      "loss": 0.9836,
      "num_input_tokens_seen": 109288520,
      "step": 6792
    },
    {
      "epoch": 0.4758377332387682,
      "grad_norm": 3.627748727798462,
      "learning_rate": 5.246089667250438e-05,
      "loss": 1.1656,
      "num_input_tokens_seen": 109304624,
      "step": 6793
    },
    {
      "epoch": 0.47590778148449747,
      "grad_norm": 4.493330478668213,
      "learning_rate": 5.245389842381787e-05,
      "loss": 1.2352,
      "num_input_tokens_seen": 109319976,
      "step": 6794
    },
    {
      "epoch": 0.4759778297302267,
      "grad_norm": 3.5947048664093018,
      "learning_rate": 5.2446900175131354e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 109336360,
      "step": 6795
    },
    {
      "epoch": 0.47604787797595594,
      "grad_norm": 4.194823741912842,
      "learning_rate": 5.2439901926444836e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 109351624,
      "step": 6796
    },
    {
      "epoch": 0.4761179262216852,
      "grad_norm": 3.9734160900115967,
      "learning_rate": 5.2432903677758324e-05,
      "loss": 1.2019,
      "num_input_tokens_seen": 109367072,
      "step": 6797
    },
    {
      "epoch": 0.47618797446741445,
      "grad_norm": 4.142136096954346,
      "learning_rate": 5.2425905429071806e-05,
      "loss": 1.1178,
      "num_input_tokens_seen": 109383408,
      "step": 6798
    },
    {
      "epoch": 0.47625802271314366,
      "grad_norm": 4.315369129180908,
      "learning_rate": 5.241890718038529e-05,
      "loss": 1.2254,
      "num_input_tokens_seen": 109398616,
      "step": 6799
    },
    {
      "epoch": 0.4763280709588729,
      "grad_norm": 4.77875280380249,
      "learning_rate": 5.241190893169877e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 109414592,
      "step": 6800
    },
    {
      "epoch": 0.4763280709588729,
      "eval_loss": 1.1252864599227905,
      "eval_runtime": 0.1585,
      "eval_samples_per_second": 6.311,
      "eval_steps_per_second": 6.311,
      "num_input_tokens_seen": 109414592,
      "step": 6800
    },
    {
      "epoch": 0.4763981192046022,
      "grad_norm": 4.054019927978516,
      "learning_rate": 5.2404910683012265e-05,
      "loss": 1.1978,
      "num_input_tokens_seen": 109430896,
      "step": 6801
    },
    {
      "epoch": 0.47646816745033144,
      "grad_norm": 4.0688276290893555,
      "learning_rate": 5.239791243432575e-05,
      "loss": 1.015,
      "num_input_tokens_seen": 109447008,
      "step": 6802
    },
    {
      "epoch": 0.47653821569606064,
      "grad_norm": 4.081553936004639,
      "learning_rate": 5.239091418563924e-05,
      "loss": 1.2566,
      "num_input_tokens_seen": 109463392,
      "step": 6803
    },
    {
      "epoch": 0.4766082639417899,
      "grad_norm": 4.719587326049805,
      "learning_rate": 5.2383915936952724e-05,
      "loss": 1.0577,
      "num_input_tokens_seen": 109478768,
      "step": 6804
    },
    {
      "epoch": 0.47667831218751916,
      "grad_norm": 3.7197132110595703,
      "learning_rate": 5.2376917688266206e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 109494808,
      "step": 6805
    },
    {
      "epoch": 0.4767483604332484,
      "grad_norm": 5.000951290130615,
      "learning_rate": 5.236991943957968e-05,
      "loss": 1.0497,
      "num_input_tokens_seen": 109511192,
      "step": 6806
    },
    {
      "epoch": 0.4768184086789776,
      "grad_norm": 3.9910333156585693,
      "learning_rate": 5.236292119089316e-05,
      "loss": 1.2905,
      "num_input_tokens_seen": 109527576,
      "step": 6807
    },
    {
      "epoch": 0.4768884569247069,
      "grad_norm": 4.522314548492432,
      "learning_rate": 5.235592294220666e-05,
      "loss": 0.9959,
      "num_input_tokens_seen": 109543960,
      "step": 6808
    },
    {
      "epoch": 0.47695850517043614,
      "grad_norm": 3.7235898971557617,
      "learning_rate": 5.234892469352014e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 109560344,
      "step": 6809
    },
    {
      "epoch": 0.4770285534161654,
      "grad_norm": 3.643763303756714,
      "learning_rate": 5.2341926444833635e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 109576728,
      "step": 6810
    },
    {
      "epoch": 0.4770986016618946,
      "grad_norm": 5.52113151550293,
      "learning_rate": 5.233492819614712e-05,
      "loss": 1.2022,
      "num_input_tokens_seen": 109592584,
      "step": 6811
    },
    {
      "epoch": 0.47716864990762387,
      "grad_norm": 4.9974188804626465,
      "learning_rate": 5.23279299474606e-05,
      "loss": 1.1755,
      "num_input_tokens_seen": 109608960,
      "step": 6812
    },
    {
      "epoch": 0.4772386981533531,
      "grad_norm": 5.266491889953613,
      "learning_rate": 5.232093169877408e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 109625104,
      "step": 6813
    },
    {
      "epoch": 0.4773087463990824,
      "grad_norm": 3.9919018745422363,
      "learning_rate": 5.231393345008757e-05,
      "loss": 1.0423,
      "num_input_tokens_seen": 109641488,
      "step": 6814
    },
    {
      "epoch": 0.4773787946448116,
      "grad_norm": 5.361277103424072,
      "learning_rate": 5.230693520140105e-05,
      "loss": 1.215,
      "num_input_tokens_seen": 109657872,
      "step": 6815
    },
    {
      "epoch": 0.47744884289054085,
      "grad_norm": 4.024937629699707,
      "learning_rate": 5.229993695271453e-05,
      "loss": 1.2601,
      "num_input_tokens_seen": 109674256,
      "step": 6816
    },
    {
      "epoch": 0.4775188911362701,
      "grad_norm": 3.7742490768432617,
      "learning_rate": 5.229293870402803e-05,
      "loss": 1.0789,
      "num_input_tokens_seen": 109690576,
      "step": 6817
    },
    {
      "epoch": 0.47758893938199937,
      "grad_norm": 3.622018814086914,
      "learning_rate": 5.228594045534151e-05,
      "loss": 0.8893,
      "num_input_tokens_seen": 109706592,
      "step": 6818
    },
    {
      "epoch": 0.47765898762772857,
      "grad_norm": 4.550981044769287,
      "learning_rate": 5.227894220665499e-05,
      "loss": 1.31,
      "num_input_tokens_seen": 109722976,
      "step": 6819
    },
    {
      "epoch": 0.47772903587345783,
      "grad_norm": 3.8553786277770996,
      "learning_rate": 5.227194395796848e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 109738920,
      "step": 6820
    },
    {
      "epoch": 0.4777990841191871,
      "grad_norm": 3.7159841060638428,
      "learning_rate": 5.226494570928196e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 109755304,
      "step": 6821
    },
    {
      "epoch": 0.47786913236491635,
      "grad_norm": 5.884495258331299,
      "learning_rate": 5.2257947460595444e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 109771576,
      "step": 6822
    },
    {
      "epoch": 0.4779391806106456,
      "grad_norm": 3.7047083377838135,
      "learning_rate": 5.2250949211908926e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 109787872,
      "step": 6823
    },
    {
      "epoch": 0.4780092288563748,
      "grad_norm": 3.485847234725952,
      "learning_rate": 5.224395096322241e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 109803384,
      "step": 6824
    },
    {
      "epoch": 0.47807927710210407,
      "grad_norm": 3.9222450256347656,
      "learning_rate": 5.223695271453592e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 109818704,
      "step": 6825
    },
    {
      "epoch": 0.47814932534783333,
      "grad_norm": 4.232855796813965,
      "learning_rate": 5.2229954465849385e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 109835088,
      "step": 6826
    },
    {
      "epoch": 0.4782193735935626,
      "grad_norm": 3.5413403511047363,
      "learning_rate": 5.222295621716288e-05,
      "loss": 0.9407,
      "num_input_tokens_seen": 109851472,
      "step": 6827
    },
    {
      "epoch": 0.4782894218392918,
      "grad_norm": 4.55118989944458,
      "learning_rate": 5.2215957968476356e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 109867856,
      "step": 6828
    },
    {
      "epoch": 0.47835947008502105,
      "grad_norm": 3.691756010055542,
      "learning_rate": 5.220895971978984e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 109884240,
      "step": 6829
    },
    {
      "epoch": 0.4784295183307503,
      "grad_norm": 3.588829755783081,
      "learning_rate": 5.220196147110333e-05,
      "loss": 1.0665,
      "num_input_tokens_seen": 109900624,
      "step": 6830
    },
    {
      "epoch": 0.4784995665764796,
      "grad_norm": 4.766005516052246,
      "learning_rate": 5.219496322241683e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 109917008,
      "step": 6831
    },
    {
      "epoch": 0.4785696148222088,
      "grad_norm": 3.7234201431274414,
      "learning_rate": 5.218796497373031e-05,
      "loss": 1.0377,
      "num_input_tokens_seen": 109933392,
      "step": 6832
    },
    {
      "epoch": 0.47863966306793804,
      "grad_norm": 3.434387683868408,
      "learning_rate": 5.218096672504378e-05,
      "loss": 0.874,
      "num_input_tokens_seen": 109949776,
      "step": 6833
    },
    {
      "epoch": 0.4787097113136673,
      "grad_norm": 3.7484259605407715,
      "learning_rate": 5.2173968476357274e-05,
      "loss": 0.9365,
      "num_input_tokens_seen": 109966016,
      "step": 6834
    },
    {
      "epoch": 0.47877975955939656,
      "grad_norm": 5.821316719055176,
      "learning_rate": 5.216697022767076e-05,
      "loss": 0.9894,
      "num_input_tokens_seen": 109981168,
      "step": 6835
    },
    {
      "epoch": 0.47884980780512576,
      "grad_norm": 5.2646484375,
      "learning_rate": 5.215997197898424e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 109996648,
      "step": 6836
    },
    {
      "epoch": 0.478919856050855,
      "grad_norm": 5.125279426574707,
      "learning_rate": 5.2152973730297726e-05,
      "loss": 0.9451,
      "num_input_tokens_seen": 110013032,
      "step": 6837
    },
    {
      "epoch": 0.4789899042965843,
      "grad_norm": 4.917844295501709,
      "learning_rate": 5.214597548161121e-05,
      "loss": 1.1573,
      "num_input_tokens_seen": 110029040,
      "step": 6838
    },
    {
      "epoch": 0.47905995254231354,
      "grad_norm": 3.6937522888183594,
      "learning_rate": 5.21389772329247e-05,
      "loss": 1.0922,
      "num_input_tokens_seen": 110045032,
      "step": 6839
    },
    {
      "epoch": 0.47913000078804274,
      "grad_norm": 4.9768757820129395,
      "learning_rate": 5.213197898423817e-05,
      "loss": 1.3347,
      "num_input_tokens_seen": 110061416,
      "step": 6840
    },
    {
      "epoch": 0.479200049033772,
      "grad_norm": 5.775148391723633,
      "learning_rate": 5.212498073555167e-05,
      "loss": 1.1443,
      "num_input_tokens_seen": 110077800,
      "step": 6841
    },
    {
      "epoch": 0.47927009727950126,
      "grad_norm": 4.3342766761779785,
      "learning_rate": 5.2117982486865155e-05,
      "loss": 1.0604,
      "num_input_tokens_seen": 110093848,
      "step": 6842
    },
    {
      "epoch": 0.4793401455252305,
      "grad_norm": 3.6098031997680664,
      "learning_rate": 5.211098423817863e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 110110232,
      "step": 6843
    },
    {
      "epoch": 0.4794101937709597,
      "grad_norm": 3.7780818939208984,
      "learning_rate": 5.210398598949212e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 110126584,
      "step": 6844
    },
    {
      "epoch": 0.479480242016689,
      "grad_norm": 3.732302188873291,
      "learning_rate": 5.2096987740805614e-05,
      "loss": 0.9158,
      "num_input_tokens_seen": 110142968,
      "step": 6845
    },
    {
      "epoch": 0.47955029026241824,
      "grad_norm": 4.920741558074951,
      "learning_rate": 5.208998949211908e-05,
      "loss": 0.9931,
      "num_input_tokens_seen": 110159352,
      "step": 6846
    },
    {
      "epoch": 0.4796203385081475,
      "grad_norm": 3.847682476043701,
      "learning_rate": 5.208299124343258e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 110175736,
      "step": 6847
    },
    {
      "epoch": 0.4796903867538767,
      "grad_norm": 3.8941121101379395,
      "learning_rate": 5.207599299474607e-05,
      "loss": 1.0896,
      "num_input_tokens_seen": 110192040,
      "step": 6848
    },
    {
      "epoch": 0.47976043499960597,
      "grad_norm": 4.254310131072998,
      "learning_rate": 5.2068994746059555e-05,
      "loss": 1.1701,
      "num_input_tokens_seen": 110208304,
      "step": 6849
    },
    {
      "epoch": 0.4798304832453352,
      "grad_norm": 3.85739803314209,
      "learning_rate": 5.206199649737302e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 110224688,
      "step": 6850
    },
    {
      "epoch": 0.4799005314910645,
      "grad_norm": 4.137633323669434,
      "learning_rate": 5.205499824868651e-05,
      "loss": 1.2111,
      "num_input_tokens_seen": 110240160,
      "step": 6851
    },
    {
      "epoch": 0.4799705797367937,
      "grad_norm": 3.827974557876587,
      "learning_rate": 5.204800000000001e-05,
      "loss": 0.9639,
      "num_input_tokens_seen": 110255952,
      "step": 6852
    },
    {
      "epoch": 0.48004062798252295,
      "grad_norm": 4.506080627441406,
      "learning_rate": 5.2041001751313475e-05,
      "loss": 1.0435,
      "num_input_tokens_seen": 110272336,
      "step": 6853
    },
    {
      "epoch": 0.4801106762282522,
      "grad_norm": 3.4824750423431396,
      "learning_rate": 5.203400350262697e-05,
      "loss": 0.8792,
      "num_input_tokens_seen": 110288720,
      "step": 6854
    },
    {
      "epoch": 0.48018072447398147,
      "grad_norm": 3.319546937942505,
      "learning_rate": 5.2027005253940466e-05,
      "loss": 0.9861,
      "num_input_tokens_seen": 110304984,
      "step": 6855
    },
    {
      "epoch": 0.48025077271971073,
      "grad_norm": 5.543242454528809,
      "learning_rate": 5.202000700525395e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 110320488,
      "step": 6856
    },
    {
      "epoch": 0.48032082096543993,
      "grad_norm": 6.7765069007873535,
      "learning_rate": 5.201300875656743e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 110336872,
      "step": 6857
    },
    {
      "epoch": 0.4803908692111692,
      "grad_norm": 3.5764353275299072,
      "learning_rate": 5.200601050788092e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 110353160,
      "step": 6858
    },
    {
      "epoch": 0.48046091745689845,
      "grad_norm": 4.938530921936035,
      "learning_rate": 5.19990122591944e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 110369544,
      "step": 6859
    },
    {
      "epoch": 0.4805309657026277,
      "grad_norm": 3.5447168350219727,
      "learning_rate": 5.1992014010507875e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 110385928,
      "step": 6860
    },
    {
      "epoch": 0.4806010139483569,
      "grad_norm": 4.1170220375061035,
      "learning_rate": 5.1985015761821364e-05,
      "loss": 1.156,
      "num_input_tokens_seen": 110402224,
      "step": 6861
    },
    {
      "epoch": 0.4806710621940862,
      "grad_norm": 3.6147382259368896,
      "learning_rate": 5.197801751313486e-05,
      "loss": 1.0212,
      "num_input_tokens_seen": 110418608,
      "step": 6862
    },
    {
      "epoch": 0.48074111043981543,
      "grad_norm": 3.745072841644287,
      "learning_rate": 5.197101926444834e-05,
      "loss": 1.1518,
      "num_input_tokens_seen": 110434792,
      "step": 6863
    },
    {
      "epoch": 0.4808111586855447,
      "grad_norm": 4.3973517417907715,
      "learning_rate": 5.196402101576182e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 110450376,
      "step": 6864
    },
    {
      "epoch": 0.4808812069312739,
      "grad_norm": 4.029878616333008,
      "learning_rate": 5.195702276707531e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 110466760,
      "step": 6865
    },
    {
      "epoch": 0.48095125517700316,
      "grad_norm": 3.5051989555358887,
      "learning_rate": 5.195002451838879e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 110483144,
      "step": 6866
    },
    {
      "epoch": 0.4810213034227324,
      "grad_norm": 3.8468475341796875,
      "learning_rate": 5.1943026269702275e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 110499528,
      "step": 6867
    },
    {
      "epoch": 0.4810913516684617,
      "grad_norm": 3.4679362773895264,
      "learning_rate": 5.193602802101577e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 110515448,
      "step": 6868
    },
    {
      "epoch": 0.4811613999141909,
      "grad_norm": 3.540043830871582,
      "learning_rate": 5.192902977232925e-05,
      "loss": 1.0163,
      "num_input_tokens_seen": 110531832,
      "step": 6869
    },
    {
      "epoch": 0.48123144815992014,
      "grad_norm": 4.2961835861206055,
      "learning_rate": 5.192203152364272e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 110548216,
      "step": 6870
    },
    {
      "epoch": 0.4813014964056494,
      "grad_norm": 4.718245029449463,
      "learning_rate": 5.191503327495623e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 110564600,
      "step": 6871
    },
    {
      "epoch": 0.48137154465137866,
      "grad_norm": 4.846748352050781,
      "learning_rate": 5.190803502626971e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 110579952,
      "step": 6872
    },
    {
      "epoch": 0.48144159289710786,
      "grad_norm": 3.5760273933410645,
      "learning_rate": 5.1901036777583186e-05,
      "loss": 1.0028,
      "num_input_tokens_seen": 110595984,
      "step": 6873
    },
    {
      "epoch": 0.4815116411428371,
      "grad_norm": 6.386372089385986,
      "learning_rate": 5.189403852889667e-05,
      "loss": 1.1695,
      "num_input_tokens_seen": 110612368,
      "step": 6874
    },
    {
      "epoch": 0.4815816893885664,
      "grad_norm": 5.007279872894287,
      "learning_rate": 5.188704028021018e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 110628752,
      "step": 6875
    },
    {
      "epoch": 0.48165173763429564,
      "grad_norm": 4.01614236831665,
      "learning_rate": 5.1880042031523645e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 110645136,
      "step": 6876
    },
    {
      "epoch": 0.48172178588002484,
      "grad_norm": 4.7416486740112305,
      "learning_rate": 5.187304378283713e-05,
      "loss": 1.3402,
      "num_input_tokens_seen": 110661400,
      "step": 6877
    },
    {
      "epoch": 0.4817918341257541,
      "grad_norm": 4.886537551879883,
      "learning_rate": 5.186604553415062e-05,
      "loss": 0.8621,
      "num_input_tokens_seen": 110677784,
      "step": 6878
    },
    {
      "epoch": 0.48186188237148336,
      "grad_norm": 4.033387660980225,
      "learning_rate": 5.1859047285464104e-05,
      "loss": 1.3515,
      "num_input_tokens_seen": 110694168,
      "step": 6879
    },
    {
      "epoch": 0.4819319306172126,
      "grad_norm": 3.7201569080352783,
      "learning_rate": 5.1852049036777586e-05,
      "loss": 1.163,
      "num_input_tokens_seen": 110710552,
      "step": 6880
    },
    {
      "epoch": 0.4820019788629418,
      "grad_norm": 3.73651123046875,
      "learning_rate": 5.1845050788091075e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 110726440,
      "step": 6881
    },
    {
      "epoch": 0.4820720271086711,
      "grad_norm": 4.395266532897949,
      "learning_rate": 5.1838052539404556e-05,
      "loss": 0.9636,
      "num_input_tokens_seen": 110742200,
      "step": 6882
    },
    {
      "epoch": 0.48214207535440035,
      "grad_norm": 3.70263409614563,
      "learning_rate": 5.183105429071804e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 110758584,
      "step": 6883
    },
    {
      "epoch": 0.4822121236001296,
      "grad_norm": 4.863175868988037,
      "learning_rate": 5.1824056042031534e-05,
      "loss": 1.1663,
      "num_input_tokens_seen": 110774040,
      "step": 6884
    },
    {
      "epoch": 0.4822821718458588,
      "grad_norm": 3.668220043182373,
      "learning_rate": 5.181705779334502e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 110790352,
      "step": 6885
    },
    {
      "epoch": 0.48235222009158807,
      "grad_norm": 4.210755825042725,
      "learning_rate": 5.18100595446585e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 110805912,
      "step": 6886
    },
    {
      "epoch": 0.48242226833731733,
      "grad_norm": 3.62275767326355,
      "learning_rate": 5.180306129597198e-05,
      "loss": 1.0383,
      "num_input_tokens_seen": 110822296,
      "step": 6887
    },
    {
      "epoch": 0.4824923165830466,
      "grad_norm": 3.498563051223755,
      "learning_rate": 5.179606304728547e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 110838680,
      "step": 6888
    },
    {
      "epoch": 0.4825623648287758,
      "grad_norm": 6.4097161293029785,
      "learning_rate": 5.178906479859895e-05,
      "loss": 0.9482,
      "num_input_tokens_seen": 110855064,
      "step": 6889
    },
    {
      "epoch": 0.48263241307450505,
      "grad_norm": 4.8159565925598145,
      "learning_rate": 5.178206654991243e-05,
      "loss": 1.2248,
      "num_input_tokens_seen": 110871328,
      "step": 6890
    },
    {
      "epoch": 0.4827024613202343,
      "grad_norm": 3.976828098297119,
      "learning_rate": 5.177506830122593e-05,
      "loss": 1.1521,
      "num_input_tokens_seen": 110886600,
      "step": 6891
    },
    {
      "epoch": 0.48277250956596357,
      "grad_norm": 3.6857738494873047,
      "learning_rate": 5.176807005253942e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 110902984,
      "step": 6892
    },
    {
      "epoch": 0.48284255781169283,
      "grad_norm": 4.129028797149658,
      "learning_rate": 5.176107180385289e-05,
      "loss": 1.1404,
      "num_input_tokens_seen": 110918808,
      "step": 6893
    },
    {
      "epoch": 0.48291260605742203,
      "grad_norm": 4.203270435333252,
      "learning_rate": 5.175407355516637e-05,
      "loss": 1.1844,
      "num_input_tokens_seen": 110935192,
      "step": 6894
    },
    {
      "epoch": 0.4829826543031513,
      "grad_norm": 3.7045552730560303,
      "learning_rate": 5.1747075306479874e-05,
      "loss": 0.9193,
      "num_input_tokens_seen": 110951168,
      "step": 6895
    },
    {
      "epoch": 0.48305270254888055,
      "grad_norm": 4.2172112464904785,
      "learning_rate": 5.174007705779334e-05,
      "loss": 0.8905,
      "num_input_tokens_seen": 110967552,
      "step": 6896
    },
    {
      "epoch": 0.4831227507946098,
      "grad_norm": 3.395329236984253,
      "learning_rate": 5.1733078809106824e-05,
      "loss": 0.9696,
      "num_input_tokens_seen": 110983736,
      "step": 6897
    },
    {
      "epoch": 0.483192799040339,
      "grad_norm": 6.649857044219971,
      "learning_rate": 5.172608056042032e-05,
      "loss": 1.2299,
      "num_input_tokens_seen": 111000120,
      "step": 6898
    },
    {
      "epoch": 0.4832628472860683,
      "grad_norm": 5.114965438842773,
      "learning_rate": 5.1719082311733815e-05,
      "loss": 1.196,
      "num_input_tokens_seen": 111016504,
      "step": 6899
    },
    {
      "epoch": 0.48333289553179754,
      "grad_norm": 4.1728410720825195,
      "learning_rate": 5.171208406304728e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 111032232,
      "step": 6900
    },
    {
      "epoch": 0.4834029437775268,
      "grad_norm": 3.674546241760254,
      "learning_rate": 5.170508581436078e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 111047576,
      "step": 6901
    },
    {
      "epoch": 0.483472992023256,
      "grad_norm": 3.4895896911621094,
      "learning_rate": 5.169808756567427e-05,
      "loss": 0.9618,
      "num_input_tokens_seen": 111063792,
      "step": 6902
    },
    {
      "epoch": 0.48354304026898526,
      "grad_norm": 8.447297096252441,
      "learning_rate": 5.169108931698774e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 111079136,
      "step": 6903
    },
    {
      "epoch": 0.4836130885147145,
      "grad_norm": 4.854581356048584,
      "learning_rate": 5.1684091068301224e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 111093808,
      "step": 6904
    },
    {
      "epoch": 0.4836831367604438,
      "grad_norm": 3.4015259742736816,
      "learning_rate": 5.167709281961471e-05,
      "loss": 0.9395,
      "num_input_tokens_seen": 111110192,
      "step": 6905
    },
    {
      "epoch": 0.483753185006173,
      "grad_norm": 3.979801654815674,
      "learning_rate": 5.1670094570928195e-05,
      "loss": 1.181,
      "num_input_tokens_seen": 111126576,
      "step": 6906
    },
    {
      "epoch": 0.48382323325190224,
      "grad_norm": 3.655245542526245,
      "learning_rate": 5.166309632224169e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 111142960,
      "step": 6907
    },
    {
      "epoch": 0.4838932814976315,
      "grad_norm": 3.820819616317749,
      "learning_rate": 5.165609807355517e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 111159344,
      "step": 6908
    },
    {
      "epoch": 0.48396332974336076,
      "grad_norm": 3.6869490146636963,
      "learning_rate": 5.164909982486866e-05,
      "loss": 0.7909,
      "num_input_tokens_seen": 111175608,
      "step": 6909
    },
    {
      "epoch": 0.48403337798908996,
      "grad_norm": 3.644277334213257,
      "learning_rate": 5.1642101576182135e-05,
      "loss": 1.0442,
      "num_input_tokens_seen": 111191992,
      "step": 6910
    },
    {
      "epoch": 0.4841034262348192,
      "grad_norm": 3.794215202331543,
      "learning_rate": 5.1635103327495624e-05,
      "loss": 1.1105,
      "num_input_tokens_seen": 111207248,
      "step": 6911
    },
    {
      "epoch": 0.4841734744805485,
      "grad_norm": 4.5081987380981445,
      "learning_rate": 5.162810507880912e-05,
      "loss": 1.3952,
      "num_input_tokens_seen": 111223632,
      "step": 6912
    },
    {
      "epoch": 0.48424352272627774,
      "grad_norm": 3.632924795150757,
      "learning_rate": 5.162110683012259e-05,
      "loss": 1.0862,
      "num_input_tokens_seen": 111240016,
      "step": 6913
    },
    {
      "epoch": 0.48431357097200695,
      "grad_norm": 3.522996425628662,
      "learning_rate": 5.161410858143607e-05,
      "loss": 0.9521,
      "num_input_tokens_seen": 111255840,
      "step": 6914
    },
    {
      "epoch": 0.4843836192177362,
      "grad_norm": 4.495186805725098,
      "learning_rate": 5.1607110332749565e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 111272224,
      "step": 6915
    },
    {
      "epoch": 0.48445366746346546,
      "grad_norm": 3.6315512657165527,
      "learning_rate": 5.160011208406306e-05,
      "loss": 1.0991,
      "num_input_tokens_seen": 111287920,
      "step": 6916
    },
    {
      "epoch": 0.4845237157091947,
      "grad_norm": 3.4649548530578613,
      "learning_rate": 5.159311383537654e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 111304304,
      "step": 6917
    },
    {
      "epoch": 0.48459376395492393,
      "grad_norm": 4.057675838470459,
      "learning_rate": 5.158611558669002e-05,
      "loss": 1.0403,
      "num_input_tokens_seen": 111320688,
      "step": 6918
    },
    {
      "epoch": 0.4846638122006532,
      "grad_norm": 4.989962100982666,
      "learning_rate": 5.157911733800351e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 111337072,
      "step": 6919
    },
    {
      "epoch": 0.48473386044638245,
      "grad_norm": 4.090515613555908,
      "learning_rate": 5.157211908931698e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 111353456,
      "step": 6920
    },
    {
      "epoch": 0.4848039086921117,
      "grad_norm": 4.017073154449463,
      "learning_rate": 5.156512084063046e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 111369840,
      "step": 6921
    },
    {
      "epoch": 0.4848739569378409,
      "grad_norm": 4.227852821350098,
      "learning_rate": 5.155812259194397e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 111386096,
      "step": 6922
    },
    {
      "epoch": 0.48494400518357017,
      "grad_norm": 5.356720447540283,
      "learning_rate": 5.155112434325745e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 111402192,
      "step": 6923
    },
    {
      "epoch": 0.48501405342929943,
      "grad_norm": 3.714996814727783,
      "learning_rate": 5.1544126094570935e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 111418120,
      "step": 6924
    },
    {
      "epoch": 0.4850841016750287,
      "grad_norm": 3.814669609069824,
      "learning_rate": 5.153712784588442e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 111434408,
      "step": 6925
    },
    {
      "epoch": 0.48515414992075795,
      "grad_norm": 4.38773250579834,
      "learning_rate": 5.1530129597197905e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 111450384,
      "step": 6926
    },
    {
      "epoch": 0.48522419816648715,
      "grad_norm": 5.492570877075195,
      "learning_rate": 5.152313134851139e-05,
      "loss": 1.0629,
      "num_input_tokens_seen": 111466768,
      "step": 6927
    },
    {
      "epoch": 0.4852942464122164,
      "grad_norm": 4.867751598358154,
      "learning_rate": 5.151613309982487e-05,
      "loss": 1.0787,
      "num_input_tokens_seen": 111481680,
      "step": 6928
    },
    {
      "epoch": 0.48536429465794567,
      "grad_norm": 3.6009931564331055,
      "learning_rate": 5.1509134851138364e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 111497784,
      "step": 6929
    },
    {
      "epoch": 0.48543434290367493,
      "grad_norm": 3.451188564300537,
      "learning_rate": 5.1502136602451846e-05,
      "loss": 0.9131,
      "num_input_tokens_seen": 111513856,
      "step": 6930
    },
    {
      "epoch": 0.48550439114940414,
      "grad_norm": 4.886107444763184,
      "learning_rate": 5.149513835376533e-05,
      "loss": 0.9234,
      "num_input_tokens_seen": 111530240,
      "step": 6931
    },
    {
      "epoch": 0.4855744393951334,
      "grad_norm": 4.033775806427002,
      "learning_rate": 5.148814010507881e-05,
      "loss": 1.0094,
      "num_input_tokens_seen": 111546160,
      "step": 6932
    },
    {
      "epoch": 0.48564448764086265,
      "grad_norm": 4.718981742858887,
      "learning_rate": 5.14811418563923e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 111562432,
      "step": 6933
    },
    {
      "epoch": 0.4857145358865919,
      "grad_norm": 3.7174808979034424,
      "learning_rate": 5.147414360770578e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 111578816,
      "step": 6934
    },
    {
      "epoch": 0.4857845841323211,
      "grad_norm": 4.0880208015441895,
      "learning_rate": 5.146714535901926e-05,
      "loss": 1.0742,
      "num_input_tokens_seen": 111593928,
      "step": 6935
    },
    {
      "epoch": 0.4858546323780504,
      "grad_norm": 3.3873400688171387,
      "learning_rate": 5.146014711033276e-05,
      "loss": 0.9752,
      "num_input_tokens_seen": 111610312,
      "step": 6936
    },
    {
      "epoch": 0.48592468062377964,
      "grad_norm": 3.6071503162384033,
      "learning_rate": 5.145314886164624e-05,
      "loss": 0.9917,
      "num_input_tokens_seen": 111626696,
      "step": 6937
    },
    {
      "epoch": 0.4859947288695089,
      "grad_norm": 3.502610445022583,
      "learning_rate": 5.1446150612959735e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 111643080,
      "step": 6938
    },
    {
      "epoch": 0.4860647771152381,
      "grad_norm": 3.5743067264556885,
      "learning_rate": 5.1439152364273216e-05,
      "loss": 1.1493,
      "num_input_tokens_seen": 111659048,
      "step": 6939
    },
    {
      "epoch": 0.48613482536096736,
      "grad_norm": 3.9423654079437256,
      "learning_rate": 5.14321541155867e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 111675432,
      "step": 6940
    },
    {
      "epoch": 0.4862048736066966,
      "grad_norm": 4.670028209686279,
      "learning_rate": 5.142515586690017e-05,
      "loss": 0.9023,
      "num_input_tokens_seen": 111691816,
      "step": 6941
    },
    {
      "epoch": 0.4862749218524259,
      "grad_norm": 3.8914809226989746,
      "learning_rate": 5.1418157618213655e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 111708200,
      "step": 6942
    },
    {
      "epoch": 0.4863449700981551,
      "grad_norm": 3.864323139190674,
      "learning_rate": 5.141115936952715e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 111724488,
      "step": 6943
    },
    {
      "epoch": 0.48641501834388434,
      "grad_norm": 3.700681447982788,
      "learning_rate": 5.140416112084063e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 111740368,
      "step": 6944
    },
    {
      "epoch": 0.4864850665896136,
      "grad_norm": 3.7225606441497803,
      "learning_rate": 5.139716287215413e-05,
      "loss": 0.971,
      "num_input_tokens_seen": 111755936,
      "step": 6945
    },
    {
      "epoch": 0.48655511483534286,
      "grad_norm": 4.638529300689697,
      "learning_rate": 5.139016462346761e-05,
      "loss": 0.9367,
      "num_input_tokens_seen": 111772152,
      "step": 6946
    },
    {
      "epoch": 0.48662516308107207,
      "grad_norm": 5.287013053894043,
      "learning_rate": 5.138316637478109e-05,
      "loss": 0.9463,
      "num_input_tokens_seen": 111787144,
      "step": 6947
    },
    {
      "epoch": 0.4866952113268013,
      "grad_norm": 3.991861343383789,
      "learning_rate": 5.137616812609457e-05,
      "loss": 0.8265,
      "num_input_tokens_seen": 111803528,
      "step": 6948
    },
    {
      "epoch": 0.4867652595725306,
      "grad_norm": 4.166889190673828,
      "learning_rate": 5.136916987740806e-05,
      "loss": 1.1667,
      "num_input_tokens_seen": 111819376,
      "step": 6949
    },
    {
      "epoch": 0.48683530781825984,
      "grad_norm": 4.159299373626709,
      "learning_rate": 5.1362171628721543e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 111835760,
      "step": 6950
    },
    {
      "epoch": 0.48690535606398905,
      "grad_norm": 5.612180709838867,
      "learning_rate": 5.1355173380035025e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 111851744,
      "step": 6951
    },
    {
      "epoch": 0.4869754043097183,
      "grad_norm": 5.82523775100708,
      "learning_rate": 5.134817513134852e-05,
      "loss": 0.9702,
      "num_input_tokens_seen": 111868128,
      "step": 6952
    },
    {
      "epoch": 0.48704545255544757,
      "grad_norm": 3.5110416412353516,
      "learning_rate": 5.1341176882662e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 111884512,
      "step": 6953
    },
    {
      "epoch": 0.4871155008011768,
      "grad_norm": 4.108850479125977,
      "learning_rate": 5.1334178633975484e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 111899544,
      "step": 6954
    },
    {
      "epoch": 0.48718554904690603,
      "grad_norm": 3.9000258445739746,
      "learning_rate": 5.132718038528897e-05,
      "loss": 0.9456,
      "num_input_tokens_seen": 111915928,
      "step": 6955
    },
    {
      "epoch": 0.4872555972926353,
      "grad_norm": 4.503340244293213,
      "learning_rate": 5.1320182136602455e-05,
      "loss": 1.004,
      "num_input_tokens_seen": 111932288,
      "step": 6956
    },
    {
      "epoch": 0.48732564553836455,
      "grad_norm": 4.052606582641602,
      "learning_rate": 5.1313183887915936e-05,
      "loss": 1.1662,
      "num_input_tokens_seen": 111948672,
      "step": 6957
    },
    {
      "epoch": 0.4873956937840938,
      "grad_norm": 3.4959487915039062,
      "learning_rate": 5.130618563922942e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 111964272,
      "step": 6958
    },
    {
      "epoch": 0.48746574202982307,
      "grad_norm": 4.654433250427246,
      "learning_rate": 5.12991873905429e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 111980656,
      "step": 6959
    },
    {
      "epoch": 0.48753579027555227,
      "grad_norm": 4.125091552734375,
      "learning_rate": 5.1292189141856395e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 111996584,
      "step": 6960
    },
    {
      "epoch": 0.48760583852128153,
      "grad_norm": 4.574272155761719,
      "learning_rate": 5.128519089316988e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 112011528,
      "step": 6961
    },
    {
      "epoch": 0.4876758867670108,
      "grad_norm": 4.110400676727295,
      "learning_rate": 5.127819264448337e-05,
      "loss": 0.9669,
      "num_input_tokens_seen": 112026256,
      "step": 6962
    },
    {
      "epoch": 0.48774593501274005,
      "grad_norm": 3.4572913646698,
      "learning_rate": 5.127119439579685e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 112042288,
      "step": 6963
    },
    {
      "epoch": 0.48781598325846925,
      "grad_norm": 4.498427391052246,
      "learning_rate": 5.126419614711033e-05,
      "loss": 1.1088,
      "num_input_tokens_seen": 112058672,
      "step": 6964
    },
    {
      "epoch": 0.4878860315041985,
      "grad_norm": 5.4692301750183105,
      "learning_rate": 5.1257197898423825e-05,
      "loss": 0.9582,
      "num_input_tokens_seen": 112073536,
      "step": 6965
    },
    {
      "epoch": 0.4879560797499278,
      "grad_norm": 3.8990654945373535,
      "learning_rate": 5.125019964973732e-05,
      "loss": 1.2729,
      "num_input_tokens_seen": 112089344,
      "step": 6966
    },
    {
      "epoch": 0.48802612799565703,
      "grad_norm": 3.5601627826690674,
      "learning_rate": 5.12432014010508e-05,
      "loss": 1.0519,
      "num_input_tokens_seen": 112105296,
      "step": 6967
    },
    {
      "epoch": 0.48809617624138624,
      "grad_norm": 3.91282057762146,
      "learning_rate": 5.123620315236427e-05,
      "loss": 1.2046,
      "num_input_tokens_seen": 112120600,
      "step": 6968
    },
    {
      "epoch": 0.4881662244871155,
      "grad_norm": 5.9246602058410645,
      "learning_rate": 5.1229204903677766e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 112136472,
      "step": 6969
    },
    {
      "epoch": 0.48823627273284476,
      "grad_norm": 8.849782943725586,
      "learning_rate": 5.122220665499125e-05,
      "loss": 1.0671,
      "num_input_tokens_seen": 112152792,
      "step": 6970
    },
    {
      "epoch": 0.488306320978574,
      "grad_norm": 4.184106349945068,
      "learning_rate": 5.121520840630473e-05,
      "loss": 1.2696,
      "num_input_tokens_seen": 112169176,
      "step": 6971
    },
    {
      "epoch": 0.4883763692243032,
      "grad_norm": 4.250857830047607,
      "learning_rate": 5.120821015761822e-05,
      "loss": 0.9724,
      "num_input_tokens_seen": 112184784,
      "step": 6972
    },
    {
      "epoch": 0.4884464174700325,
      "grad_norm": 4.522305011749268,
      "learning_rate": 5.12012119089317e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 112200960,
      "step": 6973
    },
    {
      "epoch": 0.48851646571576174,
      "grad_norm": 3.5135490894317627,
      "learning_rate": 5.1194213660245195e-05,
      "loss": 1.0713,
      "num_input_tokens_seen": 112217080,
      "step": 6974
    },
    {
      "epoch": 0.488586513961491,
      "grad_norm": 5.541810989379883,
      "learning_rate": 5.118721541155866e-05,
      "loss": 1.1885,
      "num_input_tokens_seen": 112233464,
      "step": 6975
    },
    {
      "epoch": 0.4886565622072202,
      "grad_norm": 3.7535064220428467,
      "learning_rate": 5.1180217162872165e-05,
      "loss": 1.0558,
      "num_input_tokens_seen": 112249848,
      "step": 6976
    },
    {
      "epoch": 0.48872661045294946,
      "grad_norm": 4.454082012176514,
      "learning_rate": 5.117321891418565e-05,
      "loss": 1.0892,
      "num_input_tokens_seen": 112265560,
      "step": 6977
    },
    {
      "epoch": 0.4887966586986787,
      "grad_norm": 3.770138740539551,
      "learning_rate": 5.116622066549912e-05,
      "loss": 1.082,
      "num_input_tokens_seen": 112281944,
      "step": 6978
    },
    {
      "epoch": 0.488866706944408,
      "grad_norm": 5.923669815063477,
      "learning_rate": 5.115922241681261e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 112298264,
      "step": 6979
    },
    {
      "epoch": 0.4889367551901372,
      "grad_norm": 3.9768123626708984,
      "learning_rate": 5.115222416812609e-05,
      "loss": 1.1816,
      "num_input_tokens_seen": 112314608,
      "step": 6980
    },
    {
      "epoch": 0.48900680343586644,
      "grad_norm": 5.525039196014404,
      "learning_rate": 5.1145225919439575e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 112330400,
      "step": 6981
    },
    {
      "epoch": 0.4890768516815957,
      "grad_norm": 3.8725640773773193,
      "learning_rate": 5.113822767075307e-05,
      "loss": 0.9666,
      "num_input_tokens_seen": 112345384,
      "step": 6982
    },
    {
      "epoch": 0.48914689992732496,
      "grad_norm": 4.746465682983398,
      "learning_rate": 5.1131229422066565e-05,
      "loss": 1.1162,
      "num_input_tokens_seen": 112361768,
      "step": 6983
    },
    {
      "epoch": 0.48921694817305417,
      "grad_norm": 3.774049997329712,
      "learning_rate": 5.112423117338005e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 112377432,
      "step": 6984
    },
    {
      "epoch": 0.4892869964187834,
      "grad_norm": 3.686307191848755,
      "learning_rate": 5.1117232924693515e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 112393672,
      "step": 6985
    },
    {
      "epoch": 0.4893570446645127,
      "grad_norm": 4.177459239959717,
      "learning_rate": 5.1110234676007004e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 112409600,
      "step": 6986
    },
    {
      "epoch": 0.48942709291024195,
      "grad_norm": 3.8517558574676514,
      "learning_rate": 5.1103236427320486e-05,
      "loss": 1.0947,
      "num_input_tokens_seen": 112425880,
      "step": 6987
    },
    {
      "epoch": 0.48949714115597115,
      "grad_norm": 3.3155159950256348,
      "learning_rate": 5.109623817863397e-05,
      "loss": 0.9201,
      "num_input_tokens_seen": 112442264,
      "step": 6988
    },
    {
      "epoch": 0.4895671894017004,
      "grad_norm": 4.027132987976074,
      "learning_rate": 5.108923992994746e-05,
      "loss": 1.0882,
      "num_input_tokens_seen": 112458504,
      "step": 6989
    },
    {
      "epoch": 0.48963723764742967,
      "grad_norm": 3.622421979904175,
      "learning_rate": 5.108224168126096e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 112474888,
      "step": 6990
    },
    {
      "epoch": 0.48970728589315893,
      "grad_norm": 4.16541051864624,
      "learning_rate": 5.107524343257444e-05,
      "loss": 0.967,
      "num_input_tokens_seen": 112491272,
      "step": 6991
    },
    {
      "epoch": 0.48977733413888813,
      "grad_norm": 4.473822593688965,
      "learning_rate": 5.106824518388792e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 112506632,
      "step": 6992
    },
    {
      "epoch": 0.4898473823846174,
      "grad_norm": 5.10452127456665,
      "learning_rate": 5.106124693520141e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 112521696,
      "step": 6993
    },
    {
      "epoch": 0.48991743063034665,
      "grad_norm": 4.185652732849121,
      "learning_rate": 5.105424868651489e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 112537432,
      "step": 6994
    },
    {
      "epoch": 0.4899874788760759,
      "grad_norm": 4.864262580871582,
      "learning_rate": 5.104725043782837e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 112553816,
      "step": 6995
    },
    {
      "epoch": 0.49005752712180517,
      "grad_norm": 3.859199047088623,
      "learning_rate": 5.1040252189141856e-05,
      "loss": 1.115,
      "num_input_tokens_seen": 112570200,
      "step": 6996
    },
    {
      "epoch": 0.4901275753675344,
      "grad_norm": 3.49395751953125,
      "learning_rate": 5.103325394045535e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 112586584,
      "step": 6997
    },
    {
      "epoch": 0.49019762361326363,
      "grad_norm": 4.164735317230225,
      "learning_rate": 5.102625569176883e-05,
      "loss": 0.8565,
      "num_input_tokens_seen": 112602568,
      "step": 6998
    },
    {
      "epoch": 0.4902676718589929,
      "grad_norm": 6.273041725158691,
      "learning_rate": 5.1019257443082315e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 112618952,
      "step": 6999
    },
    {
      "epoch": 0.49033772010472215,
      "grad_norm": 3.8460848331451416,
      "learning_rate": 5.1012259194395804e-05,
      "loss": 0.9854,
      "num_input_tokens_seen": 112635336,
      "step": 7000
    },
    {
      "epoch": 0.49033772010472215,
      "eval_loss": 1.1226829290390015,
      "eval_runtime": 0.157,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 6.371,
      "num_input_tokens_seen": 112635336,
      "step": 7000
    },
    {
      "epoch": 0.49040776835045136,
      "grad_norm": 4.498118877410889,
      "learning_rate": 5.1005260945709285e-05,
      "loss": 1.2191,
      "num_input_tokens_seen": 112651248,
      "step": 7001
    },
    {
      "epoch": 0.4904778165961806,
      "grad_norm": 3.9830660820007324,
      "learning_rate": 5.099826269702276e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 112667632,
      "step": 7002
    },
    {
      "epoch": 0.4905478648419099,
      "grad_norm": 3.439422130584717,
      "learning_rate": 5.099126444833626e-05,
      "loss": 0.9423,
      "num_input_tokens_seen": 112684016,
      "step": 7003
    },
    {
      "epoch": 0.49061791308763913,
      "grad_norm": 3.6636171340942383,
      "learning_rate": 5.0984266199649744e-05,
      "loss": 1.0124,
      "num_input_tokens_seen": 112700360,
      "step": 7004
    },
    {
      "epoch": 0.49068796133336834,
      "grad_norm": 4.3784589767456055,
      "learning_rate": 5.097726795096321e-05,
      "loss": 1.1782,
      "num_input_tokens_seen": 112715624,
      "step": 7005
    },
    {
      "epoch": 0.4907580095790976,
      "grad_norm": 3.716031789779663,
      "learning_rate": 5.097026970227671e-05,
      "loss": 1.0312,
      "num_input_tokens_seen": 112732008,
      "step": 7006
    },
    {
      "epoch": 0.49082805782482686,
      "grad_norm": 4.289496898651123,
      "learning_rate": 5.09632714535902e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 112748392,
      "step": 7007
    },
    {
      "epoch": 0.4908981060705561,
      "grad_norm": 4.012343883514404,
      "learning_rate": 5.095627320490368e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 112764432,
      "step": 7008
    },
    {
      "epoch": 0.4909681543162853,
      "grad_norm": 3.7599732875823975,
      "learning_rate": 5.094927495621716e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 112780544,
      "step": 7009
    },
    {
      "epoch": 0.4910382025620146,
      "grad_norm": 3.398778200149536,
      "learning_rate": 5.0942276707530656e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 112796928,
      "step": 7010
    },
    {
      "epoch": 0.49110825080774384,
      "grad_norm": 7.299741268157959,
      "learning_rate": 5.093527845884414e-05,
      "loss": 1.0963,
      "num_input_tokens_seen": 112812576,
      "step": 7011
    },
    {
      "epoch": 0.4911782990534731,
      "grad_norm": 4.2506866455078125,
      "learning_rate": 5.0928280210157606e-05,
      "loss": 1.11,
      "num_input_tokens_seen": 112828184,
      "step": 7012
    },
    {
      "epoch": 0.4912483472992023,
      "grad_norm": 4.033505916595459,
      "learning_rate": 5.09212819614711e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 112844000,
      "step": 7013
    },
    {
      "epoch": 0.49131839554493156,
      "grad_norm": 3.9474592208862305,
      "learning_rate": 5.0914283712784596e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 112860384,
      "step": 7014
    },
    {
      "epoch": 0.4913884437906608,
      "grad_norm": 5.549149036407471,
      "learning_rate": 5.090728546409808e-05,
      "loss": 0.8816,
      "num_input_tokens_seen": 112875512,
      "step": 7015
    },
    {
      "epoch": 0.4914584920363901,
      "grad_norm": 5.086400985717773,
      "learning_rate": 5.090028721541156e-05,
      "loss": 1.2013,
      "num_input_tokens_seen": 112890840,
      "step": 7016
    },
    {
      "epoch": 0.4915285402821193,
      "grad_norm": 3.8358511924743652,
      "learning_rate": 5.089328896672505e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 112907048,
      "step": 7017
    },
    {
      "epoch": 0.49159858852784855,
      "grad_norm": 3.902388095855713,
      "learning_rate": 5.088629071803853e-05,
      "loss": 0.9513,
      "num_input_tokens_seen": 112923152,
      "step": 7018
    },
    {
      "epoch": 0.4916686367735778,
      "grad_norm": 4.3525166511535645,
      "learning_rate": 5.0879292469352026e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 112939536,
      "step": 7019
    },
    {
      "epoch": 0.49173868501930706,
      "grad_norm": 4.519508361816406,
      "learning_rate": 5.087229422066551e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 112955920,
      "step": 7020
    },
    {
      "epoch": 0.49180873326503627,
      "grad_norm": 4.366591930389404,
      "learning_rate": 5.086529597197899e-05,
      "loss": 1.1246,
      "num_input_tokens_seen": 112971824,
      "step": 7021
    },
    {
      "epoch": 0.49187878151076553,
      "grad_norm": 3.9145777225494385,
      "learning_rate": 5.085829772329247e-05,
      "loss": 1.085,
      "num_input_tokens_seen": 112988208,
      "step": 7022
    },
    {
      "epoch": 0.4919488297564948,
      "grad_norm": 3.9565582275390625,
      "learning_rate": 5.085129947460595e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 113004592,
      "step": 7023
    },
    {
      "epoch": 0.49201887800222405,
      "grad_norm": 4.051690101623535,
      "learning_rate": 5.084430122591944e-05,
      "loss": 1.0569,
      "num_input_tokens_seen": 113020552,
      "step": 7024
    },
    {
      "epoch": 0.49208892624795325,
      "grad_norm": 4.020756244659424,
      "learning_rate": 5.0837302977232923e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 113036840,
      "step": 7025
    },
    {
      "epoch": 0.4921589744936825,
      "grad_norm": 4.177811622619629,
      "learning_rate": 5.0830304728546405e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 113051816,
      "step": 7026
    },
    {
      "epoch": 0.49222902273941177,
      "grad_norm": 3.997945785522461,
      "learning_rate": 5.08233064798599e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 113067968,
      "step": 7027
    },
    {
      "epoch": 0.49229907098514103,
      "grad_norm": 4.968790531158447,
      "learning_rate": 5.081630823117338e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 113084352,
      "step": 7028
    },
    {
      "epoch": 0.4923691192308703,
      "grad_norm": 3.9024367332458496,
      "learning_rate": 5.0809309982486864e-05,
      "loss": 0.9673,
      "num_input_tokens_seen": 113100672,
      "step": 7029
    },
    {
      "epoch": 0.4924391674765995,
      "grad_norm": 4.58246374130249,
      "learning_rate": 5.080231173380036e-05,
      "loss": 1.0995,
      "num_input_tokens_seen": 113115856,
      "step": 7030
    },
    {
      "epoch": 0.49250921572232875,
      "grad_norm": 4.944141387939453,
      "learning_rate": 5.0795313485113835e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 113132240,
      "step": 7031
    },
    {
      "epoch": 0.492579263968058,
      "grad_norm": 5.382607460021973,
      "learning_rate": 5.0788315236427316e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 113148624,
      "step": 7032
    },
    {
      "epoch": 0.49264931221378727,
      "grad_norm": 4.664593696594238,
      "learning_rate": 5.07813169877408e-05,
      "loss": 1.2741,
      "num_input_tokens_seen": 113164608,
      "step": 7033
    },
    {
      "epoch": 0.4927193604595165,
      "grad_norm": 3.8908638954162598,
      "learning_rate": 5.077431873905431e-05,
      "loss": 1.0828,
      "num_input_tokens_seen": 113180712,
      "step": 7034
    },
    {
      "epoch": 0.49278940870524574,
      "grad_norm": 3.93803334236145,
      "learning_rate": 5.0767320490367775e-05,
      "loss": 1.0555,
      "num_input_tokens_seen": 113196328,
      "step": 7035
    },
    {
      "epoch": 0.492859456950975,
      "grad_norm": 5.349659442901611,
      "learning_rate": 5.076032224168127e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 113212712,
      "step": 7036
    },
    {
      "epoch": 0.49292950519670425,
      "grad_norm": 5.471059322357178,
      "learning_rate": 5.075332399299475e-05,
      "loss": 1.0165,
      "num_input_tokens_seen": 113229096,
      "step": 7037
    },
    {
      "epoch": 0.49299955344243346,
      "grad_norm": 5.430948734283447,
      "learning_rate": 5.0746325744308234e-05,
      "loss": 1.1832,
      "num_input_tokens_seen": 113244752,
      "step": 7038
    },
    {
      "epoch": 0.4930696016881627,
      "grad_norm": 3.4861812591552734,
      "learning_rate": 5.0739327495621716e-05,
      "loss": 0.9185,
      "num_input_tokens_seen": 113261136,
      "step": 7039
    },
    {
      "epoch": 0.493139649933892,
      "grad_norm": 3.9184775352478027,
      "learning_rate": 5.0732329246935205e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 113277520,
      "step": 7040
    },
    {
      "epoch": 0.49320969817962124,
      "grad_norm": 3.6723365783691406,
      "learning_rate": 5.072533099824869e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 113293208,
      "step": 7041
    },
    {
      "epoch": 0.49327974642535044,
      "grad_norm": 4.273809432983398,
      "learning_rate": 5.071833274956217e-05,
      "loss": 1.2621,
      "num_input_tokens_seen": 113309592,
      "step": 7042
    },
    {
      "epoch": 0.4933497946710797,
      "grad_norm": 3.448326349258423,
      "learning_rate": 5.0711334500875664e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 113325976,
      "step": 7043
    },
    {
      "epoch": 0.49341984291680896,
      "grad_norm": 4.427138805389404,
      "learning_rate": 5.070433625218915e-05,
      "loss": 1.1113,
      "num_input_tokens_seen": 113341896,
      "step": 7044
    },
    {
      "epoch": 0.4934898911625382,
      "grad_norm": 3.837282180786133,
      "learning_rate": 5.069733800350263e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 113358280,
      "step": 7045
    },
    {
      "epoch": 0.4935599394082674,
      "grad_norm": 4.842933654785156,
      "learning_rate": 5.0690339754816116e-05,
      "loss": 1.3572,
      "num_input_tokens_seen": 113374664,
      "step": 7046
    },
    {
      "epoch": 0.4936299876539967,
      "grad_norm": 5.275210857391357,
      "learning_rate": 5.06833415061296e-05,
      "loss": 1.0208,
      "num_input_tokens_seen": 113391048,
      "step": 7047
    },
    {
      "epoch": 0.49370003589972594,
      "grad_norm": 4.120177268981934,
      "learning_rate": 5.067634325744308e-05,
      "loss": 1.1957,
      "num_input_tokens_seen": 113407432,
      "step": 7048
    },
    {
      "epoch": 0.4937700841454552,
      "grad_norm": 4.254641056060791,
      "learning_rate": 5.066934500875656e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 113423392,
      "step": 7049
    },
    {
      "epoch": 0.4938401323911844,
      "grad_norm": 4.007355690002441,
      "learning_rate": 5.066234676007006e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 113439776,
      "step": 7050
    },
    {
      "epoch": 0.49391018063691366,
      "grad_norm": 3.8774940967559814,
      "learning_rate": 5.065534851138355e-05,
      "loss": 0.8772,
      "num_input_tokens_seen": 113456160,
      "step": 7051
    },
    {
      "epoch": 0.4939802288826429,
      "grad_norm": 3.6799323558807373,
      "learning_rate": 5.064835026269702e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 113472408,
      "step": 7052
    },
    {
      "epoch": 0.4940502771283722,
      "grad_norm": 3.662111282348633,
      "learning_rate": 5.064135201401051e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 113488792,
      "step": 7053
    },
    {
      "epoch": 0.4941203253741014,
      "grad_norm": 4.95071268081665,
      "learning_rate": 5.0634353765324004e-05,
      "loss": 0.9709,
      "num_input_tokens_seen": 113505176,
      "step": 7054
    },
    {
      "epoch": 0.49419037361983065,
      "grad_norm": 4.512982368469238,
      "learning_rate": 5.062735551663747e-05,
      "loss": 1.0955,
      "num_input_tokens_seen": 113520640,
      "step": 7055
    },
    {
      "epoch": 0.4942604218655599,
      "grad_norm": 3.6250205039978027,
      "learning_rate": 5.0620357267950955e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 113535616,
      "step": 7056
    },
    {
      "epoch": 0.49433047011128917,
      "grad_norm": 3.856593132019043,
      "learning_rate": 5.0613359019264463e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 113552000,
      "step": 7057
    },
    {
      "epoch": 0.49440051835701837,
      "grad_norm": 3.655444860458374,
      "learning_rate": 5.0606360770577945e-05,
      "loss": 1.1066,
      "num_input_tokens_seen": 113568200,
      "step": 7058
    },
    {
      "epoch": 0.49447056660274763,
      "grad_norm": 4.795759677886963,
      "learning_rate": 5.0599362521891414e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 113584584,
      "step": 7059
    },
    {
      "epoch": 0.4945406148484769,
      "grad_norm": 4.4534783363342285,
      "learning_rate": 5.059236427320491e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 113600968,
      "step": 7060
    },
    {
      "epoch": 0.49461066309420615,
      "grad_norm": 4.681578159332275,
      "learning_rate": 5.05853660245184e-05,
      "loss": 1.3153,
      "num_input_tokens_seen": 113617352,
      "step": 7061
    },
    {
      "epoch": 0.49468071133993535,
      "grad_norm": 3.6734678745269775,
      "learning_rate": 5.057836777583187e-05,
      "loss": 1.0005,
      "num_input_tokens_seen": 113633736,
      "step": 7062
    },
    {
      "epoch": 0.4947507595856646,
      "grad_norm": 3.7208728790283203,
      "learning_rate": 5.057136952714536e-05,
      "loss": 1.0641,
      "num_input_tokens_seen": 113649224,
      "step": 7063
    },
    {
      "epoch": 0.49482080783139387,
      "grad_norm": 4.036924362182617,
      "learning_rate": 5.0564371278458856e-05,
      "loss": 1.0842,
      "num_input_tokens_seen": 113665552,
      "step": 7064
    },
    {
      "epoch": 0.49489085607712313,
      "grad_norm": 6.462393760681152,
      "learning_rate": 5.055737302977234e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 113681192,
      "step": 7065
    },
    {
      "epoch": 0.4949609043228524,
      "grad_norm": 4.203556537628174,
      "learning_rate": 5.055037478108582e-05,
      "loss": 1.3181,
      "num_input_tokens_seen": 113697168,
      "step": 7066
    },
    {
      "epoch": 0.4950309525685816,
      "grad_norm": 3.798896551132202,
      "learning_rate": 5.05433765323993e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 113713504,
      "step": 7067
    },
    {
      "epoch": 0.49510100081431085,
      "grad_norm": 4.175333499908447,
      "learning_rate": 5.053637828371279e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 113729872,
      "step": 7068
    },
    {
      "epoch": 0.4951710490600401,
      "grad_norm": 3.563164234161377,
      "learning_rate": 5.0529380035026266e-05,
      "loss": 0.849,
      "num_input_tokens_seen": 113745736,
      "step": 7069
    },
    {
      "epoch": 0.4952410973057694,
      "grad_norm": 3.605379104614258,
      "learning_rate": 5.0522381786339754e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 113760816,
      "step": 7070
    },
    {
      "epoch": 0.4953111455514986,
      "grad_norm": 3.849106550216675,
      "learning_rate": 5.051538353765325e-05,
      "loss": 0.8297,
      "num_input_tokens_seen": 113776600,
      "step": 7071
    },
    {
      "epoch": 0.49538119379722784,
      "grad_norm": 4.046478748321533,
      "learning_rate": 5.050838528896672e-05,
      "loss": 0.9624,
      "num_input_tokens_seen": 113792984,
      "step": 7072
    },
    {
      "epoch": 0.4954512420429571,
      "grad_norm": 4.66940450668335,
      "learning_rate": 5.050138704028021e-05,
      "loss": 1.1873,
      "num_input_tokens_seen": 113809072,
      "step": 7073
    },
    {
      "epoch": 0.49552129028868636,
      "grad_norm": 5.5129075050354,
      "learning_rate": 5.049438879159371e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 113825456,
      "step": 7074
    },
    {
      "epoch": 0.49559133853441556,
      "grad_norm": 4.045241355895996,
      "learning_rate": 5.048739054290719e-05,
      "loss": 1.0762,
      "num_input_tokens_seen": 113841840,
      "step": 7075
    },
    {
      "epoch": 0.4956613867801448,
      "grad_norm": 3.9198641777038574,
      "learning_rate": 5.0480392294220665e-05,
      "loss": 1.1765,
      "num_input_tokens_seen": 113857624,
      "step": 7076
    },
    {
      "epoch": 0.4957314350258741,
      "grad_norm": 3.836678981781006,
      "learning_rate": 5.047339404553415e-05,
      "loss": 1.1377,
      "num_input_tokens_seen": 113874008,
      "step": 7077
    },
    {
      "epoch": 0.49580148327160334,
      "grad_norm": 3.593061923980713,
      "learning_rate": 5.046639579684764e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 113889992,
      "step": 7078
    },
    {
      "epoch": 0.49587153151733254,
      "grad_norm": 5.477400302886963,
      "learning_rate": 5.045939754816111e-05,
      "loss": 1.1446,
      "num_input_tokens_seen": 113906376,
      "step": 7079
    },
    {
      "epoch": 0.4959415797630618,
      "grad_norm": 5.204897880554199,
      "learning_rate": 5.045239929947462e-05,
      "loss": 1.2356,
      "num_input_tokens_seen": 113921792,
      "step": 7080
    },
    {
      "epoch": 0.49601162800879106,
      "grad_norm": 6.132393836975098,
      "learning_rate": 5.04454010507881e-05,
      "loss": 1.0126,
      "num_input_tokens_seen": 113937720,
      "step": 7081
    },
    {
      "epoch": 0.4960816762545203,
      "grad_norm": 3.651715040206909,
      "learning_rate": 5.043840280210158e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 113953016,
      "step": 7082
    },
    {
      "epoch": 0.4961517245002495,
      "grad_norm": 4.28763484954834,
      "learning_rate": 5.0431404553415065e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 113969400,
      "step": 7083
    },
    {
      "epoch": 0.4962217727459788,
      "grad_norm": 7.6505208015441895,
      "learning_rate": 5.0424406304728554e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 113985784,
      "step": 7084
    },
    {
      "epoch": 0.49629182099170804,
      "grad_norm": 4.85219144821167,
      "learning_rate": 5.0417408056042036e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 114001592,
      "step": 7085
    },
    {
      "epoch": 0.4963618692374373,
      "grad_norm": 3.414391040802002,
      "learning_rate": 5.041040980735552e-05,
      "loss": 1.088,
      "num_input_tokens_seen": 114017976,
      "step": 7086
    },
    {
      "epoch": 0.4964319174831665,
      "grad_norm": 4.361126899719238,
      "learning_rate": 5.040341155866901e-05,
      "loss": 1.0559,
      "num_input_tokens_seen": 114034360,
      "step": 7087
    },
    {
      "epoch": 0.49650196572889577,
      "grad_norm": 3.459439754486084,
      "learning_rate": 5.0396413309982495e-05,
      "loss": 0.8409,
      "num_input_tokens_seen": 114050744,
      "step": 7088
    },
    {
      "epoch": 0.496572013974625,
      "grad_norm": 4.241810321807861,
      "learning_rate": 5.0389415061295976e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 114066776,
      "step": 7089
    },
    {
      "epoch": 0.4966420622203543,
      "grad_norm": 4.012382984161377,
      "learning_rate": 5.0382416812609465e-05,
      "loss": 1.1166,
      "num_input_tokens_seen": 114082888,
      "step": 7090
    },
    {
      "epoch": 0.4967121104660835,
      "grad_norm": 3.8776516914367676,
      "learning_rate": 5.037541856392295e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 114099072,
      "step": 7091
    },
    {
      "epoch": 0.49678215871181275,
      "grad_norm": 4.0513014793396,
      "learning_rate": 5.036842031523643e-05,
      "loss": 0.9747,
      "num_input_tokens_seen": 114115296,
      "step": 7092
    },
    {
      "epoch": 0.496852206957542,
      "grad_norm": 3.7338500022888184,
      "learning_rate": 5.036142206654991e-05,
      "loss": 1.1254,
      "num_input_tokens_seen": 114131680,
      "step": 7093
    },
    {
      "epoch": 0.49692225520327127,
      "grad_norm": 5.892488956451416,
      "learning_rate": 5.035442381786339e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 114146560,
      "step": 7094
    },
    {
      "epoch": 0.49699230344900047,
      "grad_norm": 5.1975507736206055,
      "learning_rate": 5.034742556917689e-05,
      "loss": 1.2128,
      "num_input_tokens_seen": 114162944,
      "step": 7095
    },
    {
      "epoch": 0.49706235169472973,
      "grad_norm": 4.196847438812256,
      "learning_rate": 5.034042732049037e-05,
      "loss": 1.0168,
      "num_input_tokens_seen": 114178040,
      "step": 7096
    },
    {
      "epoch": 0.497132399940459,
      "grad_norm": 4.342573642730713,
      "learning_rate": 5.0333429071803865e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 114194424,
      "step": 7097
    },
    {
      "epoch": 0.49720244818618825,
      "grad_norm": 4.113316059112549,
      "learning_rate": 5.032643082311734e-05,
      "loss": 1.0902,
      "num_input_tokens_seen": 114210808,
      "step": 7098
    },
    {
      "epoch": 0.4972724964319175,
      "grad_norm": 4.835622787475586,
      "learning_rate": 5.031943257443082e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 114225248,
      "step": 7099
    },
    {
      "epoch": 0.4973425446776467,
      "grad_norm": 4.603962421417236,
      "learning_rate": 5.031243432574432e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 114241632,
      "step": 7100
    },
    {
      "epoch": 0.497412592923376,
      "grad_norm": 4.17899227142334,
      "learning_rate": 5.030543607705781e-05,
      "loss": 1.0766,
      "num_input_tokens_seen": 114256688,
      "step": 7101
    },
    {
      "epoch": 0.49748264116910523,
      "grad_norm": 3.890780448913574,
      "learning_rate": 5.029843782837128e-05,
      "loss": 1.089,
      "num_input_tokens_seen": 114273072,
      "step": 7102
    },
    {
      "epoch": 0.4975526894148345,
      "grad_norm": 4.290158748626709,
      "learning_rate": 5.029143957968476e-05,
      "loss": 0.9706,
      "num_input_tokens_seen": 114288936,
      "step": 7103
    },
    {
      "epoch": 0.4976227376605637,
      "grad_norm": 5.222672462463379,
      "learning_rate": 5.028444133099826e-05,
      "loss": 1.0974,
      "num_input_tokens_seen": 114305320,
      "step": 7104
    },
    {
      "epoch": 0.49769278590629296,
      "grad_norm": 3.383232355117798,
      "learning_rate": 5.027744308231174e-05,
      "loss": 1.0279,
      "num_input_tokens_seen": 114321704,
      "step": 7105
    },
    {
      "epoch": 0.4977628341520222,
      "grad_norm": 3.8526852130889893,
      "learning_rate": 5.027044483362522e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 114337920,
      "step": 7106
    },
    {
      "epoch": 0.4978328823977515,
      "grad_norm": 3.699127674102783,
      "learning_rate": 5.026344658493871e-05,
      "loss": 1.1549,
      "num_input_tokens_seen": 114354304,
      "step": 7107
    },
    {
      "epoch": 0.4979029306434807,
      "grad_norm": 3.3088033199310303,
      "learning_rate": 5.025644833625219e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 114370688,
      "step": 7108
    },
    {
      "epoch": 0.49797297888920994,
      "grad_norm": 4.435497760772705,
      "learning_rate": 5.0249450087565674e-05,
      "loss": 1.0656,
      "num_input_tokens_seen": 114386664,
      "step": 7109
    },
    {
      "epoch": 0.4980430271349392,
      "grad_norm": 4.2929840087890625,
      "learning_rate": 5.0242451838879155e-05,
      "loss": 0.9749,
      "num_input_tokens_seen": 114401816,
      "step": 7110
    },
    {
      "epoch": 0.49811307538066846,
      "grad_norm": 4.264016628265381,
      "learning_rate": 5.023545359019266e-05,
      "loss": 1.008,
      "num_input_tokens_seen": 114418200,
      "step": 7111
    },
    {
      "epoch": 0.49818312362639766,
      "grad_norm": 3.5081541538238525,
      "learning_rate": 5.022845534150613e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 114434424,
      "step": 7112
    },
    {
      "epoch": 0.4982531718721269,
      "grad_norm": 5.671893119812012,
      "learning_rate": 5.0221457092819614e-05,
      "loss": 1.0707,
      "num_input_tokens_seen": 114449560,
      "step": 7113
    },
    {
      "epoch": 0.4983232201178562,
      "grad_norm": 4.350570201873779,
      "learning_rate": 5.02144588441331e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 114465504,
      "step": 7114
    },
    {
      "epoch": 0.49839326836358544,
      "grad_norm": 3.650238513946533,
      "learning_rate": 5.0207460595446585e-05,
      "loss": 1.1202,
      "num_input_tokens_seen": 114481760,
      "step": 7115
    },
    {
      "epoch": 0.49846331660931464,
      "grad_norm": 4.211227893829346,
      "learning_rate": 5.020046234676007e-05,
      "loss": 1.228,
      "num_input_tokens_seen": 114496696,
      "step": 7116
    },
    {
      "epoch": 0.4985333648550439,
      "grad_norm": 3.561427354812622,
      "learning_rate": 5.019346409807356e-05,
      "loss": 0.8941,
      "num_input_tokens_seen": 114512616,
      "step": 7117
    },
    {
      "epoch": 0.49860341310077316,
      "grad_norm": 4.558845520019531,
      "learning_rate": 5.018646584938706e-05,
      "loss": 1.1921,
      "num_input_tokens_seen": 114527600,
      "step": 7118
    },
    {
      "epoch": 0.4986734613465024,
      "grad_norm": 3.419285297393799,
      "learning_rate": 5.0179467600700526e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 114543984,
      "step": 7119
    },
    {
      "epoch": 0.4987435095922316,
      "grad_norm": 3.844834566116333,
      "learning_rate": 5.017246935201401e-05,
      "loss": 1.2369,
      "num_input_tokens_seen": 114559728,
      "step": 7120
    },
    {
      "epoch": 0.4988135578379609,
      "grad_norm": 4.457134246826172,
      "learning_rate": 5.0165471103327496e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 114575688,
      "step": 7121
    },
    {
      "epoch": 0.49888360608369015,
      "grad_norm": 4.241283893585205,
      "learning_rate": 5.015847285464098e-05,
      "loss": 1.1434,
      "num_input_tokens_seen": 114590680,
      "step": 7122
    },
    {
      "epoch": 0.4989536543294194,
      "grad_norm": 3.7781248092651367,
      "learning_rate": 5.015147460595446e-05,
      "loss": 0.9872,
      "num_input_tokens_seen": 114606960,
      "step": 7123
    },
    {
      "epoch": 0.4990237025751486,
      "grad_norm": 5.492437839508057,
      "learning_rate": 5.0144476357267955e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 114623344,
      "step": 7124
    },
    {
      "epoch": 0.49909375082087787,
      "grad_norm": 5.001891613006592,
      "learning_rate": 5.013747810858145e-05,
      "loss": 1.1999,
      "num_input_tokens_seen": 114639728,
      "step": 7125
    },
    {
      "epoch": 0.49916379906660713,
      "grad_norm": 3.78376841545105,
      "learning_rate": 5.013047985989493e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 114655984,
      "step": 7126
    },
    {
      "epoch": 0.4992338473123364,
      "grad_norm": 5.250494956970215,
      "learning_rate": 5.0123481611208414e-05,
      "loss": 1.1063,
      "num_input_tokens_seen": 114670912,
      "step": 7127
    },
    {
      "epoch": 0.4993038955580656,
      "grad_norm": 3.8290820121765137,
      "learning_rate": 5.01164833625219e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 114687240,
      "step": 7128
    },
    {
      "epoch": 0.49937394380379485,
      "grad_norm": 5.523165225982666,
      "learning_rate": 5.010948511383538e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 114703616,
      "step": 7129
    },
    {
      "epoch": 0.4994439920495241,
      "grad_norm": 3.838224172592163,
      "learning_rate": 5.010248686514886e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 114720000,
      "step": 7130
    },
    {
      "epoch": 0.49951404029525337,
      "grad_norm": 3.751004457473755,
      "learning_rate": 5.009548861646235e-05,
      "loss": 1.0345,
      "num_input_tokens_seen": 114735624,
      "step": 7131
    },
    {
      "epoch": 0.49958408854098263,
      "grad_norm": 4.485782146453857,
      "learning_rate": 5.0088490367775843e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 114751040,
      "step": 7132
    },
    {
      "epoch": 0.49965413678671183,
      "grad_norm": 4.896092891693115,
      "learning_rate": 5.0081492119089325e-05,
      "loss": 1.1213,
      "num_input_tokens_seen": 114766096,
      "step": 7133
    },
    {
      "epoch": 0.4997241850324411,
      "grad_norm": 4.27908182144165,
      "learning_rate": 5.007449387040281e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 114782480,
      "step": 7134
    },
    {
      "epoch": 0.49979423327817035,
      "grad_norm": 5.309985160827637,
      "learning_rate": 5.0067495621716296e-05,
      "loss": 1.1707,
      "num_input_tokens_seen": 114798864,
      "step": 7135
    },
    {
      "epoch": 0.4998642815238996,
      "grad_norm": 3.838355302810669,
      "learning_rate": 5.006049737302978e-05,
      "loss": 1.243,
      "num_input_tokens_seen": 114814680,
      "step": 7136
    },
    {
      "epoch": 0.4999343297696288,
      "grad_norm": 3.9620189666748047,
      "learning_rate": 5.005349912434325e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 114831064,
      "step": 7137
    },
    {
      "epoch": 0.5000043780153581,
      "grad_norm": 3.4240174293518066,
      "learning_rate": 5.004650087565674e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 114847448,
      "step": 7138
    },
    {
      "epoch": 0.5000744262610873,
      "grad_norm": 3.603026866912842,
      "learning_rate": 5.003950262697022e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 114863832,
      "step": 7139
    },
    {
      "epoch": 0.5001444745068165,
      "grad_norm": 4.349592208862305,
      "learning_rate": 5.0032504378283705e-05,
      "loss": 1.1059,
      "num_input_tokens_seen": 114879200,
      "step": 7140
    },
    {
      "epoch": 0.5002145227525459,
      "grad_norm": 5.716104984283447,
      "learning_rate": 5.00255061295972e-05,
      "loss": 1.0465,
      "num_input_tokens_seen": 114894880,
      "step": 7141
    },
    {
      "epoch": 0.5002845709982751,
      "grad_norm": 3.857797384262085,
      "learning_rate": 5.0018507880910695e-05,
      "loss": 1.0438,
      "num_input_tokens_seen": 114911264,
      "step": 7142
    },
    {
      "epoch": 0.5003546192440043,
      "grad_norm": 3.7292556762695312,
      "learning_rate": 5.001150963222417e-05,
      "loss": 1.0803,
      "num_input_tokens_seen": 114926792,
      "step": 7143
    },
    {
      "epoch": 0.5004246674897336,
      "grad_norm": 4.02719783782959,
      "learning_rate": 5.000451138353765e-05,
      "loss": 0.9635,
      "num_input_tokens_seen": 114942944,
      "step": 7144
    },
    {
      "epoch": 0.5004947157354628,
      "grad_norm": 5.39168119430542,
      "learning_rate": 4.9997513134851134e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 114958800,
      "step": 7145
    },
    {
      "epoch": 0.5005647639811921,
      "grad_norm": 4.773622512817383,
      "learning_rate": 4.999051488616463e-05,
      "loss": 1.1112,
      "num_input_tokens_seen": 114974664,
      "step": 7146
    },
    {
      "epoch": 0.5006348122269213,
      "grad_norm": 3.635557174682617,
      "learning_rate": 4.998351663747812e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 114990728,
      "step": 7147
    },
    {
      "epoch": 0.5007048604726505,
      "grad_norm": 4.165726661682129,
      "learning_rate": 4.997651838879159e-05,
      "loss": 1.1044,
      "num_input_tokens_seen": 115006552,
      "step": 7148
    },
    {
      "epoch": 0.5007749087183798,
      "grad_norm": 4.2835001945495605,
      "learning_rate": 4.996952014010508e-05,
      "loss": 0.937,
      "num_input_tokens_seen": 115022936,
      "step": 7149
    },
    {
      "epoch": 0.500844956964109,
      "grad_norm": 3.7588231563568115,
      "learning_rate": 4.9962521891418564e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 115038832,
      "step": 7150
    },
    {
      "epoch": 0.5009150052098382,
      "grad_norm": 4.017446041107178,
      "learning_rate": 4.995552364273205e-05,
      "loss": 1.0177,
      "num_input_tokens_seen": 115055216,
      "step": 7151
    },
    {
      "epoch": 0.5009850534555675,
      "grad_norm": 4.145601749420166,
      "learning_rate": 4.994852539404554e-05,
      "loss": 1.1925,
      "num_input_tokens_seen": 115071600,
      "step": 7152
    },
    {
      "epoch": 0.5010551017012967,
      "grad_norm": 4.027134895324707,
      "learning_rate": 4.994152714535902e-05,
      "loss": 1.0115,
      "num_input_tokens_seen": 115087504,
      "step": 7153
    },
    {
      "epoch": 0.5011251499470261,
      "grad_norm": 4.185591697692871,
      "learning_rate": 4.993452889667251e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 115102936,
      "step": 7154
    },
    {
      "epoch": 0.5011951981927553,
      "grad_norm": 3.262739658355713,
      "learning_rate": 4.9927530647985986e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 115119320,
      "step": 7155
    },
    {
      "epoch": 0.5012652464384845,
      "grad_norm": 3.514493465423584,
      "learning_rate": 4.992053239929948e-05,
      "loss": 0.9523,
      "num_input_tokens_seen": 115135704,
      "step": 7156
    },
    {
      "epoch": 0.5013352946842138,
      "grad_norm": 3.2577719688415527,
      "learning_rate": 4.991353415061297e-05,
      "loss": 0.9603,
      "num_input_tokens_seen": 115152088,
      "step": 7157
    },
    {
      "epoch": 0.501405342929943,
      "grad_norm": 4.475879669189453,
      "learning_rate": 4.9906535901926445e-05,
      "loss": 1.1393,
      "num_input_tokens_seen": 115168472,
      "step": 7158
    },
    {
      "epoch": 0.5014753911756722,
      "grad_norm": 4.558653354644775,
      "learning_rate": 4.9899537653239934e-05,
      "loss": 0.9628,
      "num_input_tokens_seen": 115184136,
      "step": 7159
    },
    {
      "epoch": 0.5015454394214015,
      "grad_norm": 4.034858703613281,
      "learning_rate": 4.9892539404553416e-05,
      "loss": 1.2669,
      "num_input_tokens_seen": 115200520,
      "step": 7160
    },
    {
      "epoch": 0.5016154876671307,
      "grad_norm": 4.190174579620361,
      "learning_rate": 4.9885541155866904e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 115216904,
      "step": 7161
    },
    {
      "epoch": 0.50168553591286,
      "grad_norm": 4.808748245239258,
      "learning_rate": 4.987854290718039e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 115232864,
      "step": 7162
    },
    {
      "epoch": 0.5017555841585892,
      "grad_norm": 3.7075023651123047,
      "learning_rate": 4.9871544658493875e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 115248544,
      "step": 7163
    },
    {
      "epoch": 0.5018256324043184,
      "grad_norm": 3.6593689918518066,
      "learning_rate": 4.986454640980736e-05,
      "loss": 1.0653,
      "num_input_tokens_seen": 115264616,
      "step": 7164
    },
    {
      "epoch": 0.5018956806500477,
      "grad_norm": 3.959949493408203,
      "learning_rate": 4.985754816112084e-05,
      "loss": 1.0708,
      "num_input_tokens_seen": 115281000,
      "step": 7165
    },
    {
      "epoch": 0.501965728895777,
      "grad_norm": 3.6724140644073486,
      "learning_rate": 4.9850549912434334e-05,
      "loss": 1.1306,
      "num_input_tokens_seen": 115297384,
      "step": 7166
    },
    {
      "epoch": 0.5020357771415063,
      "grad_norm": 3.9350247383117676,
      "learning_rate": 4.984355166374781e-05,
      "loss": 1.1785,
      "num_input_tokens_seen": 115312760,
      "step": 7167
    },
    {
      "epoch": 0.5021058253872355,
      "grad_norm": 3.8056607246398926,
      "learning_rate": 4.98365534150613e-05,
      "loss": 0.8915,
      "num_input_tokens_seen": 115328336,
      "step": 7168
    },
    {
      "epoch": 0.5021758736329647,
      "grad_norm": 3.995048761367798,
      "learning_rate": 4.9829555166374786e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 115344016,
      "step": 7169
    },
    {
      "epoch": 0.502245921878694,
      "grad_norm": 4.534327983856201,
      "learning_rate": 4.982255691768827e-05,
      "loss": 0.9757,
      "num_input_tokens_seen": 115359976,
      "step": 7170
    },
    {
      "epoch": 0.5023159701244232,
      "grad_norm": 5.29775333404541,
      "learning_rate": 4.9815558669001756e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 115375480,
      "step": 7171
    },
    {
      "epoch": 0.5023860183701524,
      "grad_norm": 3.8773534297943115,
      "learning_rate": 4.980856042031524e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 115391632,
      "step": 7172
    },
    {
      "epoch": 0.5024560666158817,
      "grad_norm": 4.249567985534668,
      "learning_rate": 4.9801562171628727e-05,
      "loss": 1.1145,
      "num_input_tokens_seen": 115408016,
      "step": 7173
    },
    {
      "epoch": 0.5025261148616109,
      "grad_norm": 4.293243408203125,
      "learning_rate": 4.9794563922942215e-05,
      "loss": 1.4234,
      "num_input_tokens_seen": 115423808,
      "step": 7174
    },
    {
      "epoch": 0.5025961631073402,
      "grad_norm": 4.535524845123291,
      "learning_rate": 4.978756567425569e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 115440192,
      "step": 7175
    },
    {
      "epoch": 0.5026662113530694,
      "grad_norm": 4.3390631675720215,
      "learning_rate": 4.9780567425569186e-05,
      "loss": 1.3728,
      "num_input_tokens_seen": 115455416,
      "step": 7176
    },
    {
      "epoch": 0.5027362595987986,
      "grad_norm": 3.630815267562866,
      "learning_rate": 4.977356917688266e-05,
      "loss": 1.0493,
      "num_input_tokens_seen": 115471800,
      "step": 7177
    },
    {
      "epoch": 0.502806307844528,
      "grad_norm": 3.9146728515625,
      "learning_rate": 4.976657092819615e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 115488184,
      "step": 7178
    },
    {
      "epoch": 0.5028763560902572,
      "grad_norm": 4.978190898895264,
      "learning_rate": 4.975957267950964e-05,
      "loss": 1.0014,
      "num_input_tokens_seen": 115503792,
      "step": 7179
    },
    {
      "epoch": 0.5029464043359864,
      "grad_norm": 3.8975963592529297,
      "learning_rate": 4.975257443082312e-05,
      "loss": 1.1317,
      "num_input_tokens_seen": 115519576,
      "step": 7180
    },
    {
      "epoch": 0.5030164525817157,
      "grad_norm": 4.439699649810791,
      "learning_rate": 4.974557618213661e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 115535512,
      "step": 7181
    },
    {
      "epoch": 0.5030865008274449,
      "grad_norm": 4.4080610275268555,
      "learning_rate": 4.973857793345009e-05,
      "loss": 1.3639,
      "num_input_tokens_seen": 115551880,
      "step": 7182
    },
    {
      "epoch": 0.5031565490731742,
      "grad_norm": 3.8968825340270996,
      "learning_rate": 4.973157968476358e-05,
      "loss": 1.1686,
      "num_input_tokens_seen": 115568136,
      "step": 7183
    },
    {
      "epoch": 0.5032265973189034,
      "grad_norm": 4.030379295349121,
      "learning_rate": 4.972458143607707e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 115583928,
      "step": 7184
    },
    {
      "epoch": 0.5032966455646326,
      "grad_norm": 4.46726131439209,
      "learning_rate": 4.971758318739054e-05,
      "loss": 1.1789,
      "num_input_tokens_seen": 115600016,
      "step": 7185
    },
    {
      "epoch": 0.5033666938103619,
      "grad_norm": 4.490327835083008,
      "learning_rate": 4.971058493870404e-05,
      "loss": 0.968,
      "num_input_tokens_seen": 115616400,
      "step": 7186
    },
    {
      "epoch": 0.5034367420560911,
      "grad_norm": 5.678159713745117,
      "learning_rate": 4.970358669001751e-05,
      "loss": 1.2517,
      "num_input_tokens_seen": 115632424,
      "step": 7187
    },
    {
      "epoch": 0.5035067903018203,
      "grad_norm": 4.695899963378906,
      "learning_rate": 4.9696588441331e-05,
      "loss": 1.2681,
      "num_input_tokens_seen": 115648808,
      "step": 7188
    },
    {
      "epoch": 0.5035768385475496,
      "grad_norm": 3.6823155879974365,
      "learning_rate": 4.968959019264449e-05,
      "loss": 1.0642,
      "num_input_tokens_seen": 115665192,
      "step": 7189
    },
    {
      "epoch": 0.5036468867932788,
      "grad_norm": 5.105508804321289,
      "learning_rate": 4.968259194395797e-05,
      "loss": 1.1646,
      "num_input_tokens_seen": 115681472,
      "step": 7190
    },
    {
      "epoch": 0.5037169350390082,
      "grad_norm": 4.0591607093811035,
      "learning_rate": 4.967559369527146e-05,
      "loss": 1.1231,
      "num_input_tokens_seen": 115697248,
      "step": 7191
    },
    {
      "epoch": 0.5037869832847374,
      "grad_norm": 4.097674369812012,
      "learning_rate": 4.966859544658494e-05,
      "loss": 0.9121,
      "num_input_tokens_seen": 115713632,
      "step": 7192
    },
    {
      "epoch": 0.5038570315304666,
      "grad_norm": 3.711235523223877,
      "learning_rate": 4.966159719789843e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 115730016,
      "step": 7193
    },
    {
      "epoch": 0.5039270797761959,
      "grad_norm": 3.9073588848114014,
      "learning_rate": 4.9654598949211906e-05,
      "loss": 1.0806,
      "num_input_tokens_seen": 115745872,
      "step": 7194
    },
    {
      "epoch": 0.5039971280219251,
      "grad_norm": 3.230870008468628,
      "learning_rate": 4.9647600700525394e-05,
      "loss": 0.8976,
      "num_input_tokens_seen": 115762256,
      "step": 7195
    },
    {
      "epoch": 0.5040671762676543,
      "grad_norm": 4.253819942474365,
      "learning_rate": 4.964060245183889e-05,
      "loss": 1.1024,
      "num_input_tokens_seen": 115778640,
      "step": 7196
    },
    {
      "epoch": 0.5041372245133836,
      "grad_norm": 3.6932590007781982,
      "learning_rate": 4.9633604203152365e-05,
      "loss": 1.1522,
      "num_input_tokens_seen": 115795024,
      "step": 7197
    },
    {
      "epoch": 0.5042072727591128,
      "grad_norm": 4.178073883056641,
      "learning_rate": 4.962660595446585e-05,
      "loss": 1.1629,
      "num_input_tokens_seen": 115811408,
      "step": 7198
    },
    {
      "epoch": 0.5042773210048421,
      "grad_norm": 3.4744091033935547,
      "learning_rate": 4.9619607705779335e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 115827792,
      "step": 7199
    },
    {
      "epoch": 0.5043473692505713,
      "grad_norm": 5.810272216796875,
      "learning_rate": 4.9612609457092824e-05,
      "loss": 1.0187,
      "num_input_tokens_seen": 115843352,
      "step": 7200
    },
    {
      "epoch": 0.5043473692505713,
      "eval_loss": 1.1233556270599365,
      "eval_runtime": 0.159,
      "eval_samples_per_second": 6.288,
      "eval_steps_per_second": 6.288,
      "num_input_tokens_seen": 115843352,
      "step": 7200
    },
    {
      "epoch": 0.5044174174963005,
      "grad_norm": 4.0738205909729,
      "learning_rate": 4.960561120840631e-05,
      "loss": 1.2362,
      "num_input_tokens_seen": 115859216,
      "step": 7201
    },
    {
      "epoch": 0.5044874657420299,
      "grad_norm": 4.0072102546691895,
      "learning_rate": 4.9598612959719794e-05,
      "loss": 1.1624,
      "num_input_tokens_seen": 115875600,
      "step": 7202
    },
    {
      "epoch": 0.504557513987759,
      "grad_norm": 5.232552528381348,
      "learning_rate": 4.959161471103328e-05,
      "loss": 0.9148,
      "num_input_tokens_seen": 115890976,
      "step": 7203
    },
    {
      "epoch": 0.5046275622334884,
      "grad_norm": 4.6930623054504395,
      "learning_rate": 4.958461646234676e-05,
      "loss": 1.0314,
      "num_input_tokens_seen": 115907360,
      "step": 7204
    },
    {
      "epoch": 0.5046976104792176,
      "grad_norm": 5.217222690582275,
      "learning_rate": 4.9577618213660246e-05,
      "loss": 1.1316,
      "num_input_tokens_seen": 115923320,
      "step": 7205
    },
    {
      "epoch": 0.5047676587249468,
      "grad_norm": 3.999408006668091,
      "learning_rate": 4.957061996497374e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 115939704,
      "step": 7206
    },
    {
      "epoch": 0.5048377069706761,
      "grad_norm": 4.267052173614502,
      "learning_rate": 4.956362171628722e-05,
      "loss": 1.006,
      "num_input_tokens_seen": 115955592,
      "step": 7207
    },
    {
      "epoch": 0.5049077552164053,
      "grad_norm": 4.446041584014893,
      "learning_rate": 4.9556623467600705e-05,
      "loss": 1.2351,
      "num_input_tokens_seen": 115971976,
      "step": 7208
    },
    {
      "epoch": 0.5049778034621345,
      "grad_norm": 3.8210396766662598,
      "learning_rate": 4.954962521891419e-05,
      "loss": 1.1836,
      "num_input_tokens_seen": 115987528,
      "step": 7209
    },
    {
      "epoch": 0.5050478517078638,
      "grad_norm": 5.992397785186768,
      "learning_rate": 4.9542626970227676e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 116003912,
      "step": 7210
    },
    {
      "epoch": 0.505117899953593,
      "grad_norm": 3.934375524520874,
      "learning_rate": 4.9535628721541164e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 116020296,
      "step": 7211
    },
    {
      "epoch": 0.5051879481993223,
      "grad_norm": 3.936866521835327,
      "learning_rate": 4.9528630472854646e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 116036600,
      "step": 7212
    },
    {
      "epoch": 0.5052579964450515,
      "grad_norm": 4.908316135406494,
      "learning_rate": 4.9521632224168135e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 116052984,
      "step": 7213
    },
    {
      "epoch": 0.5053280446907807,
      "grad_norm": 4.035202503204346,
      "learning_rate": 4.951463397548161e-05,
      "loss": 1.1475,
      "num_input_tokens_seen": 116068768,
      "step": 7214
    },
    {
      "epoch": 0.5053980929365101,
      "grad_norm": 3.7488014698028564,
      "learning_rate": 4.95076357267951e-05,
      "loss": 0.9506,
      "num_input_tokens_seen": 116085152,
      "step": 7215
    },
    {
      "epoch": 0.5054681411822393,
      "grad_norm": 5.226819038391113,
      "learning_rate": 4.9500637478108594e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 116100176,
      "step": 7216
    },
    {
      "epoch": 0.5055381894279685,
      "grad_norm": 4.0122857093811035,
      "learning_rate": 4.949363922942207e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 116116560,
      "step": 7217
    },
    {
      "epoch": 0.5056082376736978,
      "grad_norm": 4.160411834716797,
      "learning_rate": 4.948664098073556e-05,
      "loss": 0.9622,
      "num_input_tokens_seen": 116132464,
      "step": 7218
    },
    {
      "epoch": 0.505678285919427,
      "grad_norm": 4.860180377960205,
      "learning_rate": 4.947964273204904e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 116148848,
      "step": 7219
    },
    {
      "epoch": 0.5057483341651563,
      "grad_norm": 4.549893856048584,
      "learning_rate": 4.947264448336253e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 116165128,
      "step": 7220
    },
    {
      "epoch": 0.5058183824108855,
      "grad_norm": 3.4614131450653076,
      "learning_rate": 4.9465646234676e-05,
      "loss": 0.7666,
      "num_input_tokens_seen": 116181152,
      "step": 7221
    },
    {
      "epoch": 0.5058884306566147,
      "grad_norm": 4.7237043380737305,
      "learning_rate": 4.94586479859895e-05,
      "loss": 0.949,
      "num_input_tokens_seen": 116197032,
      "step": 7222
    },
    {
      "epoch": 0.505958478902344,
      "grad_norm": 4.4195098876953125,
      "learning_rate": 4.945164973730299e-05,
      "loss": 0.985,
      "num_input_tokens_seen": 116212752,
      "step": 7223
    },
    {
      "epoch": 0.5060285271480732,
      "grad_norm": 3.6815669536590576,
      "learning_rate": 4.944465148861646e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 116229024,
      "step": 7224
    },
    {
      "epoch": 0.5060985753938024,
      "grad_norm": 3.776644229888916,
      "learning_rate": 4.943765323992995e-05,
      "loss": 0.9953,
      "num_input_tokens_seen": 116245408,
      "step": 7225
    },
    {
      "epoch": 0.5061686236395317,
      "grad_norm": 4.3324761390686035,
      "learning_rate": 4.943065499124343e-05,
      "loss": 1.058,
      "num_input_tokens_seen": 116261192,
      "step": 7226
    },
    {
      "epoch": 0.506238671885261,
      "grad_norm": 3.499302387237549,
      "learning_rate": 4.942365674255692e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 116277576,
      "step": 7227
    },
    {
      "epoch": 0.5063087201309903,
      "grad_norm": 3.5195088386535645,
      "learning_rate": 4.941665849387041e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 116293960,
      "step": 7228
    },
    {
      "epoch": 0.5063787683767195,
      "grad_norm": 4.266250133514404,
      "learning_rate": 4.940966024518389e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 116310344,
      "step": 7229
    },
    {
      "epoch": 0.5064488166224487,
      "grad_norm": 4.53155517578125,
      "learning_rate": 4.940266199649738e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 116326536,
      "step": 7230
    },
    {
      "epoch": 0.506518864868178,
      "grad_norm": 5.224839210510254,
      "learning_rate": 4.9395663747810855e-05,
      "loss": 1.1932,
      "num_input_tokens_seen": 116342792,
      "step": 7231
    },
    {
      "epoch": 0.5065889131139072,
      "grad_norm": 4.072076797485352,
      "learning_rate": 4.938866549912435e-05,
      "loss": 1.1041,
      "num_input_tokens_seen": 116359144,
      "step": 7232
    },
    {
      "epoch": 0.5066589613596364,
      "grad_norm": 4.286440372467041,
      "learning_rate": 4.938166725043784e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 116375128,
      "step": 7233
    },
    {
      "epoch": 0.5067290096053657,
      "grad_norm": 3.684030055999756,
      "learning_rate": 4.9374669001751314e-05,
      "loss": 0.8753,
      "num_input_tokens_seen": 116391448,
      "step": 7234
    },
    {
      "epoch": 0.5067990578510949,
      "grad_norm": 3.721698045730591,
      "learning_rate": 4.93676707530648e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 116407832,
      "step": 7235
    },
    {
      "epoch": 0.5068691060968242,
      "grad_norm": 3.5029869079589844,
      "learning_rate": 4.9360672504378284e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 116424216,
      "step": 7236
    },
    {
      "epoch": 0.5069391543425534,
      "grad_norm": 3.960109233856201,
      "learning_rate": 4.935367425569177e-05,
      "loss": 1.2421,
      "num_input_tokens_seen": 116440600,
      "step": 7237
    },
    {
      "epoch": 0.5070092025882826,
      "grad_norm": 3.7146995067596436,
      "learning_rate": 4.934667600700526e-05,
      "loss": 1.0466,
      "num_input_tokens_seen": 116455776,
      "step": 7238
    },
    {
      "epoch": 0.507079250834012,
      "grad_norm": 4.000344753265381,
      "learning_rate": 4.933967775831874e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 116472160,
      "step": 7239
    },
    {
      "epoch": 0.5071492990797412,
      "grad_norm": 4.015896320343018,
      "learning_rate": 4.933267950963223e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 116488440,
      "step": 7240
    },
    {
      "epoch": 0.5072193473254705,
      "grad_norm": 4.240390777587891,
      "learning_rate": 4.932568126094571e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 116504112,
      "step": 7241
    },
    {
      "epoch": 0.5072893955711997,
      "grad_norm": 4.051314353942871,
      "learning_rate": 4.93186830122592e-05,
      "loss": 1.1068,
      "num_input_tokens_seen": 116519976,
      "step": 7242
    },
    {
      "epoch": 0.5073594438169289,
      "grad_norm": 4.370121955871582,
      "learning_rate": 4.931168476357269e-05,
      "loss": 1.1409,
      "num_input_tokens_seen": 116536360,
      "step": 7243
    },
    {
      "epoch": 0.5074294920626582,
      "grad_norm": 3.7158761024475098,
      "learning_rate": 4.9304686514886166e-05,
      "loss": 1.0234,
      "num_input_tokens_seen": 116552744,
      "step": 7244
    },
    {
      "epoch": 0.5074995403083874,
      "grad_norm": 3.6040024757385254,
      "learning_rate": 4.9297688266199654e-05,
      "loss": 0.8008,
      "num_input_tokens_seen": 116568088,
      "step": 7245
    },
    {
      "epoch": 0.5075695885541166,
      "grad_norm": 5.175736904144287,
      "learning_rate": 4.9290690017513136e-05,
      "loss": 1.09,
      "num_input_tokens_seen": 116584152,
      "step": 7246
    },
    {
      "epoch": 0.5076396367998459,
      "grad_norm": 4.735289573669434,
      "learning_rate": 4.9283691768826625e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 116599848,
      "step": 7247
    },
    {
      "epoch": 0.5077096850455751,
      "grad_norm": 5.659826278686523,
      "learning_rate": 4.9276693520140107e-05,
      "loss": 1.1248,
      "num_input_tokens_seen": 116615216,
      "step": 7248
    },
    {
      "epoch": 0.5077797332913044,
      "grad_norm": 4.524930000305176,
      "learning_rate": 4.9269695271453595e-05,
      "loss": 1.0244,
      "num_input_tokens_seen": 116631328,
      "step": 7249
    },
    {
      "epoch": 0.5078497815370336,
      "grad_norm": 3.6031768321990967,
      "learning_rate": 4.9262697022767084e-05,
      "loss": 1.0042,
      "num_input_tokens_seen": 116647016,
      "step": 7250
    },
    {
      "epoch": 0.5079198297827628,
      "grad_norm": 3.953381299972534,
      "learning_rate": 4.925569877408056e-05,
      "loss": 0.9817,
      "num_input_tokens_seen": 116663400,
      "step": 7251
    },
    {
      "epoch": 0.5079898780284922,
      "grad_norm": 7.162896633148193,
      "learning_rate": 4.9248700525394054e-05,
      "loss": 1.2361,
      "num_input_tokens_seen": 116679456,
      "step": 7252
    },
    {
      "epoch": 0.5080599262742214,
      "grad_norm": 3.416929006576538,
      "learning_rate": 4.924170227670753e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 116695840,
      "step": 7253
    },
    {
      "epoch": 0.5081299745199506,
      "grad_norm": 3.542628049850464,
      "learning_rate": 4.923470402802102e-05,
      "loss": 0.8451,
      "num_input_tokens_seen": 116712224,
      "step": 7254
    },
    {
      "epoch": 0.5082000227656799,
      "grad_norm": 5.850252151489258,
      "learning_rate": 4.9227705779334506e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 116728608,
      "step": 7255
    },
    {
      "epoch": 0.5082700710114091,
      "grad_norm": 4.921962261199951,
      "learning_rate": 4.922070753064799e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 116744992,
      "step": 7256
    },
    {
      "epoch": 0.5083401192571384,
      "grad_norm": 5.621464252471924,
      "learning_rate": 4.921370928196148e-05,
      "loss": 1.0273,
      "num_input_tokens_seen": 116761376,
      "step": 7257
    },
    {
      "epoch": 0.5084101675028676,
      "grad_norm": 3.8369336128234863,
      "learning_rate": 4.920671103327496e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 116777656,
      "step": 7258
    },
    {
      "epoch": 0.5084802157485968,
      "grad_norm": 4.033676624298096,
      "learning_rate": 4.919971278458845e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 116793432,
      "step": 7259
    },
    {
      "epoch": 0.5085502639943261,
      "grad_norm": 5.80480432510376,
      "learning_rate": 4.9192714535901936e-05,
      "loss": 0.9104,
      "num_input_tokens_seen": 116809816,
      "step": 7260
    },
    {
      "epoch": 0.5086203122400553,
      "grad_norm": 4.646456241607666,
      "learning_rate": 4.918571628721541e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 116826024,
      "step": 7261
    },
    {
      "epoch": 0.5086903604857845,
      "grad_norm": 5.681286811828613,
      "learning_rate": 4.9178718038528906e-05,
      "loss": 1.0206,
      "num_input_tokens_seen": 116841176,
      "step": 7262
    },
    {
      "epoch": 0.5087604087315138,
      "grad_norm": 4.391019821166992,
      "learning_rate": 4.917171978984238e-05,
      "loss": 0.9668,
      "num_input_tokens_seen": 116857560,
      "step": 7263
    },
    {
      "epoch": 0.508830456977243,
      "grad_norm": 3.69963002204895,
      "learning_rate": 4.916472154115587e-05,
      "loss": 1.1773,
      "num_input_tokens_seen": 116873872,
      "step": 7264
    },
    {
      "epoch": 0.5089005052229724,
      "grad_norm": 4.748251914978027,
      "learning_rate": 4.915772329246936e-05,
      "loss": 1.1401,
      "num_input_tokens_seen": 116890256,
      "step": 7265
    },
    {
      "epoch": 0.5089705534687016,
      "grad_norm": 4.31103515625,
      "learning_rate": 4.915072504378284e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 116906640,
      "step": 7266
    },
    {
      "epoch": 0.5090406017144308,
      "grad_norm": 4.616788864135742,
      "learning_rate": 4.914372679509633e-05,
      "loss": 1.0779,
      "num_input_tokens_seen": 116923024,
      "step": 7267
    },
    {
      "epoch": 0.5091106499601601,
      "grad_norm": 5.435046672821045,
      "learning_rate": 4.913672854640981e-05,
      "loss": 1.1371,
      "num_input_tokens_seen": 116939264,
      "step": 7268
    },
    {
      "epoch": 0.5091806982058893,
      "grad_norm": 3.534294843673706,
      "learning_rate": 4.91297302977233e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 116955648,
      "step": 7269
    },
    {
      "epoch": 0.5092507464516186,
      "grad_norm": 4.072021961212158,
      "learning_rate": 4.912273204903679e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 116970936,
      "step": 7270
    },
    {
      "epoch": 0.5093207946973478,
      "grad_norm": 3.853341817855835,
      "learning_rate": 4.911573380035026e-05,
      "loss": 1.0749,
      "num_input_tokens_seen": 116987064,
      "step": 7271
    },
    {
      "epoch": 0.509390842943077,
      "grad_norm": 5.853321552276611,
      "learning_rate": 4.910873555166375e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 117002752,
      "step": 7272
    },
    {
      "epoch": 0.5094608911888063,
      "grad_norm": 3.7305798530578613,
      "learning_rate": 4.910173730297723e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 117019136,
      "step": 7273
    },
    {
      "epoch": 0.5095309394345355,
      "grad_norm": 3.9224064350128174,
      "learning_rate": 4.909473905429072e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 117035504,
      "step": 7274
    },
    {
      "epoch": 0.5096009876802647,
      "grad_norm": 3.633242130279541,
      "learning_rate": 4.9087740805604204e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 117051888,
      "step": 7275
    },
    {
      "epoch": 0.509671035925994,
      "grad_norm": 3.7328341007232666,
      "learning_rate": 4.908074255691769e-05,
      "loss": 1.1986,
      "num_input_tokens_seen": 117068272,
      "step": 7276
    },
    {
      "epoch": 0.5097410841717233,
      "grad_norm": 3.492896556854248,
      "learning_rate": 4.907374430823118e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 117084656,
      "step": 7277
    },
    {
      "epoch": 0.5098111324174526,
      "grad_norm": 3.8437516689300537,
      "learning_rate": 4.906674605954466e-05,
      "loss": 1.1177,
      "num_input_tokens_seen": 117100432,
      "step": 7278
    },
    {
      "epoch": 0.5098811806631818,
      "grad_norm": 3.537297487258911,
      "learning_rate": 4.905974781085815e-05,
      "loss": 1.1694,
      "num_input_tokens_seen": 117116544,
      "step": 7279
    },
    {
      "epoch": 0.509951228908911,
      "grad_norm": 3.6758127212524414,
      "learning_rate": 4.9052749562171626e-05,
      "loss": 0.8982,
      "num_input_tokens_seen": 117132376,
      "step": 7280
    },
    {
      "epoch": 0.5100212771546403,
      "grad_norm": 3.7280797958374023,
      "learning_rate": 4.9045751313485115e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 117148760,
      "step": 7281
    },
    {
      "epoch": 0.5100913254003695,
      "grad_norm": 4.579721927642822,
      "learning_rate": 4.9038753064798603e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 117165144,
      "step": 7282
    },
    {
      "epoch": 0.5101613736460987,
      "grad_norm": 4.146833896636963,
      "learning_rate": 4.9031754816112085e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 117180472,
      "step": 7283
    },
    {
      "epoch": 0.510231421891828,
      "grad_norm": 3.7897355556488037,
      "learning_rate": 4.9024756567425574e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 117196856,
      "step": 7284
    },
    {
      "epoch": 0.5103014701375572,
      "grad_norm": 3.821641206741333,
      "learning_rate": 4.9017758318739056e-05,
      "loss": 1.0827,
      "num_input_tokens_seen": 117213240,
      "step": 7285
    },
    {
      "epoch": 0.5103715183832865,
      "grad_norm": 4.439133644104004,
      "learning_rate": 4.9010760070052544e-05,
      "loss": 1.0325,
      "num_input_tokens_seen": 117229240,
      "step": 7286
    },
    {
      "epoch": 0.5104415666290157,
      "grad_norm": 4.781843185424805,
      "learning_rate": 4.900376182136603e-05,
      "loss": 1.0871,
      "num_input_tokens_seen": 117245624,
      "step": 7287
    },
    {
      "epoch": 0.510511614874745,
      "grad_norm": 3.928457736968994,
      "learning_rate": 4.8996763572679515e-05,
      "loss": 1.0049,
      "num_input_tokens_seen": 117261496,
      "step": 7288
    },
    {
      "epoch": 0.5105816631204743,
      "grad_norm": 3.8278815746307373,
      "learning_rate": 4.8989765323993e-05,
      "loss": 0.8442,
      "num_input_tokens_seen": 117277448,
      "step": 7289
    },
    {
      "epoch": 0.5106517113662035,
      "grad_norm": 3.8238277435302734,
      "learning_rate": 4.898276707530648e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 117293832,
      "step": 7290
    },
    {
      "epoch": 0.5107217596119327,
      "grad_norm": 3.835528612136841,
      "learning_rate": 4.897576882661997e-05,
      "loss": 0.9828,
      "num_input_tokens_seen": 117310216,
      "step": 7291
    },
    {
      "epoch": 0.510791807857662,
      "grad_norm": 3.516911029815674,
      "learning_rate": 4.8968770577933455e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 117326600,
      "step": 7292
    },
    {
      "epoch": 0.5108618561033912,
      "grad_norm": 4.772302150726318,
      "learning_rate": 4.896177232924694e-05,
      "loss": 1.1043,
      "num_input_tokens_seen": 117342984,
      "step": 7293
    },
    {
      "epoch": 0.5109319043491205,
      "grad_norm": 3.933194875717163,
      "learning_rate": 4.8954774080560426e-05,
      "loss": 1.1596,
      "num_input_tokens_seen": 117359024,
      "step": 7294
    },
    {
      "epoch": 0.5110019525948497,
      "grad_norm": 5.703490734100342,
      "learning_rate": 4.894777583187391e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 117375184,
      "step": 7295
    },
    {
      "epoch": 0.5110720008405789,
      "grad_norm": 5.454290866851807,
      "learning_rate": 4.8940777583187396e-05,
      "loss": 1.1642,
      "num_input_tokens_seen": 117391480,
      "step": 7296
    },
    {
      "epoch": 0.5111420490863082,
      "grad_norm": 4.424625396728516,
      "learning_rate": 4.8933779334500885e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 117407864,
      "step": 7297
    },
    {
      "epoch": 0.5112120973320374,
      "grad_norm": 3.7400286197662354,
      "learning_rate": 4.892678108581436e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 117424248,
      "step": 7298
    },
    {
      "epoch": 0.5112821455777666,
      "grad_norm": 4.604375839233398,
      "learning_rate": 4.8919782837127855e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 117440632,
      "step": 7299
    },
    {
      "epoch": 0.511352193823496,
      "grad_norm": 4.090991497039795,
      "learning_rate": 4.891278458844133e-05,
      "loss": 0.9973,
      "num_input_tokens_seen": 117457016,
      "step": 7300
    },
    {
      "epoch": 0.5114222420692252,
      "grad_norm": 4.32021427154541,
      "learning_rate": 4.890578633975482e-05,
      "loss": 1.1599,
      "num_input_tokens_seen": 117473400,
      "step": 7301
    },
    {
      "epoch": 0.5114922903149545,
      "grad_norm": 5.156586170196533,
      "learning_rate": 4.88987880910683e-05,
      "loss": 1.0992,
      "num_input_tokens_seen": 117489520,
      "step": 7302
    },
    {
      "epoch": 0.5115623385606837,
      "grad_norm": 3.4068989753723145,
      "learning_rate": 4.889178984238179e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 117505904,
      "step": 7303
    },
    {
      "epoch": 0.5116323868064129,
      "grad_norm": 4.153528690338135,
      "learning_rate": 4.888479159369528e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 117522056,
      "step": 7304
    },
    {
      "epoch": 0.5117024350521422,
      "grad_norm": 4.036544322967529,
      "learning_rate": 4.887779334500876e-05,
      "loss": 0.9494,
      "num_input_tokens_seen": 117538392,
      "step": 7305
    },
    {
      "epoch": 0.5117724832978714,
      "grad_norm": 3.522869110107422,
      "learning_rate": 4.887079509632225e-05,
      "loss": 1.0601,
      "num_input_tokens_seen": 117554776,
      "step": 7306
    },
    {
      "epoch": 0.5118425315436007,
      "grad_norm": 5.0436530113220215,
      "learning_rate": 4.886379684763572e-05,
      "loss": 1.3193,
      "num_input_tokens_seen": 117571160,
      "step": 7307
    },
    {
      "epoch": 0.5119125797893299,
      "grad_norm": 3.9105610847473145,
      "learning_rate": 4.885679859894921e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 117587544,
      "step": 7308
    },
    {
      "epoch": 0.5119826280350591,
      "grad_norm": 3.6909308433532715,
      "learning_rate": 4.884980035026271e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 117603512,
      "step": 7309
    },
    {
      "epoch": 0.5120526762807884,
      "grad_norm": 4.3869404792785645,
      "learning_rate": 4.884280210157618e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 117619896,
      "step": 7310
    },
    {
      "epoch": 0.5121227245265176,
      "grad_norm": 4.207435607910156,
      "learning_rate": 4.883580385288967e-05,
      "loss": 1.2587,
      "num_input_tokens_seen": 117636280,
      "step": 7311
    },
    {
      "epoch": 0.5121927727722468,
      "grad_norm": 3.8486809730529785,
      "learning_rate": 4.882880560420315e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 117652664,
      "step": 7312
    },
    {
      "epoch": 0.5122628210179762,
      "grad_norm": 4.338857173919678,
      "learning_rate": 4.882180735551664e-05,
      "loss": 1.0159,
      "num_input_tokens_seen": 117669048,
      "step": 7313
    },
    {
      "epoch": 0.5123328692637054,
      "grad_norm": 4.424511432647705,
      "learning_rate": 4.881480910683013e-05,
      "loss": 1.1418,
      "num_input_tokens_seen": 117684464,
      "step": 7314
    },
    {
      "epoch": 0.5124029175094347,
      "grad_norm": 3.560134172439575,
      "learning_rate": 4.880781085814361e-05,
      "loss": 0.8744,
      "num_input_tokens_seen": 117700848,
      "step": 7315
    },
    {
      "epoch": 0.5124729657551639,
      "grad_norm": 3.687147378921509,
      "learning_rate": 4.88008126094571e-05,
      "loss": 1.0157,
      "num_input_tokens_seen": 117717232,
      "step": 7316
    },
    {
      "epoch": 0.5125430140008931,
      "grad_norm": 3.469698429107666,
      "learning_rate": 4.8793814360770575e-05,
      "loss": 1.0037,
      "num_input_tokens_seen": 117733616,
      "step": 7317
    },
    {
      "epoch": 0.5126130622466224,
      "grad_norm": 4.417529582977295,
      "learning_rate": 4.8786816112084064e-05,
      "loss": 1.0446,
      "num_input_tokens_seen": 117749912,
      "step": 7318
    },
    {
      "epoch": 0.5126831104923516,
      "grad_norm": 4.738529205322266,
      "learning_rate": 4.877981786339756e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 117766088,
      "step": 7319
    },
    {
      "epoch": 0.5127531587380808,
      "grad_norm": 3.9368979930877686,
      "learning_rate": 4.8772819614711034e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 117781896,
      "step": 7320
    },
    {
      "epoch": 0.5128232069838101,
      "grad_norm": 4.166690826416016,
      "learning_rate": 4.876582136602452e-05,
      "loss": 1.1633,
      "num_input_tokens_seen": 117798280,
      "step": 7321
    },
    {
      "epoch": 0.5128932552295393,
      "grad_norm": 5.667482376098633,
      "learning_rate": 4.8758823117338005e-05,
      "loss": 1.0399,
      "num_input_tokens_seen": 117812376,
      "step": 7322
    },
    {
      "epoch": 0.5129633034752686,
      "grad_norm": 4.0888776779174805,
      "learning_rate": 4.875182486865149e-05,
      "loss": 1.1515,
      "num_input_tokens_seen": 117828760,
      "step": 7323
    },
    {
      "epoch": 0.5130333517209978,
      "grad_norm": 3.7693824768066406,
      "learning_rate": 4.874482661996498e-05,
      "loss": 1.003,
      "num_input_tokens_seen": 117844312,
      "step": 7324
    },
    {
      "epoch": 0.513103399966727,
      "grad_norm": 4.041450500488281,
      "learning_rate": 4.8737828371278464e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 117860456,
      "step": 7325
    },
    {
      "epoch": 0.5131734482124564,
      "grad_norm": 3.697910785675049,
      "learning_rate": 4.873083012259195e-05,
      "loss": 1.0592,
      "num_input_tokens_seen": 117876840,
      "step": 7326
    },
    {
      "epoch": 0.5132434964581856,
      "grad_norm": 5.30276346206665,
      "learning_rate": 4.872383187390543e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 117891376,
      "step": 7327
    },
    {
      "epoch": 0.5133135447039148,
      "grad_norm": 4.165642261505127,
      "learning_rate": 4.8716833625218916e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 117907080,
      "step": 7328
    },
    {
      "epoch": 0.5133835929496441,
      "grad_norm": 3.727694272994995,
      "learning_rate": 4.87098353765324e-05,
      "loss": 1.0567,
      "num_input_tokens_seen": 117923464,
      "step": 7329
    },
    {
      "epoch": 0.5134536411953733,
      "grad_norm": 3.4425323009490967,
      "learning_rate": 4.8702837127845886e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 117939848,
      "step": 7330
    },
    {
      "epoch": 0.5135236894411026,
      "grad_norm": 5.108044147491455,
      "learning_rate": 4.8695838879159375e-05,
      "loss": 1.3818,
      "num_input_tokens_seen": 117956232,
      "step": 7331
    },
    {
      "epoch": 0.5135937376868318,
      "grad_norm": 3.598155975341797,
      "learning_rate": 4.868884063047286e-05,
      "loss": 0.8586,
      "num_input_tokens_seen": 117972616,
      "step": 7332
    },
    {
      "epoch": 0.513663785932561,
      "grad_norm": 5.529993057250977,
      "learning_rate": 4.8681842381786345e-05,
      "loss": 1.0782,
      "num_input_tokens_seen": 117988616,
      "step": 7333
    },
    {
      "epoch": 0.5137338341782903,
      "grad_norm": 3.80139422416687,
      "learning_rate": 4.867484413309982e-05,
      "loss": 0.9837,
      "num_input_tokens_seen": 118004720,
      "step": 7334
    },
    {
      "epoch": 0.5138038824240195,
      "grad_norm": 3.5874598026275635,
      "learning_rate": 4.8667845884413316e-05,
      "loss": 0.965,
      "num_input_tokens_seen": 118021008,
      "step": 7335
    },
    {
      "epoch": 0.5138739306697487,
      "grad_norm": 4.101084232330322,
      "learning_rate": 4.8660847635726804e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 118036976,
      "step": 7336
    },
    {
      "epoch": 0.513943978915478,
      "grad_norm": 4.202828407287598,
      "learning_rate": 4.865384938704028e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 118052632,
      "step": 7337
    },
    {
      "epoch": 0.5140140271612073,
      "grad_norm": 3.6700570583343506,
      "learning_rate": 4.864685113835377e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 118069016,
      "step": 7338
    },
    {
      "epoch": 0.5140840754069366,
      "grad_norm": 5.313467502593994,
      "learning_rate": 4.863985288966725e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 118085400,
      "step": 7339
    },
    {
      "epoch": 0.5141541236526658,
      "grad_norm": 3.78403639793396,
      "learning_rate": 4.863285464098074e-05,
      "loss": 1.1592,
      "num_input_tokens_seen": 118101784,
      "step": 7340
    },
    {
      "epoch": 0.514224171898395,
      "grad_norm": 5.284808158874512,
      "learning_rate": 4.862585639229423e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 118116584,
      "step": 7341
    },
    {
      "epoch": 0.5142942201441243,
      "grad_norm": 3.6075503826141357,
      "learning_rate": 4.861885814360771e-05,
      "loss": 1.0451,
      "num_input_tokens_seen": 118132968,
      "step": 7342
    },
    {
      "epoch": 0.5143642683898535,
      "grad_norm": 3.514549493789673,
      "learning_rate": 4.86118598949212e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 118148576,
      "step": 7343
    },
    {
      "epoch": 0.5144343166355828,
      "grad_norm": 4.394196510314941,
      "learning_rate": 4.860486164623467e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 118164168,
      "step": 7344
    },
    {
      "epoch": 0.514504364881312,
      "grad_norm": 4.1175737380981445,
      "learning_rate": 4.859786339754817e-05,
      "loss": 1.1392,
      "num_input_tokens_seen": 118180552,
      "step": 7345
    },
    {
      "epoch": 0.5145744131270412,
      "grad_norm": 4.6415581703186035,
      "learning_rate": 4.8590865148861656e-05,
      "loss": 1.1932,
      "num_input_tokens_seen": 118196936,
      "step": 7346
    },
    {
      "epoch": 0.5146444613727705,
      "grad_norm": 4.681972980499268,
      "learning_rate": 4.858386690017513e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 118213320,
      "step": 7347
    },
    {
      "epoch": 0.5147145096184997,
      "grad_norm": 6.289412021636963,
      "learning_rate": 4.857686865148862e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 118228648,
      "step": 7348
    },
    {
      "epoch": 0.5147845578642289,
      "grad_norm": 4.726010799407959,
      "learning_rate": 4.85698704028021e-05,
      "loss": 1.1635,
      "num_input_tokens_seen": 118244552,
      "step": 7349
    },
    {
      "epoch": 0.5148546061099583,
      "grad_norm": 3.7650058269500732,
      "learning_rate": 4.856287215411559e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 118260416,
      "step": 7350
    },
    {
      "epoch": 0.5149246543556875,
      "grad_norm": 5.283148765563965,
      "learning_rate": 4.855587390542907e-05,
      "loss": 1.1027,
      "num_input_tokens_seen": 118276800,
      "step": 7351
    },
    {
      "epoch": 0.5149947026014168,
      "grad_norm": 4.060834884643555,
      "learning_rate": 4.854887565674256e-05,
      "loss": 0.9387,
      "num_input_tokens_seen": 118293128,
      "step": 7352
    },
    {
      "epoch": 0.515064750847146,
      "grad_norm": 3.5670249462127686,
      "learning_rate": 4.854187740805605e-05,
      "loss": 0.9658,
      "num_input_tokens_seen": 118309512,
      "step": 7353
    },
    {
      "epoch": 0.5151347990928752,
      "grad_norm": 4.051480770111084,
      "learning_rate": 4.8534879159369524e-05,
      "loss": 1.1616,
      "num_input_tokens_seen": 118325640,
      "step": 7354
    },
    {
      "epoch": 0.5152048473386045,
      "grad_norm": 4.1467461585998535,
      "learning_rate": 4.852788091068302e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 118342024,
      "step": 7355
    },
    {
      "epoch": 0.5152748955843337,
      "grad_norm": 4.226484775543213,
      "learning_rate": 4.8520882661996495e-05,
      "loss": 1.3255,
      "num_input_tokens_seen": 118358408,
      "step": 7356
    },
    {
      "epoch": 0.5153449438300629,
      "grad_norm": 3.81172513961792,
      "learning_rate": 4.8513884413309983e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 118373000,
      "step": 7357
    },
    {
      "epoch": 0.5154149920757922,
      "grad_norm": 5.178154945373535,
      "learning_rate": 4.850688616462347e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 118389208,
      "step": 7358
    },
    {
      "epoch": 0.5154850403215214,
      "grad_norm": 3.971707582473755,
      "learning_rate": 4.8499887915936954e-05,
      "loss": 1.0011,
      "num_input_tokens_seen": 118405592,
      "step": 7359
    },
    {
      "epoch": 0.5155550885672507,
      "grad_norm": 3.4853761196136475,
      "learning_rate": 4.849288966725044e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 118421976,
      "step": 7360
    },
    {
      "epoch": 0.51562513681298,
      "grad_norm": 3.309765577316284,
      "learning_rate": 4.8485891418563924e-05,
      "loss": 0.9776,
      "num_input_tokens_seen": 118438232,
      "step": 7361
    },
    {
      "epoch": 0.5156951850587091,
      "grad_norm": 5.875191688537598,
      "learning_rate": 4.847889316987741e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 118453368,
      "step": 7362
    },
    {
      "epoch": 0.5157652333044385,
      "grad_norm": 3.633922815322876,
      "learning_rate": 4.84718949211909e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 118469752,
      "step": 7363
    },
    {
      "epoch": 0.5158352815501677,
      "grad_norm": 3.7608840465545654,
      "learning_rate": 4.8464896672504376e-05,
      "loss": 0.9275,
      "num_input_tokens_seen": 118485760,
      "step": 7364
    },
    {
      "epoch": 0.5159053297958969,
      "grad_norm": 4.625830173492432,
      "learning_rate": 4.845789842381787e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 118501944,
      "step": 7365
    },
    {
      "epoch": 0.5159753780416262,
      "grad_norm": 5.04378604888916,
      "learning_rate": 4.845090017513135e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 118518328,
      "step": 7366
    },
    {
      "epoch": 0.5160454262873554,
      "grad_norm": 3.658432960510254,
      "learning_rate": 4.8443901926444835e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 118534528,
      "step": 7367
    },
    {
      "epoch": 0.5161154745330847,
      "grad_norm": 4.299334526062012,
      "learning_rate": 4.8436903677758324e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 118550912,
      "step": 7368
    },
    {
      "epoch": 0.5161855227788139,
      "grad_norm": 5.608176231384277,
      "learning_rate": 4.8429905429071806e-05,
      "loss": 1.0708,
      "num_input_tokens_seen": 118566464,
      "step": 7369
    },
    {
      "epoch": 0.5162555710245431,
      "grad_norm": 4.76702356338501,
      "learning_rate": 4.8422907180385294e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 118582848,
      "step": 7370
    },
    {
      "epoch": 0.5163256192702724,
      "grad_norm": 4.01913595199585,
      "learning_rate": 4.8415908931698776e-05,
      "loss": 1.2714,
      "num_input_tokens_seen": 118599232,
      "step": 7371
    },
    {
      "epoch": 0.5163956675160016,
      "grad_norm": 4.569849967956543,
      "learning_rate": 4.8408910683012265e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 118615616,
      "step": 7372
    },
    {
      "epoch": 0.5164657157617308,
      "grad_norm": 4.713064670562744,
      "learning_rate": 4.8401912434325753e-05,
      "loss": 0.9028,
      "num_input_tokens_seen": 118632000,
      "step": 7373
    },
    {
      "epoch": 0.5165357640074602,
      "grad_norm": 4.8917317390441895,
      "learning_rate": 4.839491418563923e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 118646960,
      "step": 7374
    },
    {
      "epoch": 0.5166058122531894,
      "grad_norm": 4.5929460525512695,
      "learning_rate": 4.8387915936952724e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 118663344,
      "step": 7375
    },
    {
      "epoch": 0.5166758604989187,
      "grad_norm": 4.328620433807373,
      "learning_rate": 4.83809176882662e-05,
      "loss": 0.9591,
      "num_input_tokens_seen": 118679728,
      "step": 7376
    },
    {
      "epoch": 0.5167459087446479,
      "grad_norm": 4.046355724334717,
      "learning_rate": 4.837391943957969e-05,
      "loss": 1.1514,
      "num_input_tokens_seen": 118694912,
      "step": 7377
    },
    {
      "epoch": 0.5168159569903771,
      "grad_norm": 3.612791061401367,
      "learning_rate": 4.836692119089317e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 118711296,
      "step": 7378
    },
    {
      "epoch": 0.5168860052361064,
      "grad_norm": 3.2623980045318604,
      "learning_rate": 4.835992294220666e-05,
      "loss": 0.9072,
      "num_input_tokens_seen": 118727680,
      "step": 7379
    },
    {
      "epoch": 0.5169560534818356,
      "grad_norm": 3.7941362857818604,
      "learning_rate": 4.8352924693520147e-05,
      "loss": 0.8598,
      "num_input_tokens_seen": 118744064,
      "step": 7380
    },
    {
      "epoch": 0.5170261017275649,
      "grad_norm": 3.7529819011688232,
      "learning_rate": 4.834592644483363e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 118759616,
      "step": 7381
    },
    {
      "epoch": 0.5170961499732941,
      "grad_norm": 3.552791118621826,
      "learning_rate": 4.833892819614712e-05,
      "loss": 1.1577,
      "num_input_tokens_seen": 118776000,
      "step": 7382
    },
    {
      "epoch": 0.5171661982190233,
      "grad_norm": 3.719827651977539,
      "learning_rate": 4.833192994746059e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 118792360,
      "step": 7383
    },
    {
      "epoch": 0.5172362464647526,
      "grad_norm": 6.634743690490723,
      "learning_rate": 4.832493169877408e-05,
      "loss": 1.3676,
      "num_input_tokens_seen": 118808112,
      "step": 7384
    },
    {
      "epoch": 0.5173062947104818,
      "grad_norm": 4.388521194458008,
      "learning_rate": 4.8317933450087576e-05,
      "loss": 1.2442,
      "num_input_tokens_seen": 118824384,
      "step": 7385
    },
    {
      "epoch": 0.517376342956211,
      "grad_norm": 4.161240577697754,
      "learning_rate": 4.831093520140105e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 118840280,
      "step": 7386
    },
    {
      "epoch": 0.5174463912019404,
      "grad_norm": 4.272933006286621,
      "learning_rate": 4.830393695271454e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 118855176,
      "step": 7387
    },
    {
      "epoch": 0.5175164394476696,
      "grad_norm": 4.388091087341309,
      "learning_rate": 4.829693870402802e-05,
      "loss": 1.0048,
      "num_input_tokens_seen": 118869528,
      "step": 7388
    },
    {
      "epoch": 0.5175864876933989,
      "grad_norm": 3.623302698135376,
      "learning_rate": 4.828994045534151e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 118885272,
      "step": 7389
    },
    {
      "epoch": 0.5176565359391281,
      "grad_norm": 6.347753047943115,
      "learning_rate": 4.8282942206655e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 118901384,
      "step": 7390
    },
    {
      "epoch": 0.5177265841848573,
      "grad_norm": 5.049704551696777,
      "learning_rate": 4.827594395796848e-05,
      "loss": 1.1742,
      "num_input_tokens_seen": 118916392,
      "step": 7391
    },
    {
      "epoch": 0.5177966324305866,
      "grad_norm": 4.022581100463867,
      "learning_rate": 4.826894570928197e-05,
      "loss": 0.8257,
      "num_input_tokens_seen": 118932000,
      "step": 7392
    },
    {
      "epoch": 0.5178666806763158,
      "grad_norm": 4.277614593505859,
      "learning_rate": 4.8261947460595444e-05,
      "loss": 1.1972,
      "num_input_tokens_seen": 118947752,
      "step": 7393
    },
    {
      "epoch": 0.517936728922045,
      "grad_norm": 3.5294747352600098,
      "learning_rate": 4.825494921190893e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 118963312,
      "step": 7394
    },
    {
      "epoch": 0.5180067771677743,
      "grad_norm": 4.206392765045166,
      "learning_rate": 4.824795096322243e-05,
      "loss": 1.1942,
      "num_input_tokens_seen": 118979696,
      "step": 7395
    },
    {
      "epoch": 0.5180768254135035,
      "grad_norm": 5.907151222229004,
      "learning_rate": 4.82409527145359e-05,
      "loss": 1.1046,
      "num_input_tokens_seen": 118996080,
      "step": 7396
    },
    {
      "epoch": 0.5181468736592328,
      "grad_norm": 3.4217193126678467,
      "learning_rate": 4.823395446584939e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 119012064,
      "step": 7397
    },
    {
      "epoch": 0.518216921904962,
      "grad_norm": 3.4076719284057617,
      "learning_rate": 4.822695621716287e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 119028448,
      "step": 7398
    },
    {
      "epoch": 0.5182869701506913,
      "grad_norm": 3.6392581462860107,
      "learning_rate": 4.821995796847636e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 119044832,
      "step": 7399
    },
    {
      "epoch": 0.5183570183964206,
      "grad_norm": 4.3391947746276855,
      "learning_rate": 4.821295971978985e-05,
      "loss": 1.0365,
      "num_input_tokens_seen": 119061216,
      "step": 7400
    },
    {
      "epoch": 0.5183570183964206,
      "eval_loss": 1.1227985620498657,
      "eval_runtime": 0.1572,
      "eval_samples_per_second": 6.362,
      "eval_steps_per_second": 6.362,
      "num_input_tokens_seen": 119061216,
      "step": 7400
    },
    {
      "epoch": 0.5184270666421498,
      "grad_norm": 4.152843952178955,
      "learning_rate": 4.820596147110333e-05,
      "loss": 1.0123,
      "num_input_tokens_seen": 119077600,
      "step": 7401
    },
    {
      "epoch": 0.518497114887879,
      "grad_norm": 4.2847371101379395,
      "learning_rate": 4.819896322241682e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 119093984,
      "step": 7402
    },
    {
      "epoch": 0.5185671631336083,
      "grad_norm": 3.9743080139160156,
      "learning_rate": 4.8191964973730296e-05,
      "loss": 1.0229,
      "num_input_tokens_seen": 119109984,
      "step": 7403
    },
    {
      "epoch": 0.5186372113793375,
      "grad_norm": 3.8699886798858643,
      "learning_rate": 4.8184966725043785e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 119126040,
      "step": 7404
    },
    {
      "epoch": 0.5187072596250668,
      "grad_norm": 4.810290336608887,
      "learning_rate": 4.8177968476357266e-05,
      "loss": 1.1959,
      "num_input_tokens_seen": 119141512,
      "step": 7405
    },
    {
      "epoch": 0.518777307870796,
      "grad_norm": 5.068106174468994,
      "learning_rate": 4.8170970227670755e-05,
      "loss": 0.9215,
      "num_input_tokens_seen": 119156112,
      "step": 7406
    },
    {
      "epoch": 0.5188473561165252,
      "grad_norm": 5.044201374053955,
      "learning_rate": 4.8163971978984244e-05,
      "loss": 1.2169,
      "num_input_tokens_seen": 119172496,
      "step": 7407
    },
    {
      "epoch": 0.5189174043622545,
      "grad_norm": 5.986364364624023,
      "learning_rate": 4.8156973730297725e-05,
      "loss": 1.1855,
      "num_input_tokens_seen": 119188880,
      "step": 7408
    },
    {
      "epoch": 0.5189874526079837,
      "grad_norm": 4.029194355010986,
      "learning_rate": 4.8149975481611214e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 119205264,
      "step": 7409
    },
    {
      "epoch": 0.519057500853713,
      "grad_norm": 6.960198879241943,
      "learning_rate": 4.814297723292469e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 119220560,
      "step": 7410
    },
    {
      "epoch": 0.5191275490994423,
      "grad_norm": 3.8074264526367188,
      "learning_rate": 4.8135978984238184e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 119236856,
      "step": 7411
    },
    {
      "epoch": 0.5191975973451715,
      "grad_norm": 4.192657947540283,
      "learning_rate": 4.812898073555167e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 119253240,
      "step": 7412
    },
    {
      "epoch": 0.5192676455909008,
      "grad_norm": 4.494477272033691,
      "learning_rate": 4.812198248686515e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 119269264,
      "step": 7413
    },
    {
      "epoch": 0.51933769383663,
      "grad_norm": 3.808358907699585,
      "learning_rate": 4.8114984238178637e-05,
      "loss": 1.0039,
      "num_input_tokens_seen": 119285384,
      "step": 7414
    },
    {
      "epoch": 0.5194077420823592,
      "grad_norm": 4.097054481506348,
      "learning_rate": 4.810798598949212e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 119301768,
      "step": 7415
    },
    {
      "epoch": 0.5194777903280885,
      "grad_norm": 4.17371129989624,
      "learning_rate": 4.810098774080561e-05,
      "loss": 1.3124,
      "num_input_tokens_seen": 119318152,
      "step": 7416
    },
    {
      "epoch": 0.5195478385738177,
      "grad_norm": 3.7520711421966553,
      "learning_rate": 4.8093989492119096e-05,
      "loss": 0.9264,
      "num_input_tokens_seen": 119334536,
      "step": 7417
    },
    {
      "epoch": 0.519617886819547,
      "grad_norm": 3.5394372940063477,
      "learning_rate": 4.808699124343258e-05,
      "loss": 1.1861,
      "num_input_tokens_seen": 119350888,
      "step": 7418
    },
    {
      "epoch": 0.5196879350652762,
      "grad_norm": 3.490539312362671,
      "learning_rate": 4.8079992994746066e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 119367272,
      "step": 7419
    },
    {
      "epoch": 0.5197579833110054,
      "grad_norm": 3.7124273777008057,
      "learning_rate": 4.807299474605954e-05,
      "loss": 0.9956,
      "num_input_tokens_seen": 119383656,
      "step": 7420
    },
    {
      "epoch": 0.5198280315567347,
      "grad_norm": 4.070093631744385,
      "learning_rate": 4.8065996497373036e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 119399440,
      "step": 7421
    },
    {
      "epoch": 0.5198980798024639,
      "grad_norm": 7.456560134887695,
      "learning_rate": 4.8058998248686525e-05,
      "loss": 0.9437,
      "num_input_tokens_seen": 119415232,
      "step": 7422
    },
    {
      "epoch": 0.5199681280481931,
      "grad_norm": 3.812809705734253,
      "learning_rate": 4.8052e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 119431032,
      "step": 7423
    },
    {
      "epoch": 0.5200381762939225,
      "grad_norm": 4.641679763793945,
      "learning_rate": 4.804500175131349e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 119447416,
      "step": 7424
    },
    {
      "epoch": 0.5201082245396517,
      "grad_norm": 3.7425315380096436,
      "learning_rate": 4.803800350262697e-05,
      "loss": 0.922,
      "num_input_tokens_seen": 119463400,
      "step": 7425
    },
    {
      "epoch": 0.520178272785381,
      "grad_norm": 3.5548949241638184,
      "learning_rate": 4.803100525394046e-05,
      "loss": 0.8365,
      "num_input_tokens_seen": 119479784,
      "step": 7426
    },
    {
      "epoch": 0.5202483210311102,
      "grad_norm": 3.335888385772705,
      "learning_rate": 4.802400700525395e-05,
      "loss": 0.9923,
      "num_input_tokens_seen": 119495928,
      "step": 7427
    },
    {
      "epoch": 0.5203183692768394,
      "grad_norm": 3.8208446502685547,
      "learning_rate": 4.801700875656743e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 119511408,
      "step": 7428
    },
    {
      "epoch": 0.5203884175225687,
      "grad_norm": 4.35474967956543,
      "learning_rate": 4.801001050788092e-05,
      "loss": 1.1477,
      "num_input_tokens_seen": 119526656,
      "step": 7429
    },
    {
      "epoch": 0.5204584657682979,
      "grad_norm": 4.081501007080078,
      "learning_rate": 4.800301225919439e-05,
      "loss": 1.0881,
      "num_input_tokens_seen": 119543040,
      "step": 7430
    },
    {
      "epoch": 0.5205285140140271,
      "grad_norm": 4.1007866859436035,
      "learning_rate": 4.799601401050789e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 119559040,
      "step": 7431
    },
    {
      "epoch": 0.5205985622597564,
      "grad_norm": 3.955095052719116,
      "learning_rate": 4.7989015761821363e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 119574832,
      "step": 7432
    },
    {
      "epoch": 0.5206686105054856,
      "grad_norm": 5.475005626678467,
      "learning_rate": 4.798201751313485e-05,
      "loss": 1.1933,
      "num_input_tokens_seen": 119590688,
      "step": 7433
    },
    {
      "epoch": 0.520738658751215,
      "grad_norm": 4.125513553619385,
      "learning_rate": 4.797501926444834e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 119606592,
      "step": 7434
    },
    {
      "epoch": 0.5208087069969441,
      "grad_norm": 3.608366012573242,
      "learning_rate": 4.796802101576182e-05,
      "loss": 1.0557,
      "num_input_tokens_seen": 119622976,
      "step": 7435
    },
    {
      "epoch": 0.5208787552426734,
      "grad_norm": 5.827488899230957,
      "learning_rate": 4.796102276707531e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 119639360,
      "step": 7436
    },
    {
      "epoch": 0.5209488034884027,
      "grad_norm": 4.732996463775635,
      "learning_rate": 4.795402451838879e-05,
      "loss": 0.7811,
      "num_input_tokens_seen": 119654448,
      "step": 7437
    },
    {
      "epoch": 0.5210188517341319,
      "grad_norm": 3.9436864852905273,
      "learning_rate": 4.794702626970228e-05,
      "loss": 0.9169,
      "num_input_tokens_seen": 119670448,
      "step": 7438
    },
    {
      "epoch": 0.5210888999798611,
      "grad_norm": 3.787825345993042,
      "learning_rate": 4.794002802101577e-05,
      "loss": 1.1469,
      "num_input_tokens_seen": 119686832,
      "step": 7439
    },
    {
      "epoch": 0.5211589482255904,
      "grad_norm": 3.7498011589050293,
      "learning_rate": 4.7933029772329245e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 119702776,
      "step": 7440
    },
    {
      "epoch": 0.5212289964713196,
      "grad_norm": 4.7387285232543945,
      "learning_rate": 4.792603152364274e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 119719160,
      "step": 7441
    },
    {
      "epoch": 0.5212990447170489,
      "grad_norm": 4.916442394256592,
      "learning_rate": 4.7919033274956215e-05,
      "loss": 1.0036,
      "num_input_tokens_seen": 119735208,
      "step": 7442
    },
    {
      "epoch": 0.5213690929627781,
      "grad_norm": 7.069667339324951,
      "learning_rate": 4.7912035026269704e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 119750720,
      "step": 7443
    },
    {
      "epoch": 0.5214391412085073,
      "grad_norm": 4.178729057312012,
      "learning_rate": 4.790503677758319e-05,
      "loss": 0.946,
      "num_input_tokens_seen": 119767104,
      "step": 7444
    },
    {
      "epoch": 0.5215091894542366,
      "grad_norm": 4.8164801597595215,
      "learning_rate": 4.7898038528896674e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 119782680,
      "step": 7445
    },
    {
      "epoch": 0.5215792376999658,
      "grad_norm": 4.30715799331665,
      "learning_rate": 4.789104028021016e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 119798392,
      "step": 7446
    },
    {
      "epoch": 0.5216492859456952,
      "grad_norm": 4.270611763000488,
      "learning_rate": 4.7884042031523645e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 119814776,
      "step": 7447
    },
    {
      "epoch": 0.5217193341914244,
      "grad_norm": 3.3384652137756348,
      "learning_rate": 4.7877043782837133e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 119831160,
      "step": 7448
    },
    {
      "epoch": 0.5217893824371536,
      "grad_norm": 3.343555450439453,
      "learning_rate": 4.787004553415062e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 119847544,
      "step": 7449
    },
    {
      "epoch": 0.5218594306828829,
      "grad_norm": 4.26312255859375,
      "learning_rate": 4.78630472854641e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 119863928,
      "step": 7450
    },
    {
      "epoch": 0.5219294789286121,
      "grad_norm": 3.9083523750305176,
      "learning_rate": 4.785604903677759e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 119880312,
      "step": 7451
    },
    {
      "epoch": 0.5219995271743413,
      "grad_norm": 4.212127208709717,
      "learning_rate": 4.784905078809107e-05,
      "loss": 1.1562,
      "num_input_tokens_seen": 119895464,
      "step": 7452
    },
    {
      "epoch": 0.5220695754200706,
      "grad_norm": 6.6233038902282715,
      "learning_rate": 4.7842052539404556e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 119911848,
      "step": 7453
    },
    {
      "epoch": 0.5221396236657998,
      "grad_norm": 3.770444631576538,
      "learning_rate": 4.7835054290718045e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 119928232,
      "step": 7454
    },
    {
      "epoch": 0.5222096719115291,
      "grad_norm": 3.3227531909942627,
      "learning_rate": 4.7828056042031527e-05,
      "loss": 0.9593,
      "num_input_tokens_seen": 119944296,
      "step": 7455
    },
    {
      "epoch": 0.5222797201572583,
      "grad_norm": 4.297872066497803,
      "learning_rate": 4.7821057793345015e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 119960680,
      "step": 7456
    },
    {
      "epoch": 0.5223497684029875,
      "grad_norm": 3.6508405208587646,
      "learning_rate": 4.78140595446585e-05,
      "loss": 0.9298,
      "num_input_tokens_seen": 119977064,
      "step": 7457
    },
    {
      "epoch": 0.5224198166487168,
      "grad_norm": 3.5275723934173584,
      "learning_rate": 4.7807061295971986e-05,
      "loss": 1.0049,
      "num_input_tokens_seen": 119993448,
      "step": 7458
    },
    {
      "epoch": 0.522489864894446,
      "grad_norm": 3.983844757080078,
      "learning_rate": 4.780006304728546e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 120009832,
      "step": 7459
    },
    {
      "epoch": 0.5225599131401752,
      "grad_norm": 3.6752452850341797,
      "learning_rate": 4.779306479859895e-05,
      "loss": 1.0864,
      "num_input_tokens_seen": 120026216,
      "step": 7460
    },
    {
      "epoch": 0.5226299613859046,
      "grad_norm": 3.3501710891723633,
      "learning_rate": 4.7786066549912445e-05,
      "loss": 0.8639,
      "num_input_tokens_seen": 120042456,
      "step": 7461
    },
    {
      "epoch": 0.5227000096316338,
      "grad_norm": 3.448544979095459,
      "learning_rate": 4.777906830122592e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 120058392,
      "step": 7462
    },
    {
      "epoch": 0.5227700578773631,
      "grad_norm": 3.7892913818359375,
      "learning_rate": 4.777207005253941e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 120074776,
      "step": 7463
    },
    {
      "epoch": 0.5228401061230923,
      "grad_norm": 6.972007751464844,
      "learning_rate": 4.776507180385289e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 120089776,
      "step": 7464
    },
    {
      "epoch": 0.5229101543688215,
      "grad_norm": 3.6271073818206787,
      "learning_rate": 4.775807355516638e-05,
      "loss": 1.1703,
      "num_input_tokens_seen": 120106160,
      "step": 7465
    },
    {
      "epoch": 0.5229802026145508,
      "grad_norm": 4.735227108001709,
      "learning_rate": 4.775107530647987e-05,
      "loss": 1.2245,
      "num_input_tokens_seen": 120122144,
      "step": 7466
    },
    {
      "epoch": 0.52305025086028,
      "grad_norm": 4.848446846008301,
      "learning_rate": 4.774407705779335e-05,
      "loss": 1.0117,
      "num_input_tokens_seen": 120138528,
      "step": 7467
    },
    {
      "epoch": 0.5231202991060092,
      "grad_norm": 3.5538604259490967,
      "learning_rate": 4.773707880910684e-05,
      "loss": 1.0153,
      "num_input_tokens_seen": 120154912,
      "step": 7468
    },
    {
      "epoch": 0.5231903473517385,
      "grad_norm": 4.481129169464111,
      "learning_rate": 4.773008056042031e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 120171296,
      "step": 7469
    },
    {
      "epoch": 0.5232603955974677,
      "grad_norm": 4.581019401550293,
      "learning_rate": 4.77230823117338e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 120187680,
      "step": 7470
    },
    {
      "epoch": 0.523330443843197,
      "grad_norm": 5.147364139556885,
      "learning_rate": 4.7716084063047297e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 120204008,
      "step": 7471
    },
    {
      "epoch": 0.5234004920889263,
      "grad_norm": 3.716172218322754,
      "learning_rate": 4.770908581436077e-05,
      "loss": 0.9937,
      "num_input_tokens_seen": 120220392,
      "step": 7472
    },
    {
      "epoch": 0.5234705403346555,
      "grad_norm": 5.15359354019165,
      "learning_rate": 4.770208756567426e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 120236504,
      "step": 7473
    },
    {
      "epoch": 0.5235405885803848,
      "grad_norm": 3.9450008869171143,
      "learning_rate": 4.769508931698774e-05,
      "loss": 1.2084,
      "num_input_tokens_seen": 120252568,
      "step": 7474
    },
    {
      "epoch": 0.523610636826114,
      "grad_norm": 7.154159069061279,
      "learning_rate": 4.768809106830123e-05,
      "loss": 1.0837,
      "num_input_tokens_seen": 120268952,
      "step": 7475
    },
    {
      "epoch": 0.5236806850718432,
      "grad_norm": 3.838291883468628,
      "learning_rate": 4.768109281961472e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 120285336,
      "step": 7476
    },
    {
      "epoch": 0.5237507333175725,
      "grad_norm": 4.197968482971191,
      "learning_rate": 4.76740945709282e-05,
      "loss": 1.0913,
      "num_input_tokens_seen": 120301720,
      "step": 7477
    },
    {
      "epoch": 0.5238207815633017,
      "grad_norm": 4.732188701629639,
      "learning_rate": 4.766709632224169e-05,
      "loss": 1.2257,
      "num_input_tokens_seen": 120316968,
      "step": 7478
    },
    {
      "epoch": 0.523890829809031,
      "grad_norm": 4.164662837982178,
      "learning_rate": 4.7660098073555165e-05,
      "loss": 1.0551,
      "num_input_tokens_seen": 120333352,
      "step": 7479
    },
    {
      "epoch": 0.5239608780547602,
      "grad_norm": 3.6521334648132324,
      "learning_rate": 4.765309982486865e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 120349736,
      "step": 7480
    },
    {
      "epoch": 0.5240309263004894,
      "grad_norm": 3.6896276473999023,
      "learning_rate": 4.764610157618215e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 120366120,
      "step": 7481
    },
    {
      "epoch": 0.5241009745462187,
      "grad_norm": 4.352004051208496,
      "learning_rate": 4.7639103327495624e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 120382360,
      "step": 7482
    },
    {
      "epoch": 0.5241710227919479,
      "grad_norm": 6.433780670166016,
      "learning_rate": 4.763210507880911e-05,
      "loss": 1.1227,
      "num_input_tokens_seen": 120397904,
      "step": 7483
    },
    {
      "epoch": 0.5242410710376773,
      "grad_norm": 5.6717329025268555,
      "learning_rate": 4.7625106830122594e-05,
      "loss": 1.0687,
      "num_input_tokens_seen": 120414248,
      "step": 7484
    },
    {
      "epoch": 0.5243111192834065,
      "grad_norm": 3.4725184440612793,
      "learning_rate": 4.761810858143608e-05,
      "loss": 0.8943,
      "num_input_tokens_seen": 120430632,
      "step": 7485
    },
    {
      "epoch": 0.5243811675291357,
      "grad_norm": 3.803506374359131,
      "learning_rate": 4.761111033274956e-05,
      "loss": 1.0,
      "num_input_tokens_seen": 120447016,
      "step": 7486
    },
    {
      "epoch": 0.524451215774865,
      "grad_norm": 5.165005683898926,
      "learning_rate": 4.760411208406305e-05,
      "loss": 1.0664,
      "num_input_tokens_seen": 120463400,
      "step": 7487
    },
    {
      "epoch": 0.5245212640205942,
      "grad_norm": 6.133605480194092,
      "learning_rate": 4.759711383537654e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 120478432,
      "step": 7488
    },
    {
      "epoch": 0.5245913122663234,
      "grad_norm": 4.061281681060791,
      "learning_rate": 4.7590115586690017e-05,
      "loss": 1.1429,
      "num_input_tokens_seen": 120494816,
      "step": 7489
    },
    {
      "epoch": 0.5246613605120527,
      "grad_norm": 3.2192203998565674,
      "learning_rate": 4.7583117338003505e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 120511200,
      "step": 7490
    },
    {
      "epoch": 0.5247314087577819,
      "grad_norm": 3.718182325363159,
      "learning_rate": 4.757611908931699e-05,
      "loss": 0.9153,
      "num_input_tokens_seen": 120527584,
      "step": 7491
    },
    {
      "epoch": 0.5248014570035112,
      "grad_norm": 3.742267370223999,
      "learning_rate": 4.7569120840630476e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 120543968,
      "step": 7492
    },
    {
      "epoch": 0.5248715052492404,
      "grad_norm": 5.5869951248168945,
      "learning_rate": 4.7562122591943964e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 120559120,
      "step": 7493
    },
    {
      "epoch": 0.5249415534949696,
      "grad_norm": 3.9366302490234375,
      "learning_rate": 4.7555124343257446e-05,
      "loss": 1.1197,
      "num_input_tokens_seen": 120575504,
      "step": 7494
    },
    {
      "epoch": 0.5250116017406989,
      "grad_norm": 5.102993488311768,
      "learning_rate": 4.7548126094570935e-05,
      "loss": 1.0065,
      "num_input_tokens_seen": 120591592,
      "step": 7495
    },
    {
      "epoch": 0.5250816499864281,
      "grad_norm": 3.392009735107422,
      "learning_rate": 4.754112784588441e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 120607976,
      "step": 7496
    },
    {
      "epoch": 0.5251516982321573,
      "grad_norm": 5.089282512664795,
      "learning_rate": 4.7534129597197905e-05,
      "loss": 0.9714,
      "num_input_tokens_seen": 120624360,
      "step": 7497
    },
    {
      "epoch": 0.5252217464778867,
      "grad_norm": 4.311940670013428,
      "learning_rate": 4.7527131348511394e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 120640744,
      "step": 7498
    },
    {
      "epoch": 0.5252917947236159,
      "grad_norm": 3.8907923698425293,
      "learning_rate": 4.752013309982487e-05,
      "loss": 1.1852,
      "num_input_tokens_seen": 120656992,
      "step": 7499
    },
    {
      "epoch": 0.5253618429693452,
      "grad_norm": 3.856172800064087,
      "learning_rate": 4.751313485113836e-05,
      "loss": 1.2247,
      "num_input_tokens_seen": 120673376,
      "step": 7500
    },
    {
      "epoch": 0.5254318912150744,
      "grad_norm": 3.661641836166382,
      "learning_rate": 4.750613660245184e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 120689760,
      "step": 7501
    },
    {
      "epoch": 0.5255019394608036,
      "grad_norm": 3.8533976078033447,
      "learning_rate": 4.749913835376533e-05,
      "loss": 0.9485,
      "num_input_tokens_seen": 120706144,
      "step": 7502
    },
    {
      "epoch": 0.5255719877065329,
      "grad_norm": 4.209764003753662,
      "learning_rate": 4.7492140105078816e-05,
      "loss": 1.0623,
      "num_input_tokens_seen": 120722528,
      "step": 7503
    },
    {
      "epoch": 0.5256420359522621,
      "grad_norm": 4.190296649932861,
      "learning_rate": 4.74851418563923e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 120738912,
      "step": 7504
    },
    {
      "epoch": 0.5257120841979913,
      "grad_norm": 3.971188545227051,
      "learning_rate": 4.747814360770579e-05,
      "loss": 0.8885,
      "num_input_tokens_seen": 120754464,
      "step": 7505
    },
    {
      "epoch": 0.5257821324437206,
      "grad_norm": 3.8005099296569824,
      "learning_rate": 4.747114535901926e-05,
      "loss": 0.8758,
      "num_input_tokens_seen": 120770848,
      "step": 7506
    },
    {
      "epoch": 0.5258521806894498,
      "grad_norm": 4.144433498382568,
      "learning_rate": 4.746414711033276e-05,
      "loss": 1.2437,
      "num_input_tokens_seen": 120787232,
      "step": 7507
    },
    {
      "epoch": 0.5259222289351791,
      "grad_norm": 4.963449001312256,
      "learning_rate": 4.7457148861646246e-05,
      "loss": 1.2806,
      "num_input_tokens_seen": 120802792,
      "step": 7508
    },
    {
      "epoch": 0.5259922771809084,
      "grad_norm": 6.420035362243652,
      "learning_rate": 4.745015061295972e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 120819176,
      "step": 7509
    },
    {
      "epoch": 0.5260623254266376,
      "grad_norm": 3.493263006210327,
      "learning_rate": 4.744315236427321e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 120835560,
      "step": 7510
    },
    {
      "epoch": 0.5261323736723669,
      "grad_norm": 5.055440425872803,
      "learning_rate": 4.743615411558669e-05,
      "loss": 0.9479,
      "num_input_tokens_seen": 120851752,
      "step": 7511
    },
    {
      "epoch": 0.5262024219180961,
      "grad_norm": 3.632171869277954,
      "learning_rate": 4.742915586690018e-05,
      "loss": 1.0991,
      "num_input_tokens_seen": 120867568,
      "step": 7512
    },
    {
      "epoch": 0.5262724701638254,
      "grad_norm": 4.197231769561768,
      "learning_rate": 4.742215761821366e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 120883952,
      "step": 7513
    },
    {
      "epoch": 0.5263425184095546,
      "grad_norm": 5.476736068725586,
      "learning_rate": 4.741515936952715e-05,
      "loss": 0.9664,
      "num_input_tokens_seen": 120899584,
      "step": 7514
    },
    {
      "epoch": 0.5264125666552838,
      "grad_norm": 3.98996901512146,
      "learning_rate": 4.740816112084064e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 120915968,
      "step": 7515
    },
    {
      "epoch": 0.5264826149010131,
      "grad_norm": 3.881070375442505,
      "learning_rate": 4.7401162872154114e-05,
      "loss": 1.3225,
      "num_input_tokens_seen": 120932352,
      "step": 7516
    },
    {
      "epoch": 0.5265526631467423,
      "grad_norm": 4.582496643066406,
      "learning_rate": 4.739416462346761e-05,
      "loss": 1.1441,
      "num_input_tokens_seen": 120948736,
      "step": 7517
    },
    {
      "epoch": 0.5266227113924715,
      "grad_norm": 3.413455009460449,
      "learning_rate": 4.7387166374781084e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 120964928,
      "step": 7518
    },
    {
      "epoch": 0.5266927596382008,
      "grad_norm": 3.806334972381592,
      "learning_rate": 4.738016812609457e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 120981312,
      "step": 7519
    },
    {
      "epoch": 0.52676280788393,
      "grad_norm": 4.136782169342041,
      "learning_rate": 4.737316987740806e-05,
      "loss": 1.3558,
      "num_input_tokens_seen": 120997696,
      "step": 7520
    },
    {
      "epoch": 0.5268328561296594,
      "grad_norm": 4.850030899047852,
      "learning_rate": 4.736617162872154e-05,
      "loss": 0.9764,
      "num_input_tokens_seen": 121013624,
      "step": 7521
    },
    {
      "epoch": 0.5269029043753886,
      "grad_norm": 3.9870705604553223,
      "learning_rate": 4.735917338003503e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 121030008,
      "step": 7522
    },
    {
      "epoch": 0.5269729526211178,
      "grad_norm": 3.778407573699951,
      "learning_rate": 4.7352175131348513e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 121046168,
      "step": 7523
    },
    {
      "epoch": 0.5270430008668471,
      "grad_norm": 4.619656085968018,
      "learning_rate": 4.7345176882662e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 121061544,
      "step": 7524
    },
    {
      "epoch": 0.5271130491125763,
      "grad_norm": 3.794745683670044,
      "learning_rate": 4.733817863397549e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 121077928,
      "step": 7525
    },
    {
      "epoch": 0.5271830973583055,
      "grad_norm": 5.526957988739014,
      "learning_rate": 4.7331180385288966e-05,
      "loss": 0.9366,
      "num_input_tokens_seen": 121094312,
      "step": 7526
    },
    {
      "epoch": 0.5272531456040348,
      "grad_norm": 4.695877552032471,
      "learning_rate": 4.732418213660246e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 121110696,
      "step": 7527
    },
    {
      "epoch": 0.527323193849764,
      "grad_norm": 4.297723293304443,
      "learning_rate": 4.7317183887915936e-05,
      "loss": 0.9783,
      "num_input_tokens_seen": 121126832,
      "step": 7528
    },
    {
      "epoch": 0.5273932420954933,
      "grad_norm": 3.9494822025299072,
      "learning_rate": 4.7310185639229425e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 121143192,
      "step": 7529
    },
    {
      "epoch": 0.5274632903412225,
      "grad_norm": 3.572096347808838,
      "learning_rate": 4.730318739054291e-05,
      "loss": 0.9569,
      "num_input_tokens_seen": 121159552,
      "step": 7530
    },
    {
      "epoch": 0.5275333385869517,
      "grad_norm": 4.53004789352417,
      "learning_rate": 4.7296189141856395e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 121175744,
      "step": 7531
    },
    {
      "epoch": 0.527603386832681,
      "grad_norm": 4.6228203773498535,
      "learning_rate": 4.7289190893169884e-05,
      "loss": 1.0224,
      "num_input_tokens_seen": 121190304,
      "step": 7532
    },
    {
      "epoch": 0.5276734350784102,
      "grad_norm": 3.7169201374053955,
      "learning_rate": 4.7282192644483366e-05,
      "loss": 1.0556,
      "num_input_tokens_seen": 121206688,
      "step": 7533
    },
    {
      "epoch": 0.5277434833241395,
      "grad_norm": 4.113621234893799,
      "learning_rate": 4.7275194395796854e-05,
      "loss": 0.8171,
      "num_input_tokens_seen": 121223072,
      "step": 7534
    },
    {
      "epoch": 0.5278135315698688,
      "grad_norm": 3.70991849899292,
      "learning_rate": 4.726819614711034e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 121239456,
      "step": 7535
    },
    {
      "epoch": 0.527883579815598,
      "grad_norm": 4.051577091217041,
      "learning_rate": 4.726119789842382e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 121255840,
      "step": 7536
    },
    {
      "epoch": 0.5279536280613273,
      "grad_norm": 5.249835968017578,
      "learning_rate": 4.725419964973731e-05,
      "loss": 1.1282,
      "num_input_tokens_seen": 121271664,
      "step": 7537
    },
    {
      "epoch": 0.5280236763070565,
      "grad_norm": 3.727388381958008,
      "learning_rate": 4.724720140105079e-05,
      "loss": 1.1146,
      "num_input_tokens_seen": 121288048,
      "step": 7538
    },
    {
      "epoch": 0.5280937245527857,
      "grad_norm": 3.91412615776062,
      "learning_rate": 4.724020315236428e-05,
      "loss": 1.002,
      "num_input_tokens_seen": 121304432,
      "step": 7539
    },
    {
      "epoch": 0.528163772798515,
      "grad_norm": 4.474177360534668,
      "learning_rate": 4.723320490367776e-05,
      "loss": 1.1398,
      "num_input_tokens_seen": 121320816,
      "step": 7540
    },
    {
      "epoch": 0.5282338210442442,
      "grad_norm": 3.860063314437866,
      "learning_rate": 4.722620665499125e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 121337200,
      "step": 7541
    },
    {
      "epoch": 0.5283038692899734,
      "grad_norm": 3.4258196353912354,
      "learning_rate": 4.7219208406304736e-05,
      "loss": 0.8809,
      "num_input_tokens_seen": 121353584,
      "step": 7542
    },
    {
      "epoch": 0.5283739175357027,
      "grad_norm": 3.8684587478637695,
      "learning_rate": 4.721221015761822e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 121369968,
      "step": 7543
    },
    {
      "epoch": 0.5284439657814319,
      "grad_norm": 3.713254451751709,
      "learning_rate": 4.7205211908931706e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 121386352,
      "step": 7544
    },
    {
      "epoch": 0.5285140140271612,
      "grad_norm": 4.29981803894043,
      "learning_rate": 4.719821366024518e-05,
      "loss": 0.9864,
      "num_input_tokens_seen": 121402632,
      "step": 7545
    },
    {
      "epoch": 0.5285840622728905,
      "grad_norm": 3.9872469902038574,
      "learning_rate": 4.719121541155867e-05,
      "loss": 1.1266,
      "num_input_tokens_seen": 121419016,
      "step": 7546
    },
    {
      "epoch": 0.5286541105186197,
      "grad_norm": 5.582516193389893,
      "learning_rate": 4.7184217162872165e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 121434168,
      "step": 7547
    },
    {
      "epoch": 0.528724158764349,
      "grad_norm": 3.623431921005249,
      "learning_rate": 4.717721891418564e-05,
      "loss": 1.021,
      "num_input_tokens_seen": 121450552,
      "step": 7548
    },
    {
      "epoch": 0.5287942070100782,
      "grad_norm": 4.299187660217285,
      "learning_rate": 4.717022066549913e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 121466488,
      "step": 7549
    },
    {
      "epoch": 0.5288642552558075,
      "grad_norm": 3.768704652786255,
      "learning_rate": 4.716322241681261e-05,
      "loss": 1.3006,
      "num_input_tokens_seen": 121482872,
      "step": 7550
    },
    {
      "epoch": 0.5289343035015367,
      "grad_norm": 4.494194507598877,
      "learning_rate": 4.71562241681261e-05,
      "loss": 1.2086,
      "num_input_tokens_seen": 121499256,
      "step": 7551
    },
    {
      "epoch": 0.5290043517472659,
      "grad_norm": 3.676561117172241,
      "learning_rate": 4.714922591943959e-05,
      "loss": 0.9644,
      "num_input_tokens_seen": 121515272,
      "step": 7552
    },
    {
      "epoch": 0.5290743999929952,
      "grad_norm": 5.0016961097717285,
      "learning_rate": 4.714222767075307e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 121531544,
      "step": 7553
    },
    {
      "epoch": 0.5291444482387244,
      "grad_norm": 3.8496031761169434,
      "learning_rate": 4.713522942206656e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 121547928,
      "step": 7554
    },
    {
      "epoch": 0.5292144964844536,
      "grad_norm": 5.569375514984131,
      "learning_rate": 4.712823117338003e-05,
      "loss": 1.227,
      "num_input_tokens_seen": 121564312,
      "step": 7555
    },
    {
      "epoch": 0.5292845447301829,
      "grad_norm": 3.6076838970184326,
      "learning_rate": 4.712123292469352e-05,
      "loss": 1.0773,
      "num_input_tokens_seen": 121580688,
      "step": 7556
    },
    {
      "epoch": 0.5293545929759121,
      "grad_norm": 3.5435140132904053,
      "learning_rate": 4.711423467600701e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 121596496,
      "step": 7557
    },
    {
      "epoch": 0.5294246412216415,
      "grad_norm": 3.233835458755493,
      "learning_rate": 4.710723642732049e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 121612880,
      "step": 7558
    },
    {
      "epoch": 0.5294946894673707,
      "grad_norm": 5.496852874755859,
      "learning_rate": 4.710023817863398e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 121629264,
      "step": 7559
    },
    {
      "epoch": 0.5295647377130999,
      "grad_norm": 5.445659637451172,
      "learning_rate": 4.709323992994746e-05,
      "loss": 1.0318,
      "num_input_tokens_seen": 121645128,
      "step": 7560
    },
    {
      "epoch": 0.5296347859588292,
      "grad_norm": 3.8428354263305664,
      "learning_rate": 4.708624168126095e-05,
      "loss": 1.0299,
      "num_input_tokens_seen": 121661504,
      "step": 7561
    },
    {
      "epoch": 0.5297048342045584,
      "grad_norm": 3.609997272491455,
      "learning_rate": 4.707924343257444e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 121677888,
      "step": 7562
    },
    {
      "epoch": 0.5297748824502876,
      "grad_norm": 4.154750823974609,
      "learning_rate": 4.707224518388792e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 121693328,
      "step": 7563
    },
    {
      "epoch": 0.5298449306960169,
      "grad_norm": 4.595134735107422,
      "learning_rate": 4.706524693520141e-05,
      "loss": 0.9786,
      "num_input_tokens_seen": 121709512,
      "step": 7564
    },
    {
      "epoch": 0.5299149789417461,
      "grad_norm": 3.6196346282958984,
      "learning_rate": 4.7058248686514885e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 121725272,
      "step": 7565
    },
    {
      "epoch": 0.5299850271874754,
      "grad_norm": 3.8893964290618896,
      "learning_rate": 4.7051250437828374e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 121741656,
      "step": 7566
    },
    {
      "epoch": 0.5300550754332046,
      "grad_norm": 3.8492813110351562,
      "learning_rate": 4.7044252189141856e-05,
      "loss": 1.0309,
      "num_input_tokens_seen": 121758040,
      "step": 7567
    },
    {
      "epoch": 0.5301251236789338,
      "grad_norm": 3.920822858810425,
      "learning_rate": 4.7037253940455344e-05,
      "loss": 1.0483,
      "num_input_tokens_seen": 121773736,
      "step": 7568
    },
    {
      "epoch": 0.5301951719246631,
      "grad_norm": 5.269485950469971,
      "learning_rate": 4.703025569176883e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 121790120,
      "step": 7569
    },
    {
      "epoch": 0.5302652201703923,
      "grad_norm": 4.920991897583008,
      "learning_rate": 4.7023257443082315e-05,
      "loss": 0.8157,
      "num_input_tokens_seen": 121806504,
      "step": 7570
    },
    {
      "epoch": 0.5303352684161216,
      "grad_norm": 3.8457534313201904,
      "learning_rate": 4.70162591943958e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 121822256,
      "step": 7571
    },
    {
      "epoch": 0.5304053166618509,
      "grad_norm": 3.80702543258667,
      "learning_rate": 4.700926094570928e-05,
      "loss": 1.1466,
      "num_input_tokens_seen": 121838640,
      "step": 7572
    },
    {
      "epoch": 0.5304753649075801,
      "grad_norm": 3.8683180809020996,
      "learning_rate": 4.7002262697022774e-05,
      "loss": 0.8946,
      "num_input_tokens_seen": 121854704,
      "step": 7573
    },
    {
      "epoch": 0.5305454131533094,
      "grad_norm": 6.011785507202148,
      "learning_rate": 4.699526444833626e-05,
      "loss": 1.0269,
      "num_input_tokens_seen": 121871088,
      "step": 7574
    },
    {
      "epoch": 0.5306154613990386,
      "grad_norm": 4.970396995544434,
      "learning_rate": 4.698826619964974e-05,
      "loss": 0.9285,
      "num_input_tokens_seen": 121887240,
      "step": 7575
    },
    {
      "epoch": 0.5306855096447678,
      "grad_norm": 4.267600059509277,
      "learning_rate": 4.6981267950963226e-05,
      "loss": 1.2008,
      "num_input_tokens_seen": 121903624,
      "step": 7576
    },
    {
      "epoch": 0.5307555578904971,
      "grad_norm": 4.618432998657227,
      "learning_rate": 4.697426970227671e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 121919656,
      "step": 7577
    },
    {
      "epoch": 0.5308256061362263,
      "grad_norm": 3.775972366333008,
      "learning_rate": 4.6967271453590196e-05,
      "loss": 0.9733,
      "num_input_tokens_seen": 121935992,
      "step": 7578
    },
    {
      "epoch": 0.5308956543819555,
      "grad_norm": 4.6172356605529785,
      "learning_rate": 4.6960273204903685e-05,
      "loss": 1.1194,
      "num_input_tokens_seen": 121952376,
      "step": 7579
    },
    {
      "epoch": 0.5309657026276848,
      "grad_norm": 4.867498874664307,
      "learning_rate": 4.695327495621717e-05,
      "loss": 1.1188,
      "num_input_tokens_seen": 121968760,
      "step": 7580
    },
    {
      "epoch": 0.531035750873414,
      "grad_norm": 4.13311767578125,
      "learning_rate": 4.6946276707530655e-05,
      "loss": 1.0245,
      "num_input_tokens_seen": 121984304,
      "step": 7581
    },
    {
      "epoch": 0.5311057991191434,
      "grad_norm": 3.957585573196411,
      "learning_rate": 4.693927845884413e-05,
      "loss": 0.8689,
      "num_input_tokens_seen": 122000656,
      "step": 7582
    },
    {
      "epoch": 0.5311758473648726,
      "grad_norm": 4.368579387664795,
      "learning_rate": 4.693228021015762e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 122016576,
      "step": 7583
    },
    {
      "epoch": 0.5312458956106018,
      "grad_norm": 3.653543710708618,
      "learning_rate": 4.6925281961471114e-05,
      "loss": 1.034,
      "num_input_tokens_seen": 122032784,
      "step": 7584
    },
    {
      "epoch": 0.5313159438563311,
      "grad_norm": 3.5882821083068848,
      "learning_rate": 4.691828371278459e-05,
      "loss": 0.8917,
      "num_input_tokens_seen": 122049120,
      "step": 7585
    },
    {
      "epoch": 0.5313859921020603,
      "grad_norm": 3.252802848815918,
      "learning_rate": 4.691128546409808e-05,
      "loss": 0.9816,
      "num_input_tokens_seen": 122065504,
      "step": 7586
    },
    {
      "epoch": 0.5314560403477896,
      "grad_norm": 3.5019781589508057,
      "learning_rate": 4.690428721541156e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 122080800,
      "step": 7587
    },
    {
      "epoch": 0.5315260885935188,
      "grad_norm": 3.470921277999878,
      "learning_rate": 4.689728896672505e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 122097184,
      "step": 7588
    },
    {
      "epoch": 0.531596136839248,
      "grad_norm": 3.5450100898742676,
      "learning_rate": 4.689029071803854e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 122113568,
      "step": 7589
    },
    {
      "epoch": 0.5316661850849773,
      "grad_norm": 4.1188578605651855,
      "learning_rate": 4.688329246935202e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 122129800,
      "step": 7590
    },
    {
      "epoch": 0.5317362333307065,
      "grad_norm": 4.097812175750732,
      "learning_rate": 4.687629422066551e-05,
      "loss": 1.0745,
      "num_input_tokens_seen": 122145784,
      "step": 7591
    },
    {
      "epoch": 0.5318062815764357,
      "grad_norm": 3.929668664932251,
      "learning_rate": 4.686929597197898e-05,
      "loss": 1.0934,
      "num_input_tokens_seen": 122162168,
      "step": 7592
    },
    {
      "epoch": 0.531876329822165,
      "grad_norm": 5.706707954406738,
      "learning_rate": 4.686229772329247e-05,
      "loss": 1.007,
      "num_input_tokens_seen": 122178552,
      "step": 7593
    },
    {
      "epoch": 0.5319463780678942,
      "grad_norm": 4.066921234130859,
      "learning_rate": 4.685529947460595e-05,
      "loss": 0.9495,
      "num_input_tokens_seen": 122194936,
      "step": 7594
    },
    {
      "epoch": 0.5320164263136236,
      "grad_norm": 3.8690404891967773,
      "learning_rate": 4.684830122591944e-05,
      "loss": 1.2008,
      "num_input_tokens_seen": 122211184,
      "step": 7595
    },
    {
      "epoch": 0.5320864745593528,
      "grad_norm": 5.118635654449463,
      "learning_rate": 4.684130297723293e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 122227152,
      "step": 7596
    },
    {
      "epoch": 0.532156522805082,
      "grad_norm": 3.9654159545898438,
      "learning_rate": 4.683430472854641e-05,
      "loss": 1.1422,
      "num_input_tokens_seen": 122242952,
      "step": 7597
    },
    {
      "epoch": 0.5322265710508113,
      "grad_norm": 5.006682872772217,
      "learning_rate": 4.68273064798599e-05,
      "loss": 1.2078,
      "num_input_tokens_seen": 122259104,
      "step": 7598
    },
    {
      "epoch": 0.5322966192965405,
      "grad_norm": 4.747622013092041,
      "learning_rate": 4.6820308231173375e-05,
      "loss": 1.3191,
      "num_input_tokens_seen": 122274320,
      "step": 7599
    },
    {
      "epoch": 0.5323666675422697,
      "grad_norm": 4.334118843078613,
      "learning_rate": 4.681330998248687e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 122290376,
      "step": 7600
    },
    {
      "epoch": 0.5323666675422697,
      "eval_loss": 1.120592474937439,
      "eval_runtime": 0.4292,
      "eval_samples_per_second": 2.33,
      "eval_steps_per_second": 2.33,
      "num_input_tokens_seen": 122290376,
      "step": 7600
    },
    {
      "epoch": 0.532436715787999,
      "grad_norm": 3.4493050575256348,
      "learning_rate": 4.680631173380036e-05,
      "loss": 0.9147,
      "num_input_tokens_seen": 122306760,
      "step": 7601
    },
    {
      "epoch": 0.5325067640337282,
      "grad_norm": 3.933213949203491,
      "learning_rate": 4.6799313485113834e-05,
      "loss": 1.2942,
      "num_input_tokens_seen": 122322784,
      "step": 7602
    },
    {
      "epoch": 0.5325768122794575,
      "grad_norm": 3.874788284301758,
      "learning_rate": 4.679231523642732e-05,
      "loss": 1.2438,
      "num_input_tokens_seen": 122339168,
      "step": 7603
    },
    {
      "epoch": 0.5326468605251867,
      "grad_norm": 4.443728446960449,
      "learning_rate": 4.6785316987740805e-05,
      "loss": 1.138,
      "num_input_tokens_seen": 122355552,
      "step": 7604
    },
    {
      "epoch": 0.5327169087709159,
      "grad_norm": 3.7982730865478516,
      "learning_rate": 4.677831873905429e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 122371936,
      "step": 7605
    },
    {
      "epoch": 0.5327869570166452,
      "grad_norm": 3.4133870601654053,
      "learning_rate": 4.677132049036778e-05,
      "loss": 1.0804,
      "num_input_tokens_seen": 122388192,
      "step": 7606
    },
    {
      "epoch": 0.5328570052623745,
      "grad_norm": 5.178568363189697,
      "learning_rate": 4.6764322241681264e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 122404576,
      "step": 7607
    },
    {
      "epoch": 0.5329270535081037,
      "grad_norm": 3.9049737453460693,
      "learning_rate": 4.675732399299475e-05,
      "loss": 0.9998,
      "num_input_tokens_seen": 122420960,
      "step": 7608
    },
    {
      "epoch": 0.532997101753833,
      "grad_norm": 3.8209729194641113,
      "learning_rate": 4.675032574430823e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 122436808,
      "step": 7609
    },
    {
      "epoch": 0.5330671499995622,
      "grad_norm": 5.2931389808654785,
      "learning_rate": 4.674332749562172e-05,
      "loss": 1.225,
      "num_input_tokens_seen": 122453192,
      "step": 7610
    },
    {
      "epoch": 0.5331371982452915,
      "grad_norm": 3.608839273452759,
      "learning_rate": 4.673632924693521e-05,
      "loss": 1.0053,
      "num_input_tokens_seen": 122469576,
      "step": 7611
    },
    {
      "epoch": 0.5332072464910207,
      "grad_norm": 3.75544810295105,
      "learning_rate": 4.6729330998248686e-05,
      "loss": 1.0932,
      "num_input_tokens_seen": 122485960,
      "step": 7612
    },
    {
      "epoch": 0.5332772947367499,
      "grad_norm": 4.498108863830566,
      "learning_rate": 4.6722332749562175e-05,
      "loss": 1.1613,
      "num_input_tokens_seen": 122502344,
      "step": 7613
    },
    {
      "epoch": 0.5333473429824792,
      "grad_norm": 7.5673909187316895,
      "learning_rate": 4.671533450087566e-05,
      "loss": 1.1696,
      "num_input_tokens_seen": 122518728,
      "step": 7614
    },
    {
      "epoch": 0.5334173912282084,
      "grad_norm": 6.213915824890137,
      "learning_rate": 4.6708336252189145e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 122534992,
      "step": 7615
    },
    {
      "epoch": 0.5334874394739376,
      "grad_norm": 3.739473342895508,
      "learning_rate": 4.6701338003502634e-05,
      "loss": 1.0961,
      "num_input_tokens_seen": 122551376,
      "step": 7616
    },
    {
      "epoch": 0.5335574877196669,
      "grad_norm": 4.1125617027282715,
      "learning_rate": 4.6694339754816116e-05,
      "loss": 1.0056,
      "num_input_tokens_seen": 122567760,
      "step": 7617
    },
    {
      "epoch": 0.5336275359653961,
      "grad_norm": 3.5337769985198975,
      "learning_rate": 4.6687341506129604e-05,
      "loss": 0.9145,
      "num_input_tokens_seen": 122583992,
      "step": 7618
    },
    {
      "epoch": 0.5336975842111255,
      "grad_norm": 3.6304452419281006,
      "learning_rate": 4.668034325744308e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 122600376,
      "step": 7619
    },
    {
      "epoch": 0.5337676324568547,
      "grad_norm": 4.323266983032227,
      "learning_rate": 4.6673345008756575e-05,
      "loss": 1.1597,
      "num_input_tokens_seen": 122616760,
      "step": 7620
    },
    {
      "epoch": 0.5338376807025839,
      "grad_norm": 3.598428249359131,
      "learning_rate": 4.666634676007005e-05,
      "loss": 1.0098,
      "num_input_tokens_seen": 122633144,
      "step": 7621
    },
    {
      "epoch": 0.5339077289483132,
      "grad_norm": 4.279503345489502,
      "learning_rate": 4.665934851138354e-05,
      "loss": 1.1214,
      "num_input_tokens_seen": 122649528,
      "step": 7622
    },
    {
      "epoch": 0.5339777771940424,
      "grad_norm": 5.056297302246094,
      "learning_rate": 4.665235026269703e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 122665912,
      "step": 7623
    },
    {
      "epoch": 0.5340478254397717,
      "grad_norm": 7.37079381942749,
      "learning_rate": 4.664535201401051e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 122681968,
      "step": 7624
    },
    {
      "epoch": 0.5341178736855009,
      "grad_norm": 4.478328227996826,
      "learning_rate": 4.6638353765324e-05,
      "loss": 1.2127,
      "num_input_tokens_seen": 122698352,
      "step": 7625
    },
    {
      "epoch": 0.5341879219312301,
      "grad_norm": 5.164111614227295,
      "learning_rate": 4.663135551663748e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 122714736,
      "step": 7626
    },
    {
      "epoch": 0.5342579701769594,
      "grad_norm": 6.489926815032959,
      "learning_rate": 4.662435726795097e-05,
      "loss": 0.95,
      "num_input_tokens_seen": 122731120,
      "step": 7627
    },
    {
      "epoch": 0.5343280184226886,
      "grad_norm": 5.209092140197754,
      "learning_rate": 4.6617359019264456e-05,
      "loss": 1.117,
      "num_input_tokens_seen": 122747504,
      "step": 7628
    },
    {
      "epoch": 0.5343980666684178,
      "grad_norm": 4.007065773010254,
      "learning_rate": 4.661036077057793e-05,
      "loss": 0.9271,
      "num_input_tokens_seen": 122763864,
      "step": 7629
    },
    {
      "epoch": 0.5344681149141471,
      "grad_norm": 4.077016830444336,
      "learning_rate": 4.660336252189143e-05,
      "loss": 1.1052,
      "num_input_tokens_seen": 122779808,
      "step": 7630
    },
    {
      "epoch": 0.5345381631598763,
      "grad_norm": 3.764261245727539,
      "learning_rate": 4.65963642732049e-05,
      "loss": 1.1111,
      "num_input_tokens_seen": 122796192,
      "step": 7631
    },
    {
      "epoch": 0.5346082114056057,
      "grad_norm": 3.8028204441070557,
      "learning_rate": 4.658936602451839e-05,
      "loss": 0.9435,
      "num_input_tokens_seen": 122812576,
      "step": 7632
    },
    {
      "epoch": 0.5346782596513349,
      "grad_norm": 4.397709846496582,
      "learning_rate": 4.658236777583188e-05,
      "loss": 0.9155,
      "num_input_tokens_seen": 122828048,
      "step": 7633
    },
    {
      "epoch": 0.5347483078970641,
      "grad_norm": 4.138652324676514,
      "learning_rate": 4.657536952714536e-05,
      "loss": 0.9613,
      "num_input_tokens_seen": 122842976,
      "step": 7634
    },
    {
      "epoch": 0.5348183561427934,
      "grad_norm": 3.9423744678497314,
      "learning_rate": 4.656837127845885e-05,
      "loss": 1.0464,
      "num_input_tokens_seen": 122859360,
      "step": 7635
    },
    {
      "epoch": 0.5348884043885226,
      "grad_norm": 4.257975101470947,
      "learning_rate": 4.656137302977233e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 122874888,
      "step": 7636
    },
    {
      "epoch": 0.5349584526342518,
      "grad_norm": 4.364237308502197,
      "learning_rate": 4.655437478108582e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 122891272,
      "step": 7637
    },
    {
      "epoch": 0.5350285008799811,
      "grad_norm": 4.238529205322266,
      "learning_rate": 4.654737653239931e-05,
      "loss": 1.2226,
      "num_input_tokens_seen": 122906408,
      "step": 7638
    },
    {
      "epoch": 0.5350985491257103,
      "grad_norm": 3.609213352203369,
      "learning_rate": 4.654037828371278e-05,
      "loss": 0.9925,
      "num_input_tokens_seen": 122922792,
      "step": 7639
    },
    {
      "epoch": 0.5351685973714396,
      "grad_norm": 3.5299439430236816,
      "learning_rate": 4.653338003502628e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 122939120,
      "step": 7640
    },
    {
      "epoch": 0.5352386456171688,
      "grad_norm": 3.879683494567871,
      "learning_rate": 4.6526381786339754e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 122955504,
      "step": 7641
    },
    {
      "epoch": 0.535308693862898,
      "grad_norm": 3.414780616760254,
      "learning_rate": 4.651938353765324e-05,
      "loss": 1.0047,
      "num_input_tokens_seen": 122971848,
      "step": 7642
    },
    {
      "epoch": 0.5353787421086273,
      "grad_norm": 3.307396173477173,
      "learning_rate": 4.651238528896673e-05,
      "loss": 0.8771,
      "num_input_tokens_seen": 122987840,
      "step": 7643
    },
    {
      "epoch": 0.5354487903543566,
      "grad_norm": 3.41166353225708,
      "learning_rate": 4.650538704028021e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 123004224,
      "step": 7644
    },
    {
      "epoch": 0.5355188386000858,
      "grad_norm": 4.273513317108154,
      "learning_rate": 4.64983887915937e-05,
      "loss": 1.2176,
      "num_input_tokens_seen": 123020608,
      "step": 7645
    },
    {
      "epoch": 0.5355888868458151,
      "grad_norm": 6.728183269500732,
      "learning_rate": 4.649139054290718e-05,
      "loss": 1.0303,
      "num_input_tokens_seen": 123035928,
      "step": 7646
    },
    {
      "epoch": 0.5356589350915443,
      "grad_norm": 3.6993179321289062,
      "learning_rate": 4.648439229422067e-05,
      "loss": 0.9471,
      "num_input_tokens_seen": 123051784,
      "step": 7647
    },
    {
      "epoch": 0.5357289833372736,
      "grad_norm": 4.173643112182617,
      "learning_rate": 4.647739404553415e-05,
      "loss": 1.0996,
      "num_input_tokens_seen": 123068080,
      "step": 7648
    },
    {
      "epoch": 0.5357990315830028,
      "grad_norm": 4.235645294189453,
      "learning_rate": 4.6470395796847635e-05,
      "loss": 1.0112,
      "num_input_tokens_seen": 123084464,
      "step": 7649
    },
    {
      "epoch": 0.535869079828732,
      "grad_norm": 4.8372344970703125,
      "learning_rate": 4.646339754816113e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 123099480,
      "step": 7650
    },
    {
      "epoch": 0.5359391280744613,
      "grad_norm": 3.768519878387451,
      "learning_rate": 4.6456399299474606e-05,
      "loss": 1.0942,
      "num_input_tokens_seen": 123115592,
      "step": 7651
    },
    {
      "epoch": 0.5360091763201905,
      "grad_norm": 5.204262733459473,
      "learning_rate": 4.6449401050788094e-05,
      "loss": 0.954,
      "num_input_tokens_seen": 123131976,
      "step": 7652
    },
    {
      "epoch": 0.5360792245659198,
      "grad_norm": 3.371913194656372,
      "learning_rate": 4.6442402802101576e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 123148360,
      "step": 7653
    },
    {
      "epoch": 0.536149272811649,
      "grad_norm": 6.089724540710449,
      "learning_rate": 4.6435404553415065e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 123164744,
      "step": 7654
    },
    {
      "epoch": 0.5362193210573782,
      "grad_norm": 4.112463474273682,
      "learning_rate": 4.6428406304728553e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 123181128,
      "step": 7655
    },
    {
      "epoch": 0.5362893693031076,
      "grad_norm": 3.9375081062316895,
      "learning_rate": 4.6421408056042035e-05,
      "loss": 1.0524,
      "num_input_tokens_seen": 123197512,
      "step": 7656
    },
    {
      "epoch": 0.5363594175488368,
      "grad_norm": 4.459086894989014,
      "learning_rate": 4.6414409807355524e-05,
      "loss": 1.206,
      "num_input_tokens_seen": 123212816,
      "step": 7657
    },
    {
      "epoch": 0.536429465794566,
      "grad_norm": 4.026162147521973,
      "learning_rate": 4.6407411558669e-05,
      "loss": 0.8556,
      "num_input_tokens_seen": 123229200,
      "step": 7658
    },
    {
      "epoch": 0.5364995140402953,
      "grad_norm": 3.6091065406799316,
      "learning_rate": 4.640041330998249e-05,
      "loss": 1.0515,
      "num_input_tokens_seen": 123245360,
      "step": 7659
    },
    {
      "epoch": 0.5365695622860245,
      "grad_norm": 4.105917453765869,
      "learning_rate": 4.639341506129598e-05,
      "loss": 1.3174,
      "num_input_tokens_seen": 123261072,
      "step": 7660
    },
    {
      "epoch": 0.5366396105317538,
      "grad_norm": 4.645833492279053,
      "learning_rate": 4.638641681260946e-05,
      "loss": 1.0443,
      "num_input_tokens_seen": 123277216,
      "step": 7661
    },
    {
      "epoch": 0.536709658777483,
      "grad_norm": 4.012742519378662,
      "learning_rate": 4.6379418563922946e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 123293336,
      "step": 7662
    },
    {
      "epoch": 0.5367797070232122,
      "grad_norm": 3.9238698482513428,
      "learning_rate": 4.637242031523643e-05,
      "loss": 1.0772,
      "num_input_tokens_seen": 123309720,
      "step": 7663
    },
    {
      "epoch": 0.5368497552689415,
      "grad_norm": 3.3745267391204834,
      "learning_rate": 4.636542206654992e-05,
      "loss": 1.0389,
      "num_input_tokens_seen": 123326104,
      "step": 7664
    },
    {
      "epoch": 0.5369198035146707,
      "grad_norm": 4.080345630645752,
      "learning_rate": 4.6358423817863405e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 123342280,
      "step": 7665
    },
    {
      "epoch": 0.5369898517603999,
      "grad_norm": 4.973495006561279,
      "learning_rate": 4.635142556917689e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 123358664,
      "step": 7666
    },
    {
      "epoch": 0.5370599000061292,
      "grad_norm": 3.8065357208251953,
      "learning_rate": 4.6344427320490376e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 123374584,
      "step": 7667
    },
    {
      "epoch": 0.5371299482518584,
      "grad_norm": 4.49127721786499,
      "learning_rate": 4.633742907180385e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 123390208,
      "step": 7668
    },
    {
      "epoch": 0.5371999964975878,
      "grad_norm": 4.231927394866943,
      "learning_rate": 4.633043082311734e-05,
      "loss": 1.0988,
      "num_input_tokens_seen": 123406592,
      "step": 7669
    },
    {
      "epoch": 0.537270044743317,
      "grad_norm": 3.7635555267333984,
      "learning_rate": 4.6323432574430835e-05,
      "loss": 1.1342,
      "num_input_tokens_seen": 123422976,
      "step": 7670
    },
    {
      "epoch": 0.5373400929890462,
      "grad_norm": 3.9398446083068848,
      "learning_rate": 4.631643432574431e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 123439184,
      "step": 7671
    },
    {
      "epoch": 0.5374101412347755,
      "grad_norm": 3.7720675468444824,
      "learning_rate": 4.63094360770578e-05,
      "loss": 1.1092,
      "num_input_tokens_seen": 123455568,
      "step": 7672
    },
    {
      "epoch": 0.5374801894805047,
      "grad_norm": 5.23007869720459,
      "learning_rate": 4.630243782837128e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 123471952,
      "step": 7673
    },
    {
      "epoch": 0.5375502377262339,
      "grad_norm": 4.356583118438721,
      "learning_rate": 4.629543957968477e-05,
      "loss": 1.1307,
      "num_input_tokens_seen": 123487528,
      "step": 7674
    },
    {
      "epoch": 0.5376202859719632,
      "grad_norm": 3.71581768989563,
      "learning_rate": 4.6288441330998244e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 123503216,
      "step": 7675
    },
    {
      "epoch": 0.5376903342176924,
      "grad_norm": 3.9850363731384277,
      "learning_rate": 4.628144308231174e-05,
      "loss": 1.0477,
      "num_input_tokens_seen": 123519600,
      "step": 7676
    },
    {
      "epoch": 0.5377603824634217,
      "grad_norm": 3.9007675647735596,
      "learning_rate": 4.627444483362523e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 123535952,
      "step": 7677
    },
    {
      "epoch": 0.5378304307091509,
      "grad_norm": 4.420581340789795,
      "learning_rate": 4.62674465849387e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 123552336,
      "step": 7678
    },
    {
      "epoch": 0.5379004789548801,
      "grad_norm": 3.788006067276001,
      "learning_rate": 4.626044833625219e-05,
      "loss": 1.0635,
      "num_input_tokens_seen": 123568232,
      "step": 7679
    },
    {
      "epoch": 0.5379705272006094,
      "grad_norm": 3.8466997146606445,
      "learning_rate": 4.625345008756567e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 123583680,
      "step": 7680
    },
    {
      "epoch": 0.5380405754463387,
      "grad_norm": 4.214776515960693,
      "learning_rate": 4.624645183887916e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 123600064,
      "step": 7681
    },
    {
      "epoch": 0.5381106236920679,
      "grad_norm": 4.797380447387695,
      "learning_rate": 4.623945359019265e-05,
      "loss": 0.9281,
      "num_input_tokens_seen": 123615880,
      "step": 7682
    },
    {
      "epoch": 0.5381806719377972,
      "grad_norm": 3.3792150020599365,
      "learning_rate": 4.623245534150613e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 123632264,
      "step": 7683
    },
    {
      "epoch": 0.5382507201835264,
      "grad_norm": 4.794241428375244,
      "learning_rate": 4.622545709281962e-05,
      "loss": 0.9202,
      "num_input_tokens_seen": 123647104,
      "step": 7684
    },
    {
      "epoch": 0.5383207684292557,
      "grad_norm": 3.86734676361084,
      "learning_rate": 4.6218458844133096e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 123662976,
      "step": 7685
    },
    {
      "epoch": 0.5383908166749849,
      "grad_norm": 4.570960998535156,
      "learning_rate": 4.621146059544659e-05,
      "loss": 1.1882,
      "num_input_tokens_seen": 123678648,
      "step": 7686
    },
    {
      "epoch": 0.5384608649207141,
      "grad_norm": 4.53627347946167,
      "learning_rate": 4.620446234676008e-05,
      "loss": 0.996,
      "num_input_tokens_seen": 123695032,
      "step": 7687
    },
    {
      "epoch": 0.5385309131664434,
      "grad_norm": 3.517305612564087,
      "learning_rate": 4.6197464098073555e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 123711416,
      "step": 7688
    },
    {
      "epoch": 0.5386009614121726,
      "grad_norm": 3.955936908721924,
      "learning_rate": 4.6190465849387043e-05,
      "loss": 1.0099,
      "num_input_tokens_seen": 123727800,
      "step": 7689
    },
    {
      "epoch": 0.5386710096579019,
      "grad_norm": 3.3592917919158936,
      "learning_rate": 4.6183467600700525e-05,
      "loss": 1.0007,
      "num_input_tokens_seen": 123744184,
      "step": 7690
    },
    {
      "epoch": 0.5387410579036311,
      "grad_norm": 3.4240005016326904,
      "learning_rate": 4.6176469352014014e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 123760544,
      "step": 7691
    },
    {
      "epoch": 0.5388111061493603,
      "grad_norm": 4.404487609863281,
      "learning_rate": 4.61694711033275e-05,
      "loss": 1.2803,
      "num_input_tokens_seen": 123776928,
      "step": 7692
    },
    {
      "epoch": 0.5388811543950897,
      "grad_norm": 3.729642868041992,
      "learning_rate": 4.6162472854640984e-05,
      "loss": 1.0132,
      "num_input_tokens_seen": 123793312,
      "step": 7693
    },
    {
      "epoch": 0.5389512026408189,
      "grad_norm": 4.076940536499023,
      "learning_rate": 4.615547460595447e-05,
      "loss": 0.9274,
      "num_input_tokens_seen": 123809696,
      "step": 7694
    },
    {
      "epoch": 0.5390212508865481,
      "grad_norm": 3.436220407485962,
      "learning_rate": 4.614847635726795e-05,
      "loss": 1.0113,
      "num_input_tokens_seen": 123826080,
      "step": 7695
    },
    {
      "epoch": 0.5390912991322774,
      "grad_norm": 3.4559690952301025,
      "learning_rate": 4.614147810858144e-05,
      "loss": 1.0622,
      "num_input_tokens_seen": 123841856,
      "step": 7696
    },
    {
      "epoch": 0.5391613473780066,
      "grad_norm": 4.085431098937988,
      "learning_rate": 4.613447985989493e-05,
      "loss": 1.0832,
      "num_input_tokens_seen": 123858240,
      "step": 7697
    },
    {
      "epoch": 0.5392313956237359,
      "grad_norm": 4.373138427734375,
      "learning_rate": 4.612748161120841e-05,
      "loss": 1.0914,
      "num_input_tokens_seen": 123874168,
      "step": 7698
    },
    {
      "epoch": 0.5393014438694651,
      "grad_norm": 4.127585411071777,
      "learning_rate": 4.6120483362521896e-05,
      "loss": 1.1582,
      "num_input_tokens_seen": 123890552,
      "step": 7699
    },
    {
      "epoch": 0.5393714921151943,
      "grad_norm": 3.53139328956604,
      "learning_rate": 4.611348511383538e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 123906936,
      "step": 7700
    },
    {
      "epoch": 0.5394415403609236,
      "grad_norm": 3.8704042434692383,
      "learning_rate": 4.6106486865148866e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 123922640,
      "step": 7701
    },
    {
      "epoch": 0.5395115886066528,
      "grad_norm": 5.1702189445495605,
      "learning_rate": 4.609948861646235e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 123938520,
      "step": 7702
    },
    {
      "epoch": 0.539581636852382,
      "grad_norm": 4.382580280303955,
      "learning_rate": 4.6092490367775836e-05,
      "loss": 1.1366,
      "num_input_tokens_seen": 123953728,
      "step": 7703
    },
    {
      "epoch": 0.5396516850981113,
      "grad_norm": 4.136844635009766,
      "learning_rate": 4.6085492119089325e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 123970112,
      "step": 7704
    },
    {
      "epoch": 0.5397217333438405,
      "grad_norm": 3.9517366886138916,
      "learning_rate": 4.60784938704028e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 123986104,
      "step": 7705
    },
    {
      "epoch": 0.5397917815895699,
      "grad_norm": 3.5815629959106445,
      "learning_rate": 4.6071495621716295e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 124002488,
      "step": 7706
    },
    {
      "epoch": 0.5398618298352991,
      "grad_norm": 3.8689863681793213,
      "learning_rate": 4.606449737302977e-05,
      "loss": 1.0321,
      "num_input_tokens_seen": 124018184,
      "step": 7707
    },
    {
      "epoch": 0.5399318780810283,
      "grad_norm": 3.911912679672241,
      "learning_rate": 4.605749912434326e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 124034568,
      "step": 7708
    },
    {
      "epoch": 0.5400019263267576,
      "grad_norm": 4.168681621551514,
      "learning_rate": 4.605050087565675e-05,
      "loss": 0.953,
      "num_input_tokens_seen": 124050952,
      "step": 7709
    },
    {
      "epoch": 0.5400719745724868,
      "grad_norm": 3.850926160812378,
      "learning_rate": 4.604350262697023e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 124066688,
      "step": 7710
    },
    {
      "epoch": 0.540142022818216,
      "grad_norm": 4.872866630554199,
      "learning_rate": 4.603650437828372e-05,
      "loss": 1.0421,
      "num_input_tokens_seen": 124083072,
      "step": 7711
    },
    {
      "epoch": 0.5402120710639453,
      "grad_norm": 3.4104743003845215,
      "learning_rate": 4.60295061295972e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 124099144,
      "step": 7712
    },
    {
      "epoch": 0.5402821193096745,
      "grad_norm": 4.460788249969482,
      "learning_rate": 4.602250788091069e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 124115528,
      "step": 7713
    },
    {
      "epoch": 0.5403521675554038,
      "grad_norm": 4.264237880706787,
      "learning_rate": 4.601550963222418e-05,
      "loss": 0.9638,
      "num_input_tokens_seen": 124131080,
      "step": 7714
    },
    {
      "epoch": 0.540422215801133,
      "grad_norm": 3.436184883117676,
      "learning_rate": 4.600851138353765e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 124147464,
      "step": 7715
    },
    {
      "epoch": 0.5404922640468622,
      "grad_norm": 5.596844673156738,
      "learning_rate": 4.600151313485115e-05,
      "loss": 0.8719,
      "num_input_tokens_seen": 124163848,
      "step": 7716
    },
    {
      "epoch": 0.5405623122925916,
      "grad_norm": 3.6911396980285645,
      "learning_rate": 4.599451488616462e-05,
      "loss": 1.0812,
      "num_input_tokens_seen": 124180000,
      "step": 7717
    },
    {
      "epoch": 0.5406323605383208,
      "grad_norm": 5.705863952636719,
      "learning_rate": 4.598751663747811e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 124196384,
      "step": 7718
    },
    {
      "epoch": 0.54070240878405,
      "grad_norm": 3.699828863143921,
      "learning_rate": 4.59805183887916e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 124212768,
      "step": 7719
    },
    {
      "epoch": 0.5407724570297793,
      "grad_norm": 4.548577785491943,
      "learning_rate": 4.597352014010508e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 124228520,
      "step": 7720
    },
    {
      "epoch": 0.5408425052755085,
      "grad_norm": 3.9667632579803467,
      "learning_rate": 4.596652189141857e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 124244312,
      "step": 7721
    },
    {
      "epoch": 0.5409125535212378,
      "grad_norm": 5.471269130706787,
      "learning_rate": 4.595952364273205e-05,
      "loss": 1.154,
      "num_input_tokens_seen": 124260448,
      "step": 7722
    },
    {
      "epoch": 0.540982601766967,
      "grad_norm": 3.6126010417938232,
      "learning_rate": 4.595252539404554e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 124276776,
      "step": 7723
    },
    {
      "epoch": 0.5410526500126962,
      "grad_norm": 3.4256107807159424,
      "learning_rate": 4.594552714535903e-05,
      "loss": 0.8039,
      "num_input_tokens_seen": 124293160,
      "step": 7724
    },
    {
      "epoch": 0.5411226982584255,
      "grad_norm": 4.028780937194824,
      "learning_rate": 4.5938528896672504e-05,
      "loss": 0.986,
      "num_input_tokens_seen": 124308648,
      "step": 7725
    },
    {
      "epoch": 0.5411927465041547,
      "grad_norm": 4.520470142364502,
      "learning_rate": 4.5931530647986e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 124324512,
      "step": 7726
    },
    {
      "epoch": 0.541262794749884,
      "grad_norm": 6.498549938201904,
      "learning_rate": 4.5924532399299474e-05,
      "loss": 0.9941,
      "num_input_tokens_seen": 124340888,
      "step": 7727
    },
    {
      "epoch": 0.5413328429956132,
      "grad_norm": 6.183528900146484,
      "learning_rate": 4.591753415061296e-05,
      "loss": 0.9104,
      "num_input_tokens_seen": 124355024,
      "step": 7728
    },
    {
      "epoch": 0.5414028912413424,
      "grad_norm": 4.973779201507568,
      "learning_rate": 4.5910535901926445e-05,
      "loss": 1.0977,
      "num_input_tokens_seen": 124370424,
      "step": 7729
    },
    {
      "epoch": 0.5414729394870718,
      "grad_norm": 11.01496410369873,
      "learning_rate": 4.5903537653239933e-05,
      "loss": 1.2337,
      "num_input_tokens_seen": 124386808,
      "step": 7730
    },
    {
      "epoch": 0.541542987732801,
      "grad_norm": 5.617726802825928,
      "learning_rate": 4.589653940455342e-05,
      "loss": 1.1204,
      "num_input_tokens_seen": 124403192,
      "step": 7731
    },
    {
      "epoch": 0.5416130359785302,
      "grad_norm": 4.480281352996826,
      "learning_rate": 4.5889541155866904e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 124419576,
      "step": 7732
    },
    {
      "epoch": 0.5416830842242595,
      "grad_norm": 4.884644985198975,
      "learning_rate": 4.588254290718039e-05,
      "loss": 0.9479,
      "num_input_tokens_seen": 124435816,
      "step": 7733
    },
    {
      "epoch": 0.5417531324699887,
      "grad_norm": 3.540273666381836,
      "learning_rate": 4.587554465849387e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 124452200,
      "step": 7734
    },
    {
      "epoch": 0.541823180715718,
      "grad_norm": 3.7844882011413574,
      "learning_rate": 4.5868546409807356e-05,
      "loss": 0.8428,
      "num_input_tokens_seen": 124468584,
      "step": 7735
    },
    {
      "epoch": 0.5418932289614472,
      "grad_norm": 3.699333906173706,
      "learning_rate": 4.586154816112085e-05,
      "loss": 1.087,
      "num_input_tokens_seen": 124484968,
      "step": 7736
    },
    {
      "epoch": 0.5419632772071764,
      "grad_norm": 4.426324844360352,
      "learning_rate": 4.5854549912434326e-05,
      "loss": 1.0829,
      "num_input_tokens_seen": 124501352,
      "step": 7737
    },
    {
      "epoch": 0.5420333254529057,
      "grad_norm": 3.796420097351074,
      "learning_rate": 4.5847551663747815e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 124517328,
      "step": 7738
    },
    {
      "epoch": 0.5421033736986349,
      "grad_norm": 4.042966842651367,
      "learning_rate": 4.58405534150613e-05,
      "loss": 1.0737,
      "num_input_tokens_seen": 124533408,
      "step": 7739
    },
    {
      "epoch": 0.5421734219443641,
      "grad_norm": 4.4333977699279785,
      "learning_rate": 4.5833555166374785e-05,
      "loss": 1.0786,
      "num_input_tokens_seen": 124549504,
      "step": 7740
    },
    {
      "epoch": 0.5422434701900934,
      "grad_norm": 3.791276216506958,
      "learning_rate": 4.5826556917688274e-05,
      "loss": 0.9739,
      "num_input_tokens_seen": 124564864,
      "step": 7741
    },
    {
      "epoch": 0.5423135184358227,
      "grad_norm": 3.6679089069366455,
      "learning_rate": 4.5819558669001756e-05,
      "loss": 0.9717,
      "num_input_tokens_seen": 124581248,
      "step": 7742
    },
    {
      "epoch": 0.542383566681552,
      "grad_norm": 4.028548717498779,
      "learning_rate": 4.5812560420315244e-05,
      "loss": 1.2381,
      "num_input_tokens_seen": 124597632,
      "step": 7743
    },
    {
      "epoch": 0.5424536149272812,
      "grad_norm": 4.555594444274902,
      "learning_rate": 4.580556217162872e-05,
      "loss": 0.9839,
      "num_input_tokens_seen": 124614016,
      "step": 7744
    },
    {
      "epoch": 0.5425236631730104,
      "grad_norm": 4.0034589767456055,
      "learning_rate": 4.579856392294221e-05,
      "loss": 1.1126,
      "num_input_tokens_seen": 124629696,
      "step": 7745
    },
    {
      "epoch": 0.5425937114187397,
      "grad_norm": 5.23121452331543,
      "learning_rate": 4.5791565674255703e-05,
      "loss": 1.2308,
      "num_input_tokens_seen": 124646080,
      "step": 7746
    },
    {
      "epoch": 0.5426637596644689,
      "grad_norm": 3.759575605392456,
      "learning_rate": 4.578456742556918e-05,
      "loss": 0.9166,
      "num_input_tokens_seen": 124662464,
      "step": 7747
    },
    {
      "epoch": 0.5427338079101981,
      "grad_norm": 3.4041309356689453,
      "learning_rate": 4.577756917688267e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 124678848,
      "step": 7748
    },
    {
      "epoch": 0.5428038561559274,
      "grad_norm": 4.347851276397705,
      "learning_rate": 4.577057092819615e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 124695232,
      "step": 7749
    },
    {
      "epoch": 0.5428739044016566,
      "grad_norm": 3.457156181335449,
      "learning_rate": 4.576357267950964e-05,
      "loss": 0.9348,
      "num_input_tokens_seen": 124711368,
      "step": 7750
    },
    {
      "epoch": 0.5429439526473859,
      "grad_norm": 4.432048320770264,
      "learning_rate": 4.5756574430823126e-05,
      "loss": 1.1717,
      "num_input_tokens_seen": 124727752,
      "step": 7751
    },
    {
      "epoch": 0.5430140008931151,
      "grad_norm": 3.337639570236206,
      "learning_rate": 4.574957618213661e-05,
      "loss": 0.7409,
      "num_input_tokens_seen": 124744136,
      "step": 7752
    },
    {
      "epoch": 0.5430840491388443,
      "grad_norm": 5.203801155090332,
      "learning_rate": 4.5742577933450096e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 124760264,
      "step": 7753
    },
    {
      "epoch": 0.5431540973845737,
      "grad_norm": 4.642807960510254,
      "learning_rate": 4.573557968476357e-05,
      "loss": 1.0197,
      "num_input_tokens_seen": 124776456,
      "step": 7754
    },
    {
      "epoch": 0.5432241456303029,
      "grad_norm": 4.211435317993164,
      "learning_rate": 4.572858143607706e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 124792320,
      "step": 7755
    },
    {
      "epoch": 0.5432941938760322,
      "grad_norm": 4.980574607849121,
      "learning_rate": 4.572158318739054e-05,
      "loss": 1.3394,
      "num_input_tokens_seen": 124808704,
      "step": 7756
    },
    {
      "epoch": 0.5433642421217614,
      "grad_norm": 3.556262493133545,
      "learning_rate": 4.571458493870403e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 124825088,
      "step": 7757
    },
    {
      "epoch": 0.5434342903674906,
      "grad_norm": 4.556125164031982,
      "learning_rate": 4.570758669001752e-05,
      "loss": 0.9926,
      "num_input_tokens_seen": 124841032,
      "step": 7758
    },
    {
      "epoch": 0.5435043386132199,
      "grad_norm": 5.515524864196777,
      "learning_rate": 4.5700588441331e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 124857416,
      "step": 7759
    },
    {
      "epoch": 0.5435743868589491,
      "grad_norm": 3.912358283996582,
      "learning_rate": 4.569359019264449e-05,
      "loss": 0.9536,
      "num_input_tokens_seen": 124873800,
      "step": 7760
    },
    {
      "epoch": 0.5436444351046783,
      "grad_norm": 3.7982399463653564,
      "learning_rate": 4.5686591943957965e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 124890184,
      "step": 7761
    },
    {
      "epoch": 0.5437144833504076,
      "grad_norm": 3.991724967956543,
      "learning_rate": 4.567959369527146e-05,
      "loss": 0.9333,
      "num_input_tokens_seen": 124906568,
      "step": 7762
    },
    {
      "epoch": 0.5437845315961368,
      "grad_norm": 5.313719272613525,
      "learning_rate": 4.567259544658495e-05,
      "loss": 1.0243,
      "num_input_tokens_seen": 124922352,
      "step": 7763
    },
    {
      "epoch": 0.5438545798418661,
      "grad_norm": 3.796652317047119,
      "learning_rate": 4.5665597197898423e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 124938472,
      "step": 7764
    },
    {
      "epoch": 0.5439246280875953,
      "grad_norm": 4.874033451080322,
      "learning_rate": 4.565859894921191e-05,
      "loss": 1.0194,
      "num_input_tokens_seen": 124954848,
      "step": 7765
    },
    {
      "epoch": 0.5439946763333245,
      "grad_norm": 3.8010215759277344,
      "learning_rate": 4.5651600700525394e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 124970488,
      "step": 7766
    },
    {
      "epoch": 0.5440647245790539,
      "grad_norm": 4.711667060852051,
      "learning_rate": 4.564460245183888e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 124986656,
      "step": 7767
    },
    {
      "epoch": 0.5441347728247831,
      "grad_norm": 5.9820356369018555,
      "learning_rate": 4.563760420315237e-05,
      "loss": 1.0015,
      "num_input_tokens_seen": 125003040,
      "step": 7768
    },
    {
      "epoch": 0.5442048210705123,
      "grad_norm": 4.217742443084717,
      "learning_rate": 4.563060595446585e-05,
      "loss": 0.907,
      "num_input_tokens_seen": 125019008,
      "step": 7769
    },
    {
      "epoch": 0.5442748693162416,
      "grad_norm": 3.3076283931732178,
      "learning_rate": 4.562360770577934e-05,
      "loss": 0.7289,
      "num_input_tokens_seen": 125035272,
      "step": 7770
    },
    {
      "epoch": 0.5443449175619708,
      "grad_norm": 3.409607172012329,
      "learning_rate": 4.5616609457092817e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 125051656,
      "step": 7771
    },
    {
      "epoch": 0.5444149658077001,
      "grad_norm": 5.260388374328613,
      "learning_rate": 4.560961120840631e-05,
      "loss": 0.9778,
      "num_input_tokens_seen": 125068040,
      "step": 7772
    },
    {
      "epoch": 0.5444850140534293,
      "grad_norm": 4.905508518218994,
      "learning_rate": 4.56026129597198e-05,
      "loss": 1.2631,
      "num_input_tokens_seen": 125084192,
      "step": 7773
    },
    {
      "epoch": 0.5445550622991585,
      "grad_norm": 4.701261043548584,
      "learning_rate": 4.5595614711033276e-05,
      "loss": 1.1305,
      "num_input_tokens_seen": 125098976,
      "step": 7774
    },
    {
      "epoch": 0.5446251105448878,
      "grad_norm": 4.822204113006592,
      "learning_rate": 4.5588616462346764e-05,
      "loss": 1.0407,
      "num_input_tokens_seen": 125114624,
      "step": 7775
    },
    {
      "epoch": 0.544695158790617,
      "grad_norm": 3.7025883197784424,
      "learning_rate": 4.5581618213660246e-05,
      "loss": 1.0469,
      "num_input_tokens_seen": 125130560,
      "step": 7776
    },
    {
      "epoch": 0.5447652070363462,
      "grad_norm": 4.998040676116943,
      "learning_rate": 4.5574619964973735e-05,
      "loss": 1.0568,
      "num_input_tokens_seen": 125146944,
      "step": 7777
    },
    {
      "epoch": 0.5448352552820755,
      "grad_norm": 3.457750082015991,
      "learning_rate": 4.5567621716287216e-05,
      "loss": 0.8956,
      "num_input_tokens_seen": 125162520,
      "step": 7778
    },
    {
      "epoch": 0.5449053035278048,
      "grad_norm": 3.417926788330078,
      "learning_rate": 4.5560623467600705e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 125178656,
      "step": 7779
    },
    {
      "epoch": 0.5449753517735341,
      "grad_norm": 4.075389385223389,
      "learning_rate": 4.5553625218914194e-05,
      "loss": 0.9655,
      "num_input_tokens_seen": 125195040,
      "step": 7780
    },
    {
      "epoch": 0.5450454000192633,
      "grad_norm": 4.12037992477417,
      "learning_rate": 4.554662697022767e-05,
      "loss": 1.3969,
      "num_input_tokens_seen": 125211424,
      "step": 7781
    },
    {
      "epoch": 0.5451154482649925,
      "grad_norm": 4.0456671714782715,
      "learning_rate": 4.5539628721541164e-05,
      "loss": 0.9851,
      "num_input_tokens_seen": 125227752,
      "step": 7782
    },
    {
      "epoch": 0.5451854965107218,
      "grad_norm": 4.972954273223877,
      "learning_rate": 4.553263047285464e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 125242792,
      "step": 7783
    },
    {
      "epoch": 0.545255544756451,
      "grad_norm": 3.667360544204712,
      "learning_rate": 4.552563222416813e-05,
      "loss": 1.194,
      "num_input_tokens_seen": 125259176,
      "step": 7784
    },
    {
      "epoch": 0.5453255930021802,
      "grad_norm": 3.54160737991333,
      "learning_rate": 4.5518633975481616e-05,
      "loss": 0.9758,
      "num_input_tokens_seen": 125275560,
      "step": 7785
    },
    {
      "epoch": 0.5453956412479095,
      "grad_norm": 3.7189040184020996,
      "learning_rate": 4.55116357267951e-05,
      "loss": 0.8118,
      "num_input_tokens_seen": 125291944,
      "step": 7786
    },
    {
      "epoch": 0.5454656894936387,
      "grad_norm": 3.435598611831665,
      "learning_rate": 4.5504637478108587e-05,
      "loss": 0.9343,
      "num_input_tokens_seen": 125307528,
      "step": 7787
    },
    {
      "epoch": 0.545535737739368,
      "grad_norm": 3.7623162269592285,
      "learning_rate": 4.549763922942207e-05,
      "loss": 1.2752,
      "num_input_tokens_seen": 125323816,
      "step": 7788
    },
    {
      "epoch": 0.5456057859850972,
      "grad_norm": 4.6416239738464355,
      "learning_rate": 4.549064098073556e-05,
      "loss": 1.0901,
      "num_input_tokens_seen": 125340200,
      "step": 7789
    },
    {
      "epoch": 0.5456758342308264,
      "grad_norm": 4.615113258361816,
      "learning_rate": 4.5483642732049046e-05,
      "loss": 0.977,
      "num_input_tokens_seen": 125356584,
      "step": 7790
    },
    {
      "epoch": 0.5457458824765558,
      "grad_norm": 3.8960089683532715,
      "learning_rate": 4.547664448336252e-05,
      "loss": 0.9974,
      "num_input_tokens_seen": 125372968,
      "step": 7791
    },
    {
      "epoch": 0.545815930722285,
      "grad_norm": 3.9642269611358643,
      "learning_rate": 4.5469646234676016e-05,
      "loss": 0.9801,
      "num_input_tokens_seen": 125389352,
      "step": 7792
    },
    {
      "epoch": 0.5458859789680143,
      "grad_norm": 5.444625377655029,
      "learning_rate": 4.546264798598949e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 125404624,
      "step": 7793
    },
    {
      "epoch": 0.5459560272137435,
      "grad_norm": 4.1974053382873535,
      "learning_rate": 4.545564973730298e-05,
      "loss": 0.9788,
      "num_input_tokens_seen": 125420944,
      "step": 7794
    },
    {
      "epoch": 0.5460260754594727,
      "grad_norm": 5.193080425262451,
      "learning_rate": 4.544865148861647e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 125437328,
      "step": 7795
    },
    {
      "epoch": 0.546096123705202,
      "grad_norm": 5.049325942993164,
      "learning_rate": 4.544165323992995e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 125452656,
      "step": 7796
    },
    {
      "epoch": 0.5461661719509312,
      "grad_norm": 4.1581549644470215,
      "learning_rate": 4.543465499124344e-05,
      "loss": 0.873,
      "num_input_tokens_seen": 125469040,
      "step": 7797
    },
    {
      "epoch": 0.5462362201966604,
      "grad_norm": 4.484875202178955,
      "learning_rate": 4.542765674255692e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 125485168,
      "step": 7798
    },
    {
      "epoch": 0.5463062684423897,
      "grad_norm": 4.153511047363281,
      "learning_rate": 4.542065849387041e-05,
      "loss": 1.2482,
      "num_input_tokens_seen": 125501336,
      "step": 7799
    },
    {
      "epoch": 0.5463763166881189,
      "grad_norm": 4.060786724090576,
      "learning_rate": 4.54136602451839e-05,
      "loss": 1.0232,
      "num_input_tokens_seen": 125517688,
      "step": 7800
    },
    {
      "epoch": 0.5463763166881189,
      "eval_loss": 1.121619462966919,
      "eval_runtime": 0.2054,
      "eval_samples_per_second": 4.868,
      "eval_steps_per_second": 4.868,
      "num_input_tokens_seen": 125517688,
      "step": 7800
    },
    {
      "epoch": 0.5464463649338482,
      "grad_norm": 5.799534797668457,
      "learning_rate": 4.540666199649737e-05,
      "loss": 1.2131,
      "num_input_tokens_seen": 125533408,
      "step": 7801
    },
    {
      "epoch": 0.5465164131795774,
      "grad_norm": 3.7412962913513184,
      "learning_rate": 4.539966374781087e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 125549792,
      "step": 7802
    },
    {
      "epoch": 0.5465864614253066,
      "grad_norm": 3.978907346725464,
      "learning_rate": 4.539266549912434e-05,
      "loss": 1.0106,
      "num_input_tokens_seen": 125565352,
      "step": 7803
    },
    {
      "epoch": 0.546656509671036,
      "grad_norm": 4.388980865478516,
      "learning_rate": 4.538566725043783e-05,
      "loss": 1.0467,
      "num_input_tokens_seen": 125580864,
      "step": 7804
    },
    {
      "epoch": 0.5467265579167652,
      "grad_norm": 3.432842969894409,
      "learning_rate": 4.5378669001751313e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 125597248,
      "step": 7805
    },
    {
      "epoch": 0.5467966061624944,
      "grad_norm": 4.419676303863525,
      "learning_rate": 4.53716707530648e-05,
      "loss": 1.053,
      "num_input_tokens_seen": 125613448,
      "step": 7806
    },
    {
      "epoch": 0.5468666544082237,
      "grad_norm": 4.717494964599609,
      "learning_rate": 4.536467250437829e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 125629728,
      "step": 7807
    },
    {
      "epoch": 0.5469367026539529,
      "grad_norm": 3.8088278770446777,
      "learning_rate": 4.535767425569177e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 125646112,
      "step": 7808
    },
    {
      "epoch": 0.5470067508996822,
      "grad_norm": 4.486949443817139,
      "learning_rate": 4.535067600700526e-05,
      "loss": 1.0894,
      "num_input_tokens_seen": 125662496,
      "step": 7809
    },
    {
      "epoch": 0.5470767991454114,
      "grad_norm": 4.220696926116943,
      "learning_rate": 4.5343677758318736e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 125678192,
      "step": 7810
    },
    {
      "epoch": 0.5471468473911406,
      "grad_norm": 3.5514204502105713,
      "learning_rate": 4.5336679509632225e-05,
      "loss": 0.934,
      "num_input_tokens_seen": 125693824,
      "step": 7811
    },
    {
      "epoch": 0.5472168956368699,
      "grad_norm": 4.971661567687988,
      "learning_rate": 4.532968126094572e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 125709184,
      "step": 7812
    },
    {
      "epoch": 0.5472869438825991,
      "grad_norm": 4.215356349945068,
      "learning_rate": 4.5322683012259195e-05,
      "loss": 1.1258,
      "num_input_tokens_seen": 125725568,
      "step": 7813
    },
    {
      "epoch": 0.5473569921283283,
      "grad_norm": 3.7598018646240234,
      "learning_rate": 4.5315684763572684e-05,
      "loss": 0.9901,
      "num_input_tokens_seen": 125741952,
      "step": 7814
    },
    {
      "epoch": 0.5474270403740576,
      "grad_norm": 4.145439147949219,
      "learning_rate": 4.5308686514886165e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 125757792,
      "step": 7815
    },
    {
      "epoch": 0.5474970886197869,
      "grad_norm": 4.645499229431152,
      "learning_rate": 4.5301688266199654e-05,
      "loss": 0.964,
      "num_input_tokens_seen": 125774176,
      "step": 7816
    },
    {
      "epoch": 0.5475671368655162,
      "grad_norm": 3.526381015777588,
      "learning_rate": 4.529469001751314e-05,
      "loss": 1.098,
      "num_input_tokens_seen": 125790264,
      "step": 7817
    },
    {
      "epoch": 0.5476371851112454,
      "grad_norm": 4.389588356018066,
      "learning_rate": 4.5287691768826624e-05,
      "loss": 1.2777,
      "num_input_tokens_seen": 125806648,
      "step": 7818
    },
    {
      "epoch": 0.5477072333569746,
      "grad_norm": 3.5831832885742188,
      "learning_rate": 4.528069352014011e-05,
      "loss": 1.1427,
      "num_input_tokens_seen": 125822848,
      "step": 7819
    },
    {
      "epoch": 0.5477772816027039,
      "grad_norm": 5.075149059295654,
      "learning_rate": 4.527369527145359e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 125839232,
      "step": 7820
    },
    {
      "epoch": 0.5478473298484331,
      "grad_norm": 3.7986133098602295,
      "learning_rate": 4.526669702276708e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 125855616,
      "step": 7821
    },
    {
      "epoch": 0.5479173780941623,
      "grad_norm": 3.6170661449432373,
      "learning_rate": 4.525969877408057e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 125872000,
      "step": 7822
    },
    {
      "epoch": 0.5479874263398916,
      "grad_norm": 3.752514123916626,
      "learning_rate": 4.525270052539405e-05,
      "loss": 0.9744,
      "num_input_tokens_seen": 125888384,
      "step": 7823
    },
    {
      "epoch": 0.5480574745856208,
      "grad_norm": 3.609358072280884,
      "learning_rate": 4.5245702276707536e-05,
      "loss": 1.0688,
      "num_input_tokens_seen": 125904768,
      "step": 7824
    },
    {
      "epoch": 0.5481275228313501,
      "grad_norm": 5.24570369720459,
      "learning_rate": 4.523870402802102e-05,
      "loss": 1.0225,
      "num_input_tokens_seen": 125920592,
      "step": 7825
    },
    {
      "epoch": 0.5481975710770793,
      "grad_norm": 5.084728240966797,
      "learning_rate": 4.5231705779334506e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 125936976,
      "step": 7826
    },
    {
      "epoch": 0.5482676193228085,
      "grad_norm": 4.076999664306641,
      "learning_rate": 4.5224707530647995e-05,
      "loss": 1.0143,
      "num_input_tokens_seen": 125953048,
      "step": 7827
    },
    {
      "epoch": 0.5483376675685379,
      "grad_norm": 3.8320982456207275,
      "learning_rate": 4.5217709281961476e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 125969432,
      "step": 7828
    },
    {
      "epoch": 0.5484077158142671,
      "grad_norm": 5.069493293762207,
      "learning_rate": 4.5210711033274965e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 125985688,
      "step": 7829
    },
    {
      "epoch": 0.5484777640599964,
      "grad_norm": 3.8555328845977783,
      "learning_rate": 4.520371278458844e-05,
      "loss": 1.0394,
      "num_input_tokens_seen": 126002072,
      "step": 7830
    },
    {
      "epoch": 0.5485478123057256,
      "grad_norm": 3.5486679077148438,
      "learning_rate": 4.519671453590193e-05,
      "loss": 1.0865,
      "num_input_tokens_seen": 126018456,
      "step": 7831
    },
    {
      "epoch": 0.5486178605514548,
      "grad_norm": 4.256968021392822,
      "learning_rate": 4.518971628721541e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 126034272,
      "step": 7832
    },
    {
      "epoch": 0.5486879087971841,
      "grad_norm": 3.594381332397461,
      "learning_rate": 4.51827180385289e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 126050656,
      "step": 7833
    },
    {
      "epoch": 0.5487579570429133,
      "grad_norm": 3.4965176582336426,
      "learning_rate": 4.517571978984239e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 126067040,
      "step": 7834
    },
    {
      "epoch": 0.5488280052886425,
      "grad_norm": 3.762791395187378,
      "learning_rate": 4.516872154115587e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 126083384,
      "step": 7835
    },
    {
      "epoch": 0.5488980535343718,
      "grad_norm": 4.816859245300293,
      "learning_rate": 4.516172329246936e-05,
      "loss": 1.1516,
      "num_input_tokens_seen": 126099768,
      "step": 7836
    },
    {
      "epoch": 0.548968101780101,
      "grad_norm": 4.410999774932861,
      "learning_rate": 4.515472504378283e-05,
      "loss": 1.1552,
      "num_input_tokens_seen": 126114976,
      "step": 7837
    },
    {
      "epoch": 0.5490381500258303,
      "grad_norm": 3.48974609375,
      "learning_rate": 4.514772679509633e-05,
      "loss": 0.8017,
      "num_input_tokens_seen": 126131360,
      "step": 7838
    },
    {
      "epoch": 0.5491081982715595,
      "grad_norm": 4.172264575958252,
      "learning_rate": 4.514072854640982e-05,
      "loss": 0.9508,
      "num_input_tokens_seen": 126147080,
      "step": 7839
    },
    {
      "epoch": 0.5491782465172887,
      "grad_norm": 4.311397075653076,
      "learning_rate": 4.513373029772329e-05,
      "loss": 1.1269,
      "num_input_tokens_seen": 126163464,
      "step": 7840
    },
    {
      "epoch": 0.5492482947630181,
      "grad_norm": 3.950122833251953,
      "learning_rate": 4.512673204903678e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 126179848,
      "step": 7841
    },
    {
      "epoch": 0.5493183430087473,
      "grad_norm": 3.5279722213745117,
      "learning_rate": 4.511973380035026e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 126196232,
      "step": 7842
    },
    {
      "epoch": 0.5493883912544765,
      "grad_norm": 3.958651065826416,
      "learning_rate": 4.511273555166375e-05,
      "loss": 1.2114,
      "num_input_tokens_seen": 126212616,
      "step": 7843
    },
    {
      "epoch": 0.5494584395002058,
      "grad_norm": 4.567907810211182,
      "learning_rate": 4.510573730297724e-05,
      "loss": 1.0151,
      "num_input_tokens_seen": 126229000,
      "step": 7844
    },
    {
      "epoch": 0.549528487745935,
      "grad_norm": 5.533442497253418,
      "learning_rate": 4.509873905429072e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 126243808,
      "step": 7845
    },
    {
      "epoch": 0.5495985359916643,
      "grad_norm": 3.7990267276763916,
      "learning_rate": 4.509174080560421e-05,
      "loss": 1.023,
      "num_input_tokens_seen": 126260136,
      "step": 7846
    },
    {
      "epoch": 0.5496685842373935,
      "grad_norm": 3.648163318634033,
      "learning_rate": 4.5084742556917685e-05,
      "loss": 1.1331,
      "num_input_tokens_seen": 126276488,
      "step": 7847
    },
    {
      "epoch": 0.5497386324831227,
      "grad_norm": 3.247767686843872,
      "learning_rate": 4.507774430823118e-05,
      "loss": 0.8663,
      "num_input_tokens_seen": 126292872,
      "step": 7848
    },
    {
      "epoch": 0.549808680728852,
      "grad_norm": 3.6284050941467285,
      "learning_rate": 4.507074605954467e-05,
      "loss": 1.2207,
      "num_input_tokens_seen": 126309256,
      "step": 7849
    },
    {
      "epoch": 0.5498787289745812,
      "grad_norm": 3.714810848236084,
      "learning_rate": 4.5063747810858144e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 126324880,
      "step": 7850
    },
    {
      "epoch": 0.5499487772203104,
      "grad_norm": 4.455381870269775,
      "learning_rate": 4.505674956217163e-05,
      "loss": 0.9736,
      "num_input_tokens_seen": 126341264,
      "step": 7851
    },
    {
      "epoch": 0.5500188254660398,
      "grad_norm": 6.539924144744873,
      "learning_rate": 4.5049751313485115e-05,
      "loss": 1.1598,
      "num_input_tokens_seen": 126357648,
      "step": 7852
    },
    {
      "epoch": 0.550088873711769,
      "grad_norm": 4.244390487670898,
      "learning_rate": 4.50427530647986e-05,
      "loss": 1.2593,
      "num_input_tokens_seen": 126373888,
      "step": 7853
    },
    {
      "epoch": 0.5501589219574983,
      "grad_norm": 4.415054798126221,
      "learning_rate": 4.503575481611209e-05,
      "loss": 1.2661,
      "num_input_tokens_seen": 126389976,
      "step": 7854
    },
    {
      "epoch": 0.5502289702032275,
      "grad_norm": 5.152358531951904,
      "learning_rate": 4.5028756567425574e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 126406000,
      "step": 7855
    },
    {
      "epoch": 0.5502990184489567,
      "grad_norm": 4.243201732635498,
      "learning_rate": 4.502175831873906e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 126422384,
      "step": 7856
    },
    {
      "epoch": 0.550369066694686,
      "grad_norm": 4.7400898933410645,
      "learning_rate": 4.501476007005254e-05,
      "loss": 1.1462,
      "num_input_tokens_seen": 126438552,
      "step": 7857
    },
    {
      "epoch": 0.5504391149404152,
      "grad_norm": 3.5386011600494385,
      "learning_rate": 4.500776182136603e-05,
      "loss": 0.9501,
      "num_input_tokens_seen": 126454936,
      "step": 7858
    },
    {
      "epoch": 0.5505091631861445,
      "grad_norm": 3.939976692199707,
      "learning_rate": 4.500076357267951e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 126471320,
      "step": 7859
    },
    {
      "epoch": 0.5505792114318737,
      "grad_norm": 3.7520668506622314,
      "learning_rate": 4.4993765323992996e-05,
      "loss": 1.1897,
      "num_input_tokens_seen": 126487144,
      "step": 7860
    },
    {
      "epoch": 0.5506492596776029,
      "grad_norm": 3.8446950912475586,
      "learning_rate": 4.4986767075306485e-05,
      "loss": 1.1295,
      "num_input_tokens_seen": 126503208,
      "step": 7861
    },
    {
      "epoch": 0.5507193079233322,
      "grad_norm": 5.0048909187316895,
      "learning_rate": 4.4979768826619967e-05,
      "loss": 1.1618,
      "num_input_tokens_seen": 126519592,
      "step": 7862
    },
    {
      "epoch": 0.5507893561690614,
      "grad_norm": 3.4074764251708984,
      "learning_rate": 4.4972770577933455e-05,
      "loss": 0.9419,
      "num_input_tokens_seen": 126535928,
      "step": 7863
    },
    {
      "epoch": 0.5508594044147906,
      "grad_norm": 4.959311485290527,
      "learning_rate": 4.496577232924694e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 126551768,
      "step": 7864
    },
    {
      "epoch": 0.55092945266052,
      "grad_norm": 4.260848045349121,
      "learning_rate": 4.4958774080560426e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 126566968,
      "step": 7865
    },
    {
      "epoch": 0.5509995009062492,
      "grad_norm": 4.374674320220947,
      "learning_rate": 4.4951775831873914e-05,
      "loss": 1.1858,
      "num_input_tokens_seen": 126582432,
      "step": 7866
    },
    {
      "epoch": 0.5510695491519785,
      "grad_norm": 3.808882236480713,
      "learning_rate": 4.494477758318739e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 126598272,
      "step": 7867
    },
    {
      "epoch": 0.5511395973977077,
      "grad_norm": 4.428232192993164,
      "learning_rate": 4.493777933450088e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 126614656,
      "step": 7868
    },
    {
      "epoch": 0.5512096456434369,
      "grad_norm": 3.7343485355377197,
      "learning_rate": 4.493078108581436e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 126630920,
      "step": 7869
    },
    {
      "epoch": 0.5512796938891662,
      "grad_norm": 3.8342700004577637,
      "learning_rate": 4.492378283712785e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 126647304,
      "step": 7870
    },
    {
      "epoch": 0.5513497421348954,
      "grad_norm": 3.969151496887207,
      "learning_rate": 4.491678458844134e-05,
      "loss": 1.1645,
      "num_input_tokens_seen": 126663688,
      "step": 7871
    },
    {
      "epoch": 0.5514197903806246,
      "grad_norm": 4.096588134765625,
      "learning_rate": 4.490978633975482e-05,
      "loss": 1.2369,
      "num_input_tokens_seen": 126680072,
      "step": 7872
    },
    {
      "epoch": 0.5514898386263539,
      "grad_norm": 3.6431949138641357,
      "learning_rate": 4.490278809106831e-05,
      "loss": 0.8899,
      "num_input_tokens_seen": 126696456,
      "step": 7873
    },
    {
      "epoch": 0.5515598868720831,
      "grad_norm": 3.6036217212677,
      "learning_rate": 4.489578984238179e-05,
      "loss": 0.9629,
      "num_input_tokens_seen": 126712464,
      "step": 7874
    },
    {
      "epoch": 0.5516299351178124,
      "grad_norm": 4.925845146179199,
      "learning_rate": 4.488879159369528e-05,
      "loss": 1.0951,
      "num_input_tokens_seen": 126727664,
      "step": 7875
    },
    {
      "epoch": 0.5516999833635416,
      "grad_norm": 3.822420120239258,
      "learning_rate": 4.4881793345008766e-05,
      "loss": 0.9803,
      "num_input_tokens_seen": 126742280,
      "step": 7876
    },
    {
      "epoch": 0.5517700316092709,
      "grad_norm": 4.6314005851745605,
      "learning_rate": 4.487479509632224e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 126758664,
      "step": 7877
    },
    {
      "epoch": 0.5518400798550002,
      "grad_norm": 3.5483834743499756,
      "learning_rate": 4.486779684763573e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 126774160,
      "step": 7878
    },
    {
      "epoch": 0.5519101281007294,
      "grad_norm": 3.582298755645752,
      "learning_rate": 4.486079859894921e-05,
      "loss": 1.0404,
      "num_input_tokens_seen": 126790344,
      "step": 7879
    },
    {
      "epoch": 0.5519801763464586,
      "grad_norm": 6.529607772827148,
      "learning_rate": 4.48538003502627e-05,
      "loss": 0.8033,
      "num_input_tokens_seen": 126806728,
      "step": 7880
    },
    {
      "epoch": 0.5520502245921879,
      "grad_norm": 4.167466640472412,
      "learning_rate": 4.484680210157619e-05,
      "loss": 1.1219,
      "num_input_tokens_seen": 126822792,
      "step": 7881
    },
    {
      "epoch": 0.5521202728379171,
      "grad_norm": 4.332005500793457,
      "learning_rate": 4.483980385288967e-05,
      "loss": 1.093,
      "num_input_tokens_seen": 126838952,
      "step": 7882
    },
    {
      "epoch": 0.5521903210836464,
      "grad_norm": 7.588165283203125,
      "learning_rate": 4.483280560420316e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 126855336,
      "step": 7883
    },
    {
      "epoch": 0.5522603693293756,
      "grad_norm": 5.086427688598633,
      "learning_rate": 4.4825807355516634e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 126871448,
      "step": 7884
    },
    {
      "epoch": 0.5523304175751048,
      "grad_norm": 5.5746355056762695,
      "learning_rate": 4.481880910683013e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 126886424,
      "step": 7885
    },
    {
      "epoch": 0.5524004658208341,
      "grad_norm": 3.833796262741089,
      "learning_rate": 4.4811810858143605e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 126902808,
      "step": 7886
    },
    {
      "epoch": 0.5524705140665633,
      "grad_norm": 3.436943769454956,
      "learning_rate": 4.480481260945709e-05,
      "loss": 0.8928,
      "num_input_tokens_seen": 126918960,
      "step": 7887
    },
    {
      "epoch": 0.5525405623122925,
      "grad_norm": 3.8391594886779785,
      "learning_rate": 4.479781436077058e-05,
      "loss": 1.0474,
      "num_input_tokens_seen": 126934432,
      "step": 7888
    },
    {
      "epoch": 0.5526106105580219,
      "grad_norm": 5.476257801055908,
      "learning_rate": 4.4790816112084064e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 126949912,
      "step": 7889
    },
    {
      "epoch": 0.5526806588037511,
      "grad_norm": 3.628587245941162,
      "learning_rate": 4.478381786339755e-05,
      "loss": 1.013,
      "num_input_tokens_seen": 126966296,
      "step": 7890
    },
    {
      "epoch": 0.5527507070494804,
      "grad_norm": 4.087871551513672,
      "learning_rate": 4.4776819614711034e-05,
      "loss": 1.0916,
      "num_input_tokens_seen": 126982680,
      "step": 7891
    },
    {
      "epoch": 0.5528207552952096,
      "grad_norm": 4.438427448272705,
      "learning_rate": 4.476982136602452e-05,
      "loss": 1.1005,
      "num_input_tokens_seen": 126998072,
      "step": 7892
    },
    {
      "epoch": 0.5528908035409388,
      "grad_norm": 3.5243074893951416,
      "learning_rate": 4.476282311733801e-05,
      "loss": 0.7855,
      "num_input_tokens_seen": 127014104,
      "step": 7893
    },
    {
      "epoch": 0.5529608517866681,
      "grad_norm": 10.454376220703125,
      "learning_rate": 4.4755824868651486e-05,
      "loss": 0.9935,
      "num_input_tokens_seen": 127030488,
      "step": 7894
    },
    {
      "epoch": 0.5530309000323973,
      "grad_norm": 3.5278947353363037,
      "learning_rate": 4.474882661996498e-05,
      "loss": 1.1274,
      "num_input_tokens_seen": 127046536,
      "step": 7895
    },
    {
      "epoch": 0.5531009482781266,
      "grad_norm": 3.8515357971191406,
      "learning_rate": 4.474182837127846e-05,
      "loss": 1.0061,
      "num_input_tokens_seen": 127062200,
      "step": 7896
    },
    {
      "epoch": 0.5531709965238558,
      "grad_norm": 4.094476699829102,
      "learning_rate": 4.4734830122591945e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 127077656,
      "step": 7897
    },
    {
      "epoch": 0.553241044769585,
      "grad_norm": 3.9371705055236816,
      "learning_rate": 4.4727831873905434e-05,
      "loss": 1.1099,
      "num_input_tokens_seen": 127093464,
      "step": 7898
    },
    {
      "epoch": 0.5533110930153143,
      "grad_norm": 5.384475231170654,
      "learning_rate": 4.4720833625218916e-05,
      "loss": 1.0207,
      "num_input_tokens_seen": 127109848,
      "step": 7899
    },
    {
      "epoch": 0.5533811412610435,
      "grad_norm": 4.609706878662109,
      "learning_rate": 4.4713835376532404e-05,
      "loss": 0.9562,
      "num_input_tokens_seen": 127126232,
      "step": 7900
    },
    {
      "epoch": 0.5534511895067727,
      "grad_norm": 5.60554838180542,
      "learning_rate": 4.4706837127845886e-05,
      "loss": 1.0052,
      "num_input_tokens_seen": 127142616,
      "step": 7901
    },
    {
      "epoch": 0.5535212377525021,
      "grad_norm": 4.736166477203369,
      "learning_rate": 4.4699838879159375e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 127159000,
      "step": 7902
    },
    {
      "epoch": 0.5535912859982313,
      "grad_norm": 3.9779133796691895,
      "learning_rate": 4.469284063047286e-05,
      "loss": 1.0068,
      "num_input_tokens_seen": 127175384,
      "step": 7903
    },
    {
      "epoch": 0.5536613342439606,
      "grad_norm": 4.673941612243652,
      "learning_rate": 4.468584238178634e-05,
      "loss": 1.2606,
      "num_input_tokens_seen": 127191752,
      "step": 7904
    },
    {
      "epoch": 0.5537313824896898,
      "grad_norm": 3.722839117050171,
      "learning_rate": 4.4678844133099834e-05,
      "loss": 1.0554,
      "num_input_tokens_seen": 127207408,
      "step": 7905
    },
    {
      "epoch": 0.553801430735419,
      "grad_norm": 4.31935453414917,
      "learning_rate": 4.467184588441331e-05,
      "loss": 1.0523,
      "num_input_tokens_seen": 127223792,
      "step": 7906
    },
    {
      "epoch": 0.5538714789811483,
      "grad_norm": 3.686140775680542,
      "learning_rate": 4.46648476357268e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 127240176,
      "step": 7907
    },
    {
      "epoch": 0.5539415272268775,
      "grad_norm": 4.997079372406006,
      "learning_rate": 4.4657849387040286e-05,
      "loss": 1.2024,
      "num_input_tokens_seen": 127255560,
      "step": 7908
    },
    {
      "epoch": 0.5540115754726067,
      "grad_norm": 4.612823963165283,
      "learning_rate": 4.465085113835377e-05,
      "loss": 1.0067,
      "num_input_tokens_seen": 127271336,
      "step": 7909
    },
    {
      "epoch": 0.554081623718336,
      "grad_norm": 3.857698678970337,
      "learning_rate": 4.4643852889667256e-05,
      "loss": 1.2211,
      "num_input_tokens_seen": 127287720,
      "step": 7910
    },
    {
      "epoch": 0.5541516719640652,
      "grad_norm": 3.8385705947875977,
      "learning_rate": 4.463685464098074e-05,
      "loss": 1.2123,
      "num_input_tokens_seen": 127304104,
      "step": 7911
    },
    {
      "epoch": 0.5542217202097945,
      "grad_norm": 3.910621166229248,
      "learning_rate": 4.462985639229423e-05,
      "loss": 1.071,
      "num_input_tokens_seen": 127319512,
      "step": 7912
    },
    {
      "epoch": 0.5542917684555237,
      "grad_norm": 4.685849666595459,
      "learning_rate": 4.46228581436077e-05,
      "loss": 0.9889,
      "num_input_tokens_seen": 127335736,
      "step": 7913
    },
    {
      "epoch": 0.554361816701253,
      "grad_norm": 3.6313934326171875,
      "learning_rate": 4.461585989492119e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 127351896,
      "step": 7914
    },
    {
      "epoch": 0.5544318649469823,
      "grad_norm": 3.3593709468841553,
      "learning_rate": 4.4608861646234686e-05,
      "loss": 0.8789,
      "num_input_tokens_seen": 127368032,
      "step": 7915
    },
    {
      "epoch": 0.5545019131927115,
      "grad_norm": 3.924467086791992,
      "learning_rate": 4.460186339754816e-05,
      "loss": 1.0038,
      "num_input_tokens_seen": 127384416,
      "step": 7916
    },
    {
      "epoch": 0.5545719614384407,
      "grad_norm": 4.749776363372803,
      "learning_rate": 4.459486514886165e-05,
      "loss": 1.2994,
      "num_input_tokens_seen": 127400800,
      "step": 7917
    },
    {
      "epoch": 0.55464200968417,
      "grad_norm": 4.43412446975708,
      "learning_rate": 4.458786690017513e-05,
      "loss": 1.1877,
      "num_input_tokens_seen": 127416184,
      "step": 7918
    },
    {
      "epoch": 0.5547120579298992,
      "grad_norm": 3.626112699508667,
      "learning_rate": 4.458086865148862e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 127432568,
      "step": 7919
    },
    {
      "epoch": 0.5547821061756285,
      "grad_norm": 4.031342506408691,
      "learning_rate": 4.457387040280211e-05,
      "loss": 1.0058,
      "num_input_tokens_seen": 127448656,
      "step": 7920
    },
    {
      "epoch": 0.5548521544213577,
      "grad_norm": 4.554370403289795,
      "learning_rate": 4.456687215411559e-05,
      "loss": 1.101,
      "num_input_tokens_seen": 127462584,
      "step": 7921
    },
    {
      "epoch": 0.5549222026670869,
      "grad_norm": 4.46144437789917,
      "learning_rate": 4.455987390542908e-05,
      "loss": 1.1095,
      "num_input_tokens_seen": 127478968,
      "step": 7922
    },
    {
      "epoch": 0.5549922509128162,
      "grad_norm": 3.759772539138794,
      "learning_rate": 4.4552875656742554e-05,
      "loss": 1.1723,
      "num_input_tokens_seen": 127494856,
      "step": 7923
    },
    {
      "epoch": 0.5550622991585454,
      "grad_norm": 6.424376010894775,
      "learning_rate": 4.454587740805604e-05,
      "loss": 1.1691,
      "num_input_tokens_seen": 127511240,
      "step": 7924
    },
    {
      "epoch": 0.5551323474042746,
      "grad_norm": 4.5455708503723145,
      "learning_rate": 4.453887915936954e-05,
      "loss": 1.0578,
      "num_input_tokens_seen": 127527624,
      "step": 7925
    },
    {
      "epoch": 0.555202395650004,
      "grad_norm": 3.673654794692993,
      "learning_rate": 4.453188091068301e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 127543200,
      "step": 7926
    },
    {
      "epoch": 0.5552724438957332,
      "grad_norm": 5.912845134735107,
      "learning_rate": 4.45248826619965e-05,
      "loss": 1.1171,
      "num_input_tokens_seen": 127558704,
      "step": 7927
    },
    {
      "epoch": 0.5553424921414625,
      "grad_norm": 4.603636741638184,
      "learning_rate": 4.451788441330998e-05,
      "loss": 1.0536,
      "num_input_tokens_seen": 127575088,
      "step": 7928
    },
    {
      "epoch": 0.5554125403871917,
      "grad_norm": 4.067967414855957,
      "learning_rate": 4.451088616462347e-05,
      "loss": 1.3762,
      "num_input_tokens_seen": 127590584,
      "step": 7929
    },
    {
      "epoch": 0.5554825886329209,
      "grad_norm": 4.261424541473389,
      "learning_rate": 4.450388791593696e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 127606320,
      "step": 7930
    },
    {
      "epoch": 0.5555526368786502,
      "grad_norm": 3.4832777976989746,
      "learning_rate": 4.449688966725044e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 127622704,
      "step": 7931
    },
    {
      "epoch": 0.5556226851243794,
      "grad_norm": 3.5916316509246826,
      "learning_rate": 4.448989141856393e-05,
      "loss": 1.014,
      "num_input_tokens_seen": 127639088,
      "step": 7932
    },
    {
      "epoch": 0.5556927333701087,
      "grad_norm": 4.151430130004883,
      "learning_rate": 4.4482893169877406e-05,
      "loss": 1.0182,
      "num_input_tokens_seen": 127655240,
      "step": 7933
    },
    {
      "epoch": 0.5557627816158379,
      "grad_norm": 4.024296283721924,
      "learning_rate": 4.4475894921190894e-05,
      "loss": 0.9699,
      "num_input_tokens_seen": 127671624,
      "step": 7934
    },
    {
      "epoch": 0.5558328298615671,
      "grad_norm": 3.977968454360962,
      "learning_rate": 4.446889667250439e-05,
      "loss": 1.1159,
      "num_input_tokens_seen": 127688008,
      "step": 7935
    },
    {
      "epoch": 0.5559028781072964,
      "grad_norm": 3.4853708744049072,
      "learning_rate": 4.4461898423817865e-05,
      "loss": 0.9936,
      "num_input_tokens_seen": 127704352,
      "step": 7936
    },
    {
      "epoch": 0.5559729263530256,
      "grad_norm": 3.8649024963378906,
      "learning_rate": 4.445490017513135e-05,
      "loss": 0.9168,
      "num_input_tokens_seen": 127720488,
      "step": 7937
    },
    {
      "epoch": 0.5560429745987548,
      "grad_norm": 3.8322534561157227,
      "learning_rate": 4.4447901926444835e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 127736280,
      "step": 7938
    },
    {
      "epoch": 0.5561130228444842,
      "grad_norm": 3.2827885150909424,
      "learning_rate": 4.4440903677758324e-05,
      "loss": 0.8871,
      "num_input_tokens_seen": 127752664,
      "step": 7939
    },
    {
      "epoch": 0.5561830710902134,
      "grad_norm": 3.632049560546875,
      "learning_rate": 4.44339054290718e-05,
      "loss": 0.9965,
      "num_input_tokens_seen": 127769048,
      "step": 7940
    },
    {
      "epoch": 0.5562531193359427,
      "grad_norm": 3.9161887168884277,
      "learning_rate": 4.4426907180385294e-05,
      "loss": 1.3189,
      "num_input_tokens_seen": 127784576,
      "step": 7941
    },
    {
      "epoch": 0.5563231675816719,
      "grad_norm": 3.6720736026763916,
      "learning_rate": 4.441990893169878e-05,
      "loss": 1.2914,
      "num_input_tokens_seen": 127800960,
      "step": 7942
    },
    {
      "epoch": 0.5563932158274011,
      "grad_norm": 4.027480602264404,
      "learning_rate": 4.441291068301226e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 127816800,
      "step": 7943
    },
    {
      "epoch": 0.5564632640731304,
      "grad_norm": 5.149120807647705,
      "learning_rate": 4.4405912434325746e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 127832760,
      "step": 7944
    },
    {
      "epoch": 0.5565333123188596,
      "grad_norm": 4.779052734375,
      "learning_rate": 4.439891418563923e-05,
      "loss": 1.0751,
      "num_input_tokens_seen": 127848888,
      "step": 7945
    },
    {
      "epoch": 0.5566033605645888,
      "grad_norm": 4.37583065032959,
      "learning_rate": 4.439191593695272e-05,
      "loss": 1.1268,
      "num_input_tokens_seen": 127865216,
      "step": 7946
    },
    {
      "epoch": 0.5566734088103181,
      "grad_norm": 4.8968963623046875,
      "learning_rate": 4.4384917688266205e-05,
      "loss": 1.139,
      "num_input_tokens_seen": 127881600,
      "step": 7947
    },
    {
      "epoch": 0.5567434570560473,
      "grad_norm": 3.862400770187378,
      "learning_rate": 4.437791943957969e-05,
      "loss": 1.0398,
      "num_input_tokens_seen": 127897424,
      "step": 7948
    },
    {
      "epoch": 0.5568135053017766,
      "grad_norm": 3.653930902481079,
      "learning_rate": 4.4370921190893176e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 127913808,
      "step": 7949
    },
    {
      "epoch": 0.5568835535475058,
      "grad_norm": 4.557390213012695,
      "learning_rate": 4.436392294220665e-05,
      "loss": 1.1048,
      "num_input_tokens_seen": 127930192,
      "step": 7950
    },
    {
      "epoch": 0.556953601793235,
      "grad_norm": 6.2530364990234375,
      "learning_rate": 4.4356924693520146e-05,
      "loss": 1.2491,
      "num_input_tokens_seen": 127946576,
      "step": 7951
    },
    {
      "epoch": 0.5570236500389644,
      "grad_norm": 3.9652791023254395,
      "learning_rate": 4.4349926444833635e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 127962960,
      "step": 7952
    },
    {
      "epoch": 0.5570936982846936,
      "grad_norm": 5.276636600494385,
      "learning_rate": 4.434292819614711e-05,
      "loss": 1.0179,
      "num_input_tokens_seen": 127979016,
      "step": 7953
    },
    {
      "epoch": 0.5571637465304228,
      "grad_norm": 3.445674180984497,
      "learning_rate": 4.43359299474606e-05,
      "loss": 1.033,
      "num_input_tokens_seen": 127995400,
      "step": 7954
    },
    {
      "epoch": 0.5572337947761521,
      "grad_norm": 4.23106575012207,
      "learning_rate": 4.432893169877408e-05,
      "loss": 1.1025,
      "num_input_tokens_seen": 128011784,
      "step": 7955
    },
    {
      "epoch": 0.5573038430218813,
      "grad_norm": 3.795984983444214,
      "learning_rate": 4.432193345008757e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 128027504,
      "step": 7956
    },
    {
      "epoch": 0.5573738912676106,
      "grad_norm": 4.286993980407715,
      "learning_rate": 4.431493520140106e-05,
      "loss": 1.0134,
      "num_input_tokens_seen": 128043832,
      "step": 7957
    },
    {
      "epoch": 0.5574439395133398,
      "grad_norm": 5.160306930541992,
      "learning_rate": 4.430793695271454e-05,
      "loss": 1.1087,
      "num_input_tokens_seen": 128059152,
      "step": 7958
    },
    {
      "epoch": 0.557513987759069,
      "grad_norm": 5.083519458770752,
      "learning_rate": 4.430093870402803e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 128074664,
      "step": 7959
    },
    {
      "epoch": 0.5575840360047983,
      "grad_norm": 3.746807336807251,
      "learning_rate": 4.42939404553415e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 128090984,
      "step": 7960
    },
    {
      "epoch": 0.5576540842505275,
      "grad_norm": 5.527798175811768,
      "learning_rate": 4.4286942206655e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 128107368,
      "step": 7961
    },
    {
      "epoch": 0.5577241324962567,
      "grad_norm": 5.1027116775512695,
      "learning_rate": 4.427994395796849e-05,
      "loss": 1.116,
      "num_input_tokens_seen": 128123752,
      "step": 7962
    },
    {
      "epoch": 0.5577941807419861,
      "grad_norm": 3.7059028148651123,
      "learning_rate": 4.427294570928196e-05,
      "loss": 1.2798,
      "num_input_tokens_seen": 128140136,
      "step": 7963
    },
    {
      "epoch": 0.5578642289877153,
      "grad_norm": 3.936305522918701,
      "learning_rate": 4.426594746059545e-05,
      "loss": 0.88,
      "num_input_tokens_seen": 128156520,
      "step": 7964
    },
    {
      "epoch": 0.5579342772334446,
      "grad_norm": 4.8260369300842285,
      "learning_rate": 4.425894921190893e-05,
      "loss": 0.8951,
      "num_input_tokens_seen": 128172904,
      "step": 7965
    },
    {
      "epoch": 0.5580043254791738,
      "grad_norm": 4.908670425415039,
      "learning_rate": 4.425195096322242e-05,
      "loss": 0.9753,
      "num_input_tokens_seen": 128188672,
      "step": 7966
    },
    {
      "epoch": 0.558074373724903,
      "grad_norm": 4.453614711761475,
      "learning_rate": 4.42449527145359e-05,
      "loss": 1.1764,
      "num_input_tokens_seen": 128205056,
      "step": 7967
    },
    {
      "epoch": 0.5581444219706323,
      "grad_norm": 3.9060006141662598,
      "learning_rate": 4.423795446584939e-05,
      "loss": 1.1539,
      "num_input_tokens_seen": 128221440,
      "step": 7968
    },
    {
      "epoch": 0.5582144702163615,
      "grad_norm": 4.674314498901367,
      "learning_rate": 4.423095621716288e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 128237824,
      "step": 7969
    },
    {
      "epoch": 0.5582845184620908,
      "grad_norm": 5.411530017852783,
      "learning_rate": 4.4223957968476355e-05,
      "loss": 1.1191,
      "num_input_tokens_seen": 128253488,
      "step": 7970
    },
    {
      "epoch": 0.55835456670782,
      "grad_norm": 3.7647812366485596,
      "learning_rate": 4.421695971978985e-05,
      "loss": 1.0836,
      "num_input_tokens_seen": 128269600,
      "step": 7971
    },
    {
      "epoch": 0.5584246149535492,
      "grad_norm": 5.809626579284668,
      "learning_rate": 4.4209961471103325e-05,
      "loss": 1.2228,
      "num_input_tokens_seen": 128285592,
      "step": 7972
    },
    {
      "epoch": 0.5584946631992785,
      "grad_norm": 4.234646320343018,
      "learning_rate": 4.4202963222416814e-05,
      "loss": 1.2126,
      "num_input_tokens_seen": 128301976,
      "step": 7973
    },
    {
      "epoch": 0.5585647114450077,
      "grad_norm": 3.390713930130005,
      "learning_rate": 4.41959649737303e-05,
      "loss": 1.0575,
      "num_input_tokens_seen": 128318360,
      "step": 7974
    },
    {
      "epoch": 0.558634759690737,
      "grad_norm": 3.582789182662964,
      "learning_rate": 4.4188966725043784e-05,
      "loss": 1.0472,
      "num_input_tokens_seen": 128334744,
      "step": 7975
    },
    {
      "epoch": 0.5587048079364663,
      "grad_norm": 4.052979946136475,
      "learning_rate": 4.418196847635727e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 128350920,
      "step": 7976
    },
    {
      "epoch": 0.5587748561821955,
      "grad_norm": 4.101496696472168,
      "learning_rate": 4.4174970227670755e-05,
      "loss": 1.0432,
      "num_input_tokens_seen": 128367144,
      "step": 7977
    },
    {
      "epoch": 0.5588449044279248,
      "grad_norm": 4.708171367645264,
      "learning_rate": 4.416797197898424e-05,
      "loss": 0.9654,
      "num_input_tokens_seen": 128381272,
      "step": 7978
    },
    {
      "epoch": 0.558914952673654,
      "grad_norm": 4.02223539352417,
      "learning_rate": 4.416097373029773e-05,
      "loss": 1.1984,
      "num_input_tokens_seen": 128397160,
      "step": 7979
    },
    {
      "epoch": 0.5589850009193832,
      "grad_norm": 4.594113349914551,
      "learning_rate": 4.415397548161121e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 128413248,
      "step": 7980
    },
    {
      "epoch": 0.5590550491651125,
      "grad_norm": 4.255110263824463,
      "learning_rate": 4.41469772329247e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 128429632,
      "step": 7981
    },
    {
      "epoch": 0.5591250974108417,
      "grad_norm": 4.862103462219238,
      "learning_rate": 4.413997898423818e-05,
      "loss": 0.8901,
      "num_input_tokens_seen": 128446016,
      "step": 7982
    },
    {
      "epoch": 0.5591951456565709,
      "grad_norm": 4.902801036834717,
      "learning_rate": 4.4132980735551666e-05,
      "loss": 1.1788,
      "num_input_tokens_seen": 128461344,
      "step": 7983
    },
    {
      "epoch": 0.5592651939023002,
      "grad_norm": 3.762756824493408,
      "learning_rate": 4.4125982486865154e-05,
      "loss": 1.2135,
      "num_input_tokens_seen": 128477080,
      "step": 7984
    },
    {
      "epoch": 0.5593352421480294,
      "grad_norm": 4.081719398498535,
      "learning_rate": 4.4118984238178636e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 128493112,
      "step": 7985
    },
    {
      "epoch": 0.5594052903937587,
      "grad_norm": 4.764441013336182,
      "learning_rate": 4.4111985989492125e-05,
      "loss": 1.1384,
      "num_input_tokens_seen": 128508816,
      "step": 7986
    },
    {
      "epoch": 0.559475338639488,
      "grad_norm": 3.709524631500244,
      "learning_rate": 4.410498774080561e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 128525200,
      "step": 7987
    },
    {
      "epoch": 0.5595453868852172,
      "grad_norm": 3.648374557495117,
      "learning_rate": 4.4097989492119095e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 128541584,
      "step": 7988
    },
    {
      "epoch": 0.5596154351309465,
      "grad_norm": 6.713209629058838,
      "learning_rate": 4.4090991243432584e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 128557048,
      "step": 7989
    },
    {
      "epoch": 0.5596854833766757,
      "grad_norm": 5.627123832702637,
      "learning_rate": 4.408399299474606e-05,
      "loss": 0.8656,
      "num_input_tokens_seen": 128572232,
      "step": 7990
    },
    {
      "epoch": 0.5597555316224049,
      "grad_norm": 3.928884267807007,
      "learning_rate": 4.4076994746059554e-05,
      "loss": 0.8768,
      "num_input_tokens_seen": 128588616,
      "step": 7991
    },
    {
      "epoch": 0.5598255798681342,
      "grad_norm": 4.144782543182373,
      "learning_rate": 4.406999649737303e-05,
      "loss": 1.2362,
      "num_input_tokens_seen": 128604824,
      "step": 7992
    },
    {
      "epoch": 0.5598956281138634,
      "grad_norm": 3.7674214839935303,
      "learning_rate": 4.406299824868652e-05,
      "loss": 1.0108,
      "num_input_tokens_seen": 128621208,
      "step": 7993
    },
    {
      "epoch": 0.5599656763595927,
      "grad_norm": 3.4347217082977295,
      "learning_rate": 4.4056e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 128637592,
      "step": 7994
    },
    {
      "epoch": 0.5600357246053219,
      "grad_norm": 4.309778690338135,
      "learning_rate": 4.404900175131349e-05,
      "loss": 1.1468,
      "num_input_tokens_seen": 128653976,
      "step": 7995
    },
    {
      "epoch": 0.5601057728510511,
      "grad_norm": 3.567622423171997,
      "learning_rate": 4.404200350262698e-05,
      "loss": 0.9106,
      "num_input_tokens_seen": 128670360,
      "step": 7996
    },
    {
      "epoch": 0.5601758210967804,
      "grad_norm": 5.184731483459473,
      "learning_rate": 4.403500525394046e-05,
      "loss": 0.8277,
      "num_input_tokens_seen": 128684744,
      "step": 7997
    },
    {
      "epoch": 0.5602458693425096,
      "grad_norm": 4.807730674743652,
      "learning_rate": 4.402800700525395e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 128701128,
      "step": 7998
    },
    {
      "epoch": 0.560315917588239,
      "grad_norm": 6.368520736694336,
      "learning_rate": 4.402100875656742e-05,
      "loss": 1.3107,
      "num_input_tokens_seen": 128717288,
      "step": 7999
    },
    {
      "epoch": 0.5603859658339682,
      "grad_norm": 3.997955083847046,
      "learning_rate": 4.401401050788091e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 128733328,
      "step": 8000
    },
    {
      "epoch": 0.5603859658339682,
      "eval_loss": 1.1226634979248047,
      "eval_runtime": 0.2085,
      "eval_samples_per_second": 4.797,
      "eval_steps_per_second": 4.797,
      "num_input_tokens_seen": 128733328,
      "step": 8000
    },
    {
      "epoch": 0.5604560140796974,
      "grad_norm": 7.1005635261535645,
      "learning_rate": 4.4007012259194406e-05,
      "loss": 1.0116,
      "num_input_tokens_seen": 128749712,
      "step": 8001
    },
    {
      "epoch": 0.5605260623254267,
      "grad_norm": 3.9482359886169434,
      "learning_rate": 4.400001401050788e-05,
      "loss": 0.8402,
      "num_input_tokens_seen": 128765408,
      "step": 8002
    },
    {
      "epoch": 0.5605961105711559,
      "grad_norm": 5.755213260650635,
      "learning_rate": 4.399301576182137e-05,
      "loss": 1.0476,
      "num_input_tokens_seen": 128781256,
      "step": 8003
    },
    {
      "epoch": 0.5606661588168851,
      "grad_norm": 4.391782760620117,
      "learning_rate": 4.398601751313485e-05,
      "loss": 0.9619,
      "num_input_tokens_seen": 128796952,
      "step": 8004
    },
    {
      "epoch": 0.5607362070626144,
      "grad_norm": 4.010289192199707,
      "learning_rate": 4.397901926444834e-05,
      "loss": 0.963,
      "num_input_tokens_seen": 128813160,
      "step": 8005
    },
    {
      "epoch": 0.5608062553083436,
      "grad_norm": 4.009171485900879,
      "learning_rate": 4.397202101576183e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 128828712,
      "step": 8006
    },
    {
      "epoch": 0.5608763035540729,
      "grad_norm": 3.7310292720794678,
      "learning_rate": 4.396502276707531e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 128845096,
      "step": 8007
    },
    {
      "epoch": 0.5609463517998021,
      "grad_norm": 4.350787162780762,
      "learning_rate": 4.39580245183888e-05,
      "loss": 1.1524,
      "num_input_tokens_seen": 128861480,
      "step": 8008
    },
    {
      "epoch": 0.5610164000455313,
      "grad_norm": 3.7701098918914795,
      "learning_rate": 4.3951026269702274e-05,
      "loss": 1.0272,
      "num_input_tokens_seen": 128877160,
      "step": 8009
    },
    {
      "epoch": 0.5610864482912606,
      "grad_norm": 3.878599166870117,
      "learning_rate": 4.394402802101576e-05,
      "loss": 0.9247,
      "num_input_tokens_seen": 128893544,
      "step": 8010
    },
    {
      "epoch": 0.5611564965369898,
      "grad_norm": 4.865396499633789,
      "learning_rate": 4.393702977232926e-05,
      "loss": 1.2614,
      "num_input_tokens_seen": 128909496,
      "step": 8011
    },
    {
      "epoch": 0.561226544782719,
      "grad_norm": 3.691734552383423,
      "learning_rate": 4.393003152364273e-05,
      "loss": 1.1862,
      "num_input_tokens_seen": 128925440,
      "step": 8012
    },
    {
      "epoch": 0.5612965930284484,
      "grad_norm": 3.890925884246826,
      "learning_rate": 4.392303327495622e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 128941824,
      "step": 8013
    },
    {
      "epoch": 0.5613666412741776,
      "grad_norm": 5.752060413360596,
      "learning_rate": 4.3916035026269704e-05,
      "loss": 1.1932,
      "num_input_tokens_seen": 128955672,
      "step": 8014
    },
    {
      "epoch": 0.5614366895199069,
      "grad_norm": 3.931739091873169,
      "learning_rate": 4.390903677758319e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 128972056,
      "step": 8015
    },
    {
      "epoch": 0.5615067377656361,
      "grad_norm": 5.384754657745361,
      "learning_rate": 4.390203852889668e-05,
      "loss": 1.0261,
      "num_input_tokens_seen": 128988440,
      "step": 8016
    },
    {
      "epoch": 0.5615767860113653,
      "grad_norm": 4.624659061431885,
      "learning_rate": 4.389504028021016e-05,
      "loss": 0.8866,
      "num_input_tokens_seen": 129004528,
      "step": 8017
    },
    {
      "epoch": 0.5616468342570946,
      "grad_norm": 3.9661507606506348,
      "learning_rate": 4.388804203152365e-05,
      "loss": 1.1405,
      "num_input_tokens_seen": 129020912,
      "step": 8018
    },
    {
      "epoch": 0.5617168825028238,
      "grad_norm": 3.5319230556488037,
      "learning_rate": 4.3881043782837126e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 129037296,
      "step": 8019
    },
    {
      "epoch": 0.561786930748553,
      "grad_norm": 4.236444473266602,
      "learning_rate": 4.3874045534150615e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 129053680,
      "step": 8020
    },
    {
      "epoch": 0.5618569789942823,
      "grad_norm": 3.69328236579895,
      "learning_rate": 4.38670472854641e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 129070064,
      "step": 8021
    },
    {
      "epoch": 0.5619270272400115,
      "grad_norm": 5.341061115264893,
      "learning_rate": 4.3860049036777585e-05,
      "loss": 0.9455,
      "num_input_tokens_seen": 129086224,
      "step": 8022
    },
    {
      "epoch": 0.5619970754857408,
      "grad_norm": 4.005245208740234,
      "learning_rate": 4.3853050788091074e-05,
      "loss": 0.9985,
      "num_input_tokens_seen": 129102608,
      "step": 8023
    },
    {
      "epoch": 0.56206712373147,
      "grad_norm": 4.377266883850098,
      "learning_rate": 4.3846052539404556e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 129118992,
      "step": 8024
    },
    {
      "epoch": 0.5621371719771993,
      "grad_norm": 3.7815823554992676,
      "learning_rate": 4.3839054290718044e-05,
      "loss": 1.192,
      "num_input_tokens_seen": 129135376,
      "step": 8025
    },
    {
      "epoch": 0.5622072202229286,
      "grad_norm": 4.8340744972229,
      "learning_rate": 4.383205604203152e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 129151352,
      "step": 8026
    },
    {
      "epoch": 0.5622772684686578,
      "grad_norm": 5.007735252380371,
      "learning_rate": 4.3825057793345015e-05,
      "loss": 1.094,
      "num_input_tokens_seen": 129166992,
      "step": 8027
    },
    {
      "epoch": 0.562347316714387,
      "grad_norm": 5.447760105133057,
      "learning_rate": 4.38180595446585e-05,
      "loss": 1.2057,
      "num_input_tokens_seen": 129183376,
      "step": 8028
    },
    {
      "epoch": 0.5624173649601163,
      "grad_norm": 4.995473861694336,
      "learning_rate": 4.381106129597198e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 129198816,
      "step": 8029
    },
    {
      "epoch": 0.5624874132058455,
      "grad_norm": 4.708920955657959,
      "learning_rate": 4.380406304728547e-05,
      "loss": 0.9984,
      "num_input_tokens_seen": 129215200,
      "step": 8030
    },
    {
      "epoch": 0.5625574614515748,
      "grad_norm": 3.8864386081695557,
      "learning_rate": 4.379706479859895e-05,
      "loss": 0.734,
      "num_input_tokens_seen": 129231504,
      "step": 8031
    },
    {
      "epoch": 0.562627509697304,
      "grad_norm": 3.467696189880371,
      "learning_rate": 4.379006654991244e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 129247624,
      "step": 8032
    },
    {
      "epoch": 0.5626975579430332,
      "grad_norm": 4.07413387298584,
      "learning_rate": 4.3783068301225926e-05,
      "loss": 1.0158,
      "num_input_tokens_seen": 129264008,
      "step": 8033
    },
    {
      "epoch": 0.5627676061887625,
      "grad_norm": 4.135556697845459,
      "learning_rate": 4.377607005253941e-05,
      "loss": 1.1483,
      "num_input_tokens_seen": 129279136,
      "step": 8034
    },
    {
      "epoch": 0.5628376544344917,
      "grad_norm": 4.153659820556641,
      "learning_rate": 4.3769071803852896e-05,
      "loss": 0.9118,
      "num_input_tokens_seen": 129294192,
      "step": 8035
    },
    {
      "epoch": 0.5629077026802211,
      "grad_norm": 3.6717982292175293,
      "learning_rate": 4.376207355516637e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 129310312,
      "step": 8036
    },
    {
      "epoch": 0.5629777509259503,
      "grad_norm": 4.013020038604736,
      "learning_rate": 4.375507530647987e-05,
      "loss": 1.1211,
      "num_input_tokens_seen": 129326696,
      "step": 8037
    },
    {
      "epoch": 0.5630477991716795,
      "grad_norm": 4.093492031097412,
      "learning_rate": 4.3748077057793355e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 129343080,
      "step": 8038
    },
    {
      "epoch": 0.5631178474174088,
      "grad_norm": 5.1094512939453125,
      "learning_rate": 4.374107880910683e-05,
      "loss": 1.1122,
      "num_input_tokens_seen": 129359232,
      "step": 8039
    },
    {
      "epoch": 0.563187895663138,
      "grad_norm": 4.3890275955200195,
      "learning_rate": 4.373408056042032e-05,
      "loss": 1.2691,
      "num_input_tokens_seen": 129374992,
      "step": 8040
    },
    {
      "epoch": 0.5632579439088672,
      "grad_norm": 3.4645395278930664,
      "learning_rate": 4.37270823117338e-05,
      "loss": 0.945,
      "num_input_tokens_seen": 129390728,
      "step": 8041
    },
    {
      "epoch": 0.5633279921545965,
      "grad_norm": 3.5011980533599854,
      "learning_rate": 4.372008406304729e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 129406616,
      "step": 8042
    },
    {
      "epoch": 0.5633980404003257,
      "grad_norm": 3.694739580154419,
      "learning_rate": 4.371308581436078e-05,
      "loss": 1.0331,
      "num_input_tokens_seen": 129423000,
      "step": 8043
    },
    {
      "epoch": 0.563468088646055,
      "grad_norm": 4.7144293785095215,
      "learning_rate": 4.370608756567426e-05,
      "loss": 0.9954,
      "num_input_tokens_seen": 129438048,
      "step": 8044
    },
    {
      "epoch": 0.5635381368917842,
      "grad_norm": 3.9834823608398438,
      "learning_rate": 4.369908931698775e-05,
      "loss": 1.1571,
      "num_input_tokens_seen": 129454432,
      "step": 8045
    },
    {
      "epoch": 0.5636081851375134,
      "grad_norm": 5.334307670593262,
      "learning_rate": 4.3692091068301223e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 129470816,
      "step": 8046
    },
    {
      "epoch": 0.5636782333832427,
      "grad_norm": 4.415291786193848,
      "learning_rate": 4.368509281961472e-05,
      "loss": 1.2554,
      "num_input_tokens_seen": 129487200,
      "step": 8047
    },
    {
      "epoch": 0.563748281628972,
      "grad_norm": 5.6152262687683105,
      "learning_rate": 4.3678094570928194e-05,
      "loss": 1.0074,
      "num_input_tokens_seen": 129503584,
      "step": 8048
    },
    {
      "epoch": 0.5638183298747012,
      "grad_norm": 5.171108245849609,
      "learning_rate": 4.367109632224168e-05,
      "loss": 0.9731,
      "num_input_tokens_seen": 129519968,
      "step": 8049
    },
    {
      "epoch": 0.5638883781204305,
      "grad_norm": 4.773746013641357,
      "learning_rate": 4.366409807355517e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 129535432,
      "step": 8050
    },
    {
      "epoch": 0.5639584263661597,
      "grad_norm": 4.129162311553955,
      "learning_rate": 4.365709982486865e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 129550880,
      "step": 8051
    },
    {
      "epoch": 0.564028474611889,
      "grad_norm": 4.851277828216553,
      "learning_rate": 4.365010157618214e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 129565592,
      "step": 8052
    },
    {
      "epoch": 0.5640985228576182,
      "grad_norm": 4.487455368041992,
      "learning_rate": 4.364310332749562e-05,
      "loss": 1.2058,
      "num_input_tokens_seen": 129581976,
      "step": 8053
    },
    {
      "epoch": 0.5641685711033474,
      "grad_norm": 4.226871013641357,
      "learning_rate": 4.363610507880911e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 129598360,
      "step": 8054
    },
    {
      "epoch": 0.5642386193490767,
      "grad_norm": 4.204067230224609,
      "learning_rate": 4.36291068301226e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 129614232,
      "step": 8055
    },
    {
      "epoch": 0.5643086675948059,
      "grad_norm": 3.792025566101074,
      "learning_rate": 4.3622108581436075e-05,
      "loss": 1.2165,
      "num_input_tokens_seen": 129630616,
      "step": 8056
    },
    {
      "epoch": 0.5643787158405351,
      "grad_norm": 3.966076135635376,
      "learning_rate": 4.361511033274957e-05,
      "loss": 1.1306,
      "num_input_tokens_seen": 129647000,
      "step": 8057
    },
    {
      "epoch": 0.5644487640862644,
      "grad_norm": 5.704814910888672,
      "learning_rate": 4.3608112084063046e-05,
      "loss": 1.0289,
      "num_input_tokens_seen": 129663384,
      "step": 8058
    },
    {
      "epoch": 0.5645188123319936,
      "grad_norm": 4.2781081199646,
      "learning_rate": 4.3601113835376534e-05,
      "loss": 1.173,
      "num_input_tokens_seen": 129679768,
      "step": 8059
    },
    {
      "epoch": 0.564588860577723,
      "grad_norm": 3.806344985961914,
      "learning_rate": 4.359411558669002e-05,
      "loss": 1.1536,
      "num_input_tokens_seen": 129696152,
      "step": 8060
    },
    {
      "epoch": 0.5646589088234522,
      "grad_norm": 4.551416873931885,
      "learning_rate": 4.3587117338003505e-05,
      "loss": 1.078,
      "num_input_tokens_seen": 129711184,
      "step": 8061
    },
    {
      "epoch": 0.5647289570691814,
      "grad_norm": 3.6475703716278076,
      "learning_rate": 4.3580119089316993e-05,
      "loss": 1.0125,
      "num_input_tokens_seen": 129727568,
      "step": 8062
    },
    {
      "epoch": 0.5647990053149107,
      "grad_norm": 4.113941192626953,
      "learning_rate": 4.3573120840630475e-05,
      "loss": 1.0104,
      "num_input_tokens_seen": 129743952,
      "step": 8063
    },
    {
      "epoch": 0.5648690535606399,
      "grad_norm": 3.9017693996429443,
      "learning_rate": 4.3566122591943964e-05,
      "loss": 0.9712,
      "num_input_tokens_seen": 129760336,
      "step": 8064
    },
    {
      "epoch": 0.5649391018063691,
      "grad_norm": 4.13060188293457,
      "learning_rate": 4.355912434325745e-05,
      "loss": 0.9723,
      "num_input_tokens_seen": 129776144,
      "step": 8065
    },
    {
      "epoch": 0.5650091500520984,
      "grad_norm": 4.518004417419434,
      "learning_rate": 4.355212609457093e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 129792000,
      "step": 8066
    },
    {
      "epoch": 0.5650791982978276,
      "grad_norm": 4.141806602478027,
      "learning_rate": 4.354512784588442e-05,
      "loss": 1.0458,
      "num_input_tokens_seen": 129808384,
      "step": 8067
    },
    {
      "epoch": 0.5651492465435569,
      "grad_norm": 4.177087783813477,
      "learning_rate": 4.35381295971979e-05,
      "loss": 1.1853,
      "num_input_tokens_seen": 129824256,
      "step": 8068
    },
    {
      "epoch": 0.5652192947892861,
      "grad_norm": 3.939929485321045,
      "learning_rate": 4.3531131348511386e-05,
      "loss": 0.995,
      "num_input_tokens_seen": 129840640,
      "step": 8069
    },
    {
      "epoch": 0.5652893430350153,
      "grad_norm": 4.335320472717285,
      "learning_rate": 4.3524133099824875e-05,
      "loss": 1.048,
      "num_input_tokens_seen": 129856952,
      "step": 8070
    },
    {
      "epoch": 0.5653593912807446,
      "grad_norm": 6.816842079162598,
      "learning_rate": 4.351713485113836e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 129872552,
      "step": 8071
    },
    {
      "epoch": 0.5654294395264738,
      "grad_norm": 4.088340759277344,
      "learning_rate": 4.3510136602451845e-05,
      "loss": 1.0941,
      "num_input_tokens_seen": 129888936,
      "step": 8072
    },
    {
      "epoch": 0.5654994877722032,
      "grad_norm": 3.65846586227417,
      "learning_rate": 4.350313835376533e-05,
      "loss": 1.1355,
      "num_input_tokens_seen": 129905320,
      "step": 8073
    },
    {
      "epoch": 0.5655695360179324,
      "grad_norm": 3.5550310611724854,
      "learning_rate": 4.3496140105078816e-05,
      "loss": 1.0121,
      "num_input_tokens_seen": 129921704,
      "step": 8074
    },
    {
      "epoch": 0.5656395842636616,
      "grad_norm": 4.011558532714844,
      "learning_rate": 4.348914185639229e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 129937968,
      "step": 8075
    },
    {
      "epoch": 0.5657096325093909,
      "grad_norm": 3.63883638381958,
      "learning_rate": 4.348214360770578e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 129954352,
      "step": 8076
    },
    {
      "epoch": 0.5657796807551201,
      "grad_norm": 3.9487674236297607,
      "learning_rate": 4.3475145359019275e-05,
      "loss": 1.1023,
      "num_input_tokens_seen": 129970528,
      "step": 8077
    },
    {
      "epoch": 0.5658497290008493,
      "grad_norm": 4.0052266120910645,
      "learning_rate": 4.346814711033275e-05,
      "loss": 1.0462,
      "num_input_tokens_seen": 129985568,
      "step": 8078
    },
    {
      "epoch": 0.5659197772465786,
      "grad_norm": 4.147900581359863,
      "learning_rate": 4.346114886164624e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 130000832,
      "step": 8079
    },
    {
      "epoch": 0.5659898254923078,
      "grad_norm": 3.9152534008026123,
      "learning_rate": 4.345415061295972e-05,
      "loss": 1.1202,
      "num_input_tokens_seen": 130017216,
      "step": 8080
    },
    {
      "epoch": 0.5660598737380371,
      "grad_norm": 5.270138263702393,
      "learning_rate": 4.344715236427321e-05,
      "loss": 1.1368,
      "num_input_tokens_seen": 130033600,
      "step": 8081
    },
    {
      "epoch": 0.5661299219837663,
      "grad_norm": 3.7850892543792725,
      "learning_rate": 4.34401541155867e-05,
      "loss": 1.0064,
      "num_input_tokens_seen": 130049576,
      "step": 8082
    },
    {
      "epoch": 0.5661999702294955,
      "grad_norm": 3.8164186477661133,
      "learning_rate": 4.343315586690018e-05,
      "loss": 0.9948,
      "num_input_tokens_seen": 130065960,
      "step": 8083
    },
    {
      "epoch": 0.5662700184752248,
      "grad_norm": 4.439040660858154,
      "learning_rate": 4.342615761821367e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 130082344,
      "step": 8084
    },
    {
      "epoch": 0.566340066720954,
      "grad_norm": 4.9009599685668945,
      "learning_rate": 4.341915936952714e-05,
      "loss": 1.2654,
      "num_input_tokens_seen": 130098432,
      "step": 8085
    },
    {
      "epoch": 0.5664101149666833,
      "grad_norm": 3.7141706943511963,
      "learning_rate": 4.341216112084063e-05,
      "loss": 0.9878,
      "num_input_tokens_seen": 130114816,
      "step": 8086
    },
    {
      "epoch": 0.5664801632124126,
      "grad_norm": 3.661966323852539,
      "learning_rate": 4.340516287215413e-05,
      "loss": 0.9867,
      "num_input_tokens_seen": 130131112,
      "step": 8087
    },
    {
      "epoch": 0.5665502114581418,
      "grad_norm": 5.044811725616455,
      "learning_rate": 4.33981646234676e-05,
      "loss": 0.9719,
      "num_input_tokens_seen": 130147496,
      "step": 8088
    },
    {
      "epoch": 0.5666202597038711,
      "grad_norm": 5.159415245056152,
      "learning_rate": 4.339116637478109e-05,
      "loss": 1.1301,
      "num_input_tokens_seen": 130163880,
      "step": 8089
    },
    {
      "epoch": 0.5666903079496003,
      "grad_norm": 3.651123523712158,
      "learning_rate": 4.338416812609457e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 130179888,
      "step": 8090
    },
    {
      "epoch": 0.5667603561953295,
      "grad_norm": 4.067645072937012,
      "learning_rate": 4.337716987740806e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 130196272,
      "step": 8091
    },
    {
      "epoch": 0.5668304044410588,
      "grad_norm": 3.646764039993286,
      "learning_rate": 4.337017162872155e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 130212656,
      "step": 8092
    },
    {
      "epoch": 0.566900452686788,
      "grad_norm": 5.098438739776611,
      "learning_rate": 4.336317338003503e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 130228608,
      "step": 8093
    },
    {
      "epoch": 0.5669705009325172,
      "grad_norm": 3.536712408065796,
      "learning_rate": 4.335617513134852e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 130244768,
      "step": 8094
    },
    {
      "epoch": 0.5670405491782465,
      "grad_norm": 4.888411998748779,
      "learning_rate": 4.3349176882661995e-05,
      "loss": 1.1445,
      "num_input_tokens_seen": 130260936,
      "step": 8095
    },
    {
      "epoch": 0.5671105974239757,
      "grad_norm": 4.441121578216553,
      "learning_rate": 4.3342178633975484e-05,
      "loss": 1.1489,
      "num_input_tokens_seen": 130277320,
      "step": 8096
    },
    {
      "epoch": 0.567180645669705,
      "grad_norm": 4.9977545738220215,
      "learning_rate": 4.333518038528898e-05,
      "loss": 1.0849,
      "num_input_tokens_seen": 130292984,
      "step": 8097
    },
    {
      "epoch": 0.5672506939154343,
      "grad_norm": 3.602999448776245,
      "learning_rate": 4.3328182136602454e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 130308376,
      "step": 8098
    },
    {
      "epoch": 0.5673207421611635,
      "grad_norm": 4.618570804595947,
      "learning_rate": 4.332118388791594e-05,
      "loss": 1.1175,
      "num_input_tokens_seen": 130324440,
      "step": 8099
    },
    {
      "epoch": 0.5673907904068928,
      "grad_norm": 4.875861644744873,
      "learning_rate": 4.3314185639229424e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 130340160,
      "step": 8100
    },
    {
      "epoch": 0.567460838652622,
      "grad_norm": 3.3949697017669678,
      "learning_rate": 4.330718739054291e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 130355984,
      "step": 8101
    },
    {
      "epoch": 0.5675308868983513,
      "grad_norm": 5.05001163482666,
      "learning_rate": 4.330018914185639e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 130372368,
      "step": 8102
    },
    {
      "epoch": 0.5676009351440805,
      "grad_norm": 3.913780689239502,
      "learning_rate": 4.329319089316988e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 130388752,
      "step": 8103
    },
    {
      "epoch": 0.5676709833898097,
      "grad_norm": 3.778477668762207,
      "learning_rate": 4.328619264448337e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 130405136,
      "step": 8104
    },
    {
      "epoch": 0.567741031635539,
      "grad_norm": 4.065064430236816,
      "learning_rate": 4.327919439579685e-05,
      "loss": 1.1245,
      "num_input_tokens_seen": 130421384,
      "step": 8105
    },
    {
      "epoch": 0.5678110798812682,
      "grad_norm": 3.609297037124634,
      "learning_rate": 4.3272196147110336e-05,
      "loss": 1.0434,
      "num_input_tokens_seen": 130437768,
      "step": 8106
    },
    {
      "epoch": 0.5678811281269974,
      "grad_norm": 4.521945476531982,
      "learning_rate": 4.326519789842382e-05,
      "loss": 1.0285,
      "num_input_tokens_seen": 130452640,
      "step": 8107
    },
    {
      "epoch": 0.5679511763727267,
      "grad_norm": 4.647178649902344,
      "learning_rate": 4.3258199649737306e-05,
      "loss": 1.1492,
      "num_input_tokens_seen": 130469024,
      "step": 8108
    },
    {
      "epoch": 0.5680212246184559,
      "grad_norm": 3.594043254852295,
      "learning_rate": 4.3251201401050795e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 130484648,
      "step": 8109
    },
    {
      "epoch": 0.5680912728641853,
      "grad_norm": 4.731708526611328,
      "learning_rate": 4.3244203152364276e-05,
      "loss": 1.3175,
      "num_input_tokens_seen": 130500024,
      "step": 8110
    },
    {
      "epoch": 0.5681613211099145,
      "grad_norm": 4.488846302032471,
      "learning_rate": 4.3237204903677765e-05,
      "loss": 1.0714,
      "num_input_tokens_seen": 130516408,
      "step": 8111
    },
    {
      "epoch": 0.5682313693556437,
      "grad_norm": 4.828577995300293,
      "learning_rate": 4.323020665499124e-05,
      "loss": 1.1976,
      "num_input_tokens_seen": 130532752,
      "step": 8112
    },
    {
      "epoch": 0.568301417601373,
      "grad_norm": 5.491752624511719,
      "learning_rate": 4.3223208406304735e-05,
      "loss": 1.249,
      "num_input_tokens_seen": 130549136,
      "step": 8113
    },
    {
      "epoch": 0.5683714658471022,
      "grad_norm": 3.4913344383239746,
      "learning_rate": 4.3216210157618224e-05,
      "loss": 1.0322,
      "num_input_tokens_seen": 130565448,
      "step": 8114
    },
    {
      "epoch": 0.5684415140928314,
      "grad_norm": 3.519788980484009,
      "learning_rate": 4.32092119089317e-05,
      "loss": 0.8711,
      "num_input_tokens_seen": 130581832,
      "step": 8115
    },
    {
      "epoch": 0.5685115623385607,
      "grad_norm": 4.691020488739014,
      "learning_rate": 4.320221366024519e-05,
      "loss": 1.1926,
      "num_input_tokens_seen": 130597608,
      "step": 8116
    },
    {
      "epoch": 0.5685816105842899,
      "grad_norm": 4.807389736175537,
      "learning_rate": 4.319521541155867e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 130613992,
      "step": 8117
    },
    {
      "epoch": 0.5686516588300192,
      "grad_norm": 4.060730457305908,
      "learning_rate": 4.318821716287216e-05,
      "loss": 1.2357,
      "num_input_tokens_seen": 130630376,
      "step": 8118
    },
    {
      "epoch": 0.5687217070757484,
      "grad_norm": 4.58277702331543,
      "learning_rate": 4.3181218914185647e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 130646112,
      "step": 8119
    },
    {
      "epoch": 0.5687917553214776,
      "grad_norm": 4.290525913238525,
      "learning_rate": 4.317422066549913e-05,
      "loss": 1.177,
      "num_input_tokens_seen": 130662496,
      "step": 8120
    },
    {
      "epoch": 0.568861803567207,
      "grad_norm": 4.339941501617432,
      "learning_rate": 4.316722241681262e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 130678880,
      "step": 8121
    },
    {
      "epoch": 0.5689318518129362,
      "grad_norm": 7.363986492156982,
      "learning_rate": 4.316022416812609e-05,
      "loss": 1.0718,
      "num_input_tokens_seen": 130693176,
      "step": 8122
    },
    {
      "epoch": 0.5690019000586654,
      "grad_norm": 4.054739475250244,
      "learning_rate": 4.315322591943959e-05,
      "loss": 1.147,
      "num_input_tokens_seen": 130709152,
      "step": 8123
    },
    {
      "epoch": 0.5690719483043947,
      "grad_norm": 3.5982446670532227,
      "learning_rate": 4.3146227670753076e-05,
      "loss": 0.9578,
      "num_input_tokens_seen": 130725424,
      "step": 8124
    },
    {
      "epoch": 0.5691419965501239,
      "grad_norm": 4.437469005584717,
      "learning_rate": 4.313922942206655e-05,
      "loss": 1.0498,
      "num_input_tokens_seen": 130740184,
      "step": 8125
    },
    {
      "epoch": 0.5692120447958532,
      "grad_norm": 3.761885643005371,
      "learning_rate": 4.313223117338004e-05,
      "loss": 1.0214,
      "num_input_tokens_seen": 130756568,
      "step": 8126
    },
    {
      "epoch": 0.5692820930415824,
      "grad_norm": 4.0984416007995605,
      "learning_rate": 4.312523292469352e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 130772248,
      "step": 8127
    },
    {
      "epoch": 0.5693521412873116,
      "grad_norm": 3.3904314041137695,
      "learning_rate": 4.311823467600701e-05,
      "loss": 0.9048,
      "num_input_tokens_seen": 130788632,
      "step": 8128
    },
    {
      "epoch": 0.5694221895330409,
      "grad_norm": 4.1015214920043945,
      "learning_rate": 4.311123642732049e-05,
      "loss": 1.0574,
      "num_input_tokens_seen": 130805016,
      "step": 8129
    },
    {
      "epoch": 0.5694922377787701,
      "grad_norm": 4.696201801300049,
      "learning_rate": 4.310423817863398e-05,
      "loss": 1.0341,
      "num_input_tokens_seen": 130821160,
      "step": 8130
    },
    {
      "epoch": 0.5695622860244993,
      "grad_norm": 4.361652374267578,
      "learning_rate": 4.309723992994747e-05,
      "loss": 0.9995,
      "num_input_tokens_seen": 130835640,
      "step": 8131
    },
    {
      "epoch": 0.5696323342702286,
      "grad_norm": 3.435509204864502,
      "learning_rate": 4.3090241681260944e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 130851176,
      "step": 8132
    },
    {
      "epoch": 0.5697023825159578,
      "grad_norm": 5.87947416305542,
      "learning_rate": 4.308324343257444e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 130866568,
      "step": 8133
    },
    {
      "epoch": 0.5697724307616872,
      "grad_norm": 3.353489398956299,
      "learning_rate": 4.3076245183887914e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 130882536,
      "step": 8134
    },
    {
      "epoch": 0.5698424790074164,
      "grad_norm": 3.4409422874450684,
      "learning_rate": 4.30692469352014e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 130898848,
      "step": 8135
    },
    {
      "epoch": 0.5699125272531456,
      "grad_norm": 4.050583362579346,
      "learning_rate": 4.306224868651489e-05,
      "loss": 1.0886,
      "num_input_tokens_seen": 130915224,
      "step": 8136
    },
    {
      "epoch": 0.5699825754988749,
      "grad_norm": 3.630647659301758,
      "learning_rate": 4.3055250437828373e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 130931608,
      "step": 8137
    },
    {
      "epoch": 0.5700526237446041,
      "grad_norm": 3.3654730319976807,
      "learning_rate": 4.304825218914186e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 130947992,
      "step": 8138
    },
    {
      "epoch": 0.5701226719903334,
      "grad_norm": 4.898937702178955,
      "learning_rate": 4.3041253940455344e-05,
      "loss": 1.0703,
      "num_input_tokens_seen": 130963560,
      "step": 8139
    },
    {
      "epoch": 0.5701927202360626,
      "grad_norm": 3.9583377838134766,
      "learning_rate": 4.303425569176883e-05,
      "loss": 0.9444,
      "num_input_tokens_seen": 130978024,
      "step": 8140
    },
    {
      "epoch": 0.5702627684817918,
      "grad_norm": 3.5538089275360107,
      "learning_rate": 4.302725744308232e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 130993672,
      "step": 8141
    },
    {
      "epoch": 0.5703328167275211,
      "grad_norm": 3.731968879699707,
      "learning_rate": 4.3020259194395796e-05,
      "loss": 1.0219,
      "num_input_tokens_seen": 131010056,
      "step": 8142
    },
    {
      "epoch": 0.5704028649732503,
      "grad_norm": 4.178092002868652,
      "learning_rate": 4.301326094570929e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 131026440,
      "step": 8143
    },
    {
      "epoch": 0.5704729132189795,
      "grad_norm": 4.041304588317871,
      "learning_rate": 4.3006262697022766e-05,
      "loss": 1.0485,
      "num_input_tokens_seen": 131041816,
      "step": 8144
    },
    {
      "epoch": 0.5705429614647088,
      "grad_norm": 3.405008316040039,
      "learning_rate": 4.2999264448336255e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 131058200,
      "step": 8145
    },
    {
      "epoch": 0.570613009710438,
      "grad_norm": 4.503276348114014,
      "learning_rate": 4.2992266199649744e-05,
      "loss": 0.8806,
      "num_input_tokens_seen": 131074584,
      "step": 8146
    },
    {
      "epoch": 0.5706830579561674,
      "grad_norm": 3.5301196575164795,
      "learning_rate": 4.2985267950963225e-05,
      "loss": 0.918,
      "num_input_tokens_seen": 131090968,
      "step": 8147
    },
    {
      "epoch": 0.5707531062018966,
      "grad_norm": 3.807494878768921,
      "learning_rate": 4.2978269702276714e-05,
      "loss": 1.1168,
      "num_input_tokens_seen": 131106840,
      "step": 8148
    },
    {
      "epoch": 0.5708231544476258,
      "grad_norm": 3.9565365314483643,
      "learning_rate": 4.2971271453590196e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 131123224,
      "step": 8149
    },
    {
      "epoch": 0.5708932026933551,
      "grad_norm": 3.4169907569885254,
      "learning_rate": 4.2964273204903684e-05,
      "loss": 0.9627,
      "num_input_tokens_seen": 131139608,
      "step": 8150
    },
    {
      "epoch": 0.5709632509390843,
      "grad_norm": 3.7201099395751953,
      "learning_rate": 4.295727495621717e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 131155896,
      "step": 8151
    },
    {
      "epoch": 0.5710332991848135,
      "grad_norm": 4.128871440887451,
      "learning_rate": 4.295027670753065e-05,
      "loss": 1.1233,
      "num_input_tokens_seen": 131171824,
      "step": 8152
    },
    {
      "epoch": 0.5711033474305428,
      "grad_norm": 4.046292304992676,
      "learning_rate": 4.294327845884414e-05,
      "loss": 1.0088,
      "num_input_tokens_seen": 131188208,
      "step": 8153
    },
    {
      "epoch": 0.571173395676272,
      "grad_norm": 4.006393909454346,
      "learning_rate": 4.293628021015762e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 131204592,
      "step": 8154
    },
    {
      "epoch": 0.5712434439220013,
      "grad_norm": 3.7372918128967285,
      "learning_rate": 4.292928196147111e-05,
      "loss": 0.8374,
      "num_input_tokens_seen": 131220552,
      "step": 8155
    },
    {
      "epoch": 0.5713134921677305,
      "grad_norm": 3.778796672821045,
      "learning_rate": 4.292228371278459e-05,
      "loss": 1.193,
      "num_input_tokens_seen": 131236936,
      "step": 8156
    },
    {
      "epoch": 0.5713835404134597,
      "grad_norm": 5.442931652069092,
      "learning_rate": 4.291528546409808e-05,
      "loss": 1.3328,
      "num_input_tokens_seen": 131253320,
      "step": 8157
    },
    {
      "epoch": 0.571453588659189,
      "grad_norm": 3.9510107040405273,
      "learning_rate": 4.2908287215411566e-05,
      "loss": 1.1776,
      "num_input_tokens_seen": 131269704,
      "step": 8158
    },
    {
      "epoch": 0.5715236369049183,
      "grad_norm": 3.787173271179199,
      "learning_rate": 4.290128896672505e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 131286088,
      "step": 8159
    },
    {
      "epoch": 0.5715936851506475,
      "grad_norm": 4.486179828643799,
      "learning_rate": 4.2894290718038536e-05,
      "loss": 0.8241,
      "num_input_tokens_seen": 131302472,
      "step": 8160
    },
    {
      "epoch": 0.5716637333963768,
      "grad_norm": 5.035008907318115,
      "learning_rate": 4.288729246935201e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 131318520,
      "step": 8161
    },
    {
      "epoch": 0.571733781642106,
      "grad_norm": 3.7599833011627197,
      "learning_rate": 4.28802942206655e-05,
      "loss": 0.8591,
      "num_input_tokens_seen": 131334904,
      "step": 8162
    },
    {
      "epoch": 0.5718038298878353,
      "grad_norm": 4.14766263961792,
      "learning_rate": 4.287329597197899e-05,
      "loss": 1.1655,
      "num_input_tokens_seen": 131351288,
      "step": 8163
    },
    {
      "epoch": 0.5718738781335645,
      "grad_norm": 3.914499044418335,
      "learning_rate": 4.286629772329247e-05,
      "loss": 1.0247,
      "num_input_tokens_seen": 131367672,
      "step": 8164
    },
    {
      "epoch": 0.5719439263792937,
      "grad_norm": 3.5725250244140625,
      "learning_rate": 4.285929947460596e-05,
      "loss": 1.0252,
      "num_input_tokens_seen": 131383968,
      "step": 8165
    },
    {
      "epoch": 0.572013974625023,
      "grad_norm": 3.5275251865386963,
      "learning_rate": 4.285230122591944e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 131400352,
      "step": 8166
    },
    {
      "epoch": 0.5720840228707522,
      "grad_norm": 5.734055042266846,
      "learning_rate": 4.284530297723293e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 131416168,
      "step": 8167
    },
    {
      "epoch": 0.5721540711164814,
      "grad_norm": 6.255549907684326,
      "learning_rate": 4.283830472854642e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 131432072,
      "step": 8168
    },
    {
      "epoch": 0.5722241193622107,
      "grad_norm": 5.589477062225342,
      "learning_rate": 4.283130647985989e-05,
      "loss": 0.9766,
      "num_input_tokens_seen": 131448456,
      "step": 8169
    },
    {
      "epoch": 0.5722941676079399,
      "grad_norm": 4.749863147735596,
      "learning_rate": 4.282430823117339e-05,
      "loss": 1.122,
      "num_input_tokens_seen": 131464840,
      "step": 8170
    },
    {
      "epoch": 0.5723642158536693,
      "grad_norm": 4.904588222503662,
      "learning_rate": 4.2817309982486864e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 131481224,
      "step": 8171
    },
    {
      "epoch": 0.5724342640993985,
      "grad_norm": 3.5946271419525146,
      "learning_rate": 4.281031173380035e-05,
      "loss": 0.8804,
      "num_input_tokens_seen": 131497608,
      "step": 8172
    },
    {
      "epoch": 0.5725043123451277,
      "grad_norm": 4.187933921813965,
      "learning_rate": 4.280331348511384e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 131513776,
      "step": 8173
    },
    {
      "epoch": 0.572574360590857,
      "grad_norm": 4.336525917053223,
      "learning_rate": 4.279631523642732e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 131529544,
      "step": 8174
    },
    {
      "epoch": 0.5726444088365862,
      "grad_norm": 4.156813621520996,
      "learning_rate": 4.278931698774081e-05,
      "loss": 0.9311,
      "num_input_tokens_seen": 131544880,
      "step": 8175
    },
    {
      "epoch": 0.5727144570823155,
      "grad_norm": 4.064977169036865,
      "learning_rate": 4.278231873905429e-05,
      "loss": 1.2461,
      "num_input_tokens_seen": 131560600,
      "step": 8176
    },
    {
      "epoch": 0.5727845053280447,
      "grad_norm": 3.8093082904815674,
      "learning_rate": 4.277532049036778e-05,
      "loss": 1.0079,
      "num_input_tokens_seen": 131576984,
      "step": 8177
    },
    {
      "epoch": 0.5728545535737739,
      "grad_norm": 3.8263261318206787,
      "learning_rate": 4.276832224168127e-05,
      "loss": 1.0441,
      "num_input_tokens_seen": 131592848,
      "step": 8178
    },
    {
      "epoch": 0.5729246018195032,
      "grad_norm": 4.268642425537109,
      "learning_rate": 4.2761323992994745e-05,
      "loss": 1.0139,
      "num_input_tokens_seen": 131609232,
      "step": 8179
    },
    {
      "epoch": 0.5729946500652324,
      "grad_norm": 3.6199145317077637,
      "learning_rate": 4.275432574430824e-05,
      "loss": 1.0993,
      "num_input_tokens_seen": 131625160,
      "step": 8180
    },
    {
      "epoch": 0.5730646983109616,
      "grad_norm": 5.106734275817871,
      "learning_rate": 4.2747327495621716e-05,
      "loss": 1.2588,
      "num_input_tokens_seen": 131641240,
      "step": 8181
    },
    {
      "epoch": 0.5731347465566909,
      "grad_norm": 4.9245381355285645,
      "learning_rate": 4.2740329246935204e-05,
      "loss": 0.9496,
      "num_input_tokens_seen": 131657624,
      "step": 8182
    },
    {
      "epoch": 0.5732047948024201,
      "grad_norm": 4.108707904815674,
      "learning_rate": 4.2733330998248686e-05,
      "loss": 1.0043,
      "num_input_tokens_seen": 131673400,
      "step": 8183
    },
    {
      "epoch": 0.5732748430481495,
      "grad_norm": 4.672266006469727,
      "learning_rate": 4.2726332749562175e-05,
      "loss": 1.1272,
      "num_input_tokens_seen": 131689784,
      "step": 8184
    },
    {
      "epoch": 0.5733448912938787,
      "grad_norm": 5.033571720123291,
      "learning_rate": 4.271933450087566e-05,
      "loss": 1.067,
      "num_input_tokens_seen": 131706168,
      "step": 8185
    },
    {
      "epoch": 0.5734149395396079,
      "grad_norm": 4.336825847625732,
      "learning_rate": 4.2712336252189145e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 131722496,
      "step": 8186
    },
    {
      "epoch": 0.5734849877853372,
      "grad_norm": 3.6326375007629395,
      "learning_rate": 4.2705338003502634e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 131738880,
      "step": 8187
    },
    {
      "epoch": 0.5735550360310664,
      "grad_norm": 3.407284736633301,
      "learning_rate": 4.269833975481611e-05,
      "loss": 1.099,
      "num_input_tokens_seen": 131755080,
      "step": 8188
    },
    {
      "epoch": 0.5736250842767956,
      "grad_norm": 5.604672431945801,
      "learning_rate": 4.26913415061296e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 131771464,
      "step": 8189
    },
    {
      "epoch": 0.5736951325225249,
      "grad_norm": 3.4375531673431396,
      "learning_rate": 4.268434325744309e-05,
      "loss": 0.9977,
      "num_input_tokens_seen": 131787848,
      "step": 8190
    },
    {
      "epoch": 0.5737651807682541,
      "grad_norm": 4.113184928894043,
      "learning_rate": 4.267734500875657e-05,
      "loss": 1.0283,
      "num_input_tokens_seen": 131804232,
      "step": 8191
    },
    {
      "epoch": 0.5738352290139834,
      "grad_norm": 5.918920516967773,
      "learning_rate": 4.2670346760070056e-05,
      "loss": 1.168,
      "num_input_tokens_seen": 131819608,
      "step": 8192
    },
    {
      "epoch": 0.5739052772597126,
      "grad_norm": 4.387043476104736,
      "learning_rate": 4.266334851138354e-05,
      "loss": 1.2334,
      "num_input_tokens_seen": 131835992,
      "step": 8193
    },
    {
      "epoch": 0.5739753255054418,
      "grad_norm": 4.381311416625977,
      "learning_rate": 4.2656350262697027e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 131851920,
      "step": 8194
    },
    {
      "epoch": 0.5740453737511712,
      "grad_norm": 6.177340507507324,
      "learning_rate": 4.2649352014010515e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 131867240,
      "step": 8195
    },
    {
      "epoch": 0.5741154219969004,
      "grad_norm": 4.505295276641846,
      "learning_rate": 4.2642353765324e-05,
      "loss": 1.1147,
      "num_input_tokens_seen": 131883408,
      "step": 8196
    },
    {
      "epoch": 0.5741854702426296,
      "grad_norm": 3.9150309562683105,
      "learning_rate": 4.2635355516637486e-05,
      "loss": 0.9609,
      "num_input_tokens_seen": 131899144,
      "step": 8197
    },
    {
      "epoch": 0.5742555184883589,
      "grad_norm": 4.148092269897461,
      "learning_rate": 4.262835726795096e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 131915528,
      "step": 8198
    },
    {
      "epoch": 0.5743255667340881,
      "grad_norm": 3.353017807006836,
      "learning_rate": 4.262135901926445e-05,
      "loss": 1.0293,
      "num_input_tokens_seen": 131931912,
      "step": 8199
    },
    {
      "epoch": 0.5743956149798174,
      "grad_norm": 3.949207305908203,
      "learning_rate": 4.2614360770577945e-05,
      "loss": 0.8957,
      "num_input_tokens_seen": 131947712,
      "step": 8200
    },
    {
      "epoch": 0.5743956149798174,
      "eval_loss": 1.1200522184371948,
      "eval_runtime": 0.2045,
      "eval_samples_per_second": 4.889,
      "eval_steps_per_second": 4.889,
      "num_input_tokens_seen": 131947712,
      "step": 8200
    },
    {
      "epoch": 0.5744656632255466,
      "grad_norm": 3.978220224380493,
      "learning_rate": 4.260736252189142e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 131963208,
      "step": 8201
    },
    {
      "epoch": 0.5745357114712758,
      "grad_norm": 3.7986531257629395,
      "learning_rate": 4.260036427320491e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 131978048,
      "step": 8202
    },
    {
      "epoch": 0.5746057597170051,
      "grad_norm": 3.9384713172912598,
      "learning_rate": 4.259336602451839e-05,
      "loss": 1.1815,
      "num_input_tokens_seen": 131994432,
      "step": 8203
    },
    {
      "epoch": 0.5746758079627343,
      "grad_norm": 4.360903739929199,
      "learning_rate": 4.258636777583188e-05,
      "loss": 1.107,
      "num_input_tokens_seen": 132010504,
      "step": 8204
    },
    {
      "epoch": 0.5747458562084636,
      "grad_norm": 4.957514762878418,
      "learning_rate": 4.2579369527145354e-05,
      "loss": 1.1793,
      "num_input_tokens_seen": 132026192,
      "step": 8205
    },
    {
      "epoch": 0.5748159044541928,
      "grad_norm": 4.4821882247924805,
      "learning_rate": 4.257237127845885e-05,
      "loss": 1.1506,
      "num_input_tokens_seen": 132041624,
      "step": 8206
    },
    {
      "epoch": 0.574885952699922,
      "grad_norm": 3.915302276611328,
      "learning_rate": 4.256537302977234e-05,
      "loss": 1.0654,
      "num_input_tokens_seen": 132056600,
      "step": 8207
    },
    {
      "epoch": 0.5749560009456514,
      "grad_norm": 3.67526912689209,
      "learning_rate": 4.255837478108581e-05,
      "loss": 1.101,
      "num_input_tokens_seen": 132072984,
      "step": 8208
    },
    {
      "epoch": 0.5750260491913806,
      "grad_norm": 5.455066204071045,
      "learning_rate": 4.25513765323993e-05,
      "loss": 1.1665,
      "num_input_tokens_seen": 132089368,
      "step": 8209
    },
    {
      "epoch": 0.5750960974371098,
      "grad_norm": 3.7879862785339355,
      "learning_rate": 4.254437828371278e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 132105752,
      "step": 8210
    },
    {
      "epoch": 0.5751661456828391,
      "grad_norm": 4.033114910125732,
      "learning_rate": 4.253738003502627e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 132122136,
      "step": 8211
    },
    {
      "epoch": 0.5752361939285683,
      "grad_norm": 3.596369743347168,
      "learning_rate": 4.253038178633976e-05,
      "loss": 1.0748,
      "num_input_tokens_seen": 132138520,
      "step": 8212
    },
    {
      "epoch": 0.5753062421742976,
      "grad_norm": 4.42296838760376,
      "learning_rate": 4.252338353765324e-05,
      "loss": 1.0508,
      "num_input_tokens_seen": 132154904,
      "step": 8213
    },
    {
      "epoch": 0.5753762904200268,
      "grad_norm": 3.5020558834075928,
      "learning_rate": 4.251638528896673e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 132171248,
      "step": 8214
    },
    {
      "epoch": 0.575446338665756,
      "grad_norm": 4.176201343536377,
      "learning_rate": 4.2509387040280206e-05,
      "loss": 1.0606,
      "num_input_tokens_seen": 132187296,
      "step": 8215
    },
    {
      "epoch": 0.5755163869114853,
      "grad_norm": 3.459134817123413,
      "learning_rate": 4.25023887915937e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 132203680,
      "step": 8216
    },
    {
      "epoch": 0.5755864351572145,
      "grad_norm": 3.870122194290161,
      "learning_rate": 4.249539054290719e-05,
      "loss": 1.005,
      "num_input_tokens_seen": 132220064,
      "step": 8217
    },
    {
      "epoch": 0.5756564834029437,
      "grad_norm": 4.246501922607422,
      "learning_rate": 4.2488392294220665e-05,
      "loss": 1.246,
      "num_input_tokens_seen": 132236248,
      "step": 8218
    },
    {
      "epoch": 0.575726531648673,
      "grad_norm": 4.028504371643066,
      "learning_rate": 4.248139404553415e-05,
      "loss": 1.1537,
      "num_input_tokens_seen": 132252624,
      "step": 8219
    },
    {
      "epoch": 0.5757965798944022,
      "grad_norm": 5.083249092102051,
      "learning_rate": 4.2474395796847635e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 132269008,
      "step": 8220
    },
    {
      "epoch": 0.5758666281401316,
      "grad_norm": 4.459508419036865,
      "learning_rate": 4.2467397548161124e-05,
      "loss": 0.9581,
      "num_input_tokens_seen": 132285392,
      "step": 8221
    },
    {
      "epoch": 0.5759366763858608,
      "grad_norm": 3.5667686462402344,
      "learning_rate": 4.246039929947461e-05,
      "loss": 0.9399,
      "num_input_tokens_seen": 132301584,
      "step": 8222
    },
    {
      "epoch": 0.57600672463159,
      "grad_norm": 3.982933759689331,
      "learning_rate": 4.2453401050788094e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 132317968,
      "step": 8223
    },
    {
      "epoch": 0.5760767728773193,
      "grad_norm": 3.59114408493042,
      "learning_rate": 4.244640280210158e-05,
      "loss": 0.9939,
      "num_input_tokens_seen": 132334352,
      "step": 8224
    },
    {
      "epoch": 0.5761468211230485,
      "grad_norm": 4.596956729888916,
      "learning_rate": 4.243940455341506e-05,
      "loss": 1.1058,
      "num_input_tokens_seen": 132350736,
      "step": 8225
    },
    {
      "epoch": 0.5762168693687777,
      "grad_norm": 4.798111915588379,
      "learning_rate": 4.243240630472855e-05,
      "loss": 1.1164,
      "num_input_tokens_seen": 132367120,
      "step": 8226
    },
    {
      "epoch": 0.576286917614507,
      "grad_norm": 3.9915149211883545,
      "learning_rate": 4.242540805604204e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 132382712,
      "step": 8227
    },
    {
      "epoch": 0.5763569658602362,
      "grad_norm": 3.8071064949035645,
      "learning_rate": 4.241840980735552e-05,
      "loss": 0.8643,
      "num_input_tokens_seen": 132399096,
      "step": 8228
    },
    {
      "epoch": 0.5764270141059655,
      "grad_norm": 3.679708242416382,
      "learning_rate": 4.2411411558669005e-05,
      "loss": 1.1226,
      "num_input_tokens_seen": 132415288,
      "step": 8229
    },
    {
      "epoch": 0.5764970623516947,
      "grad_norm": 3.723848581314087,
      "learning_rate": 4.240441330998249e-05,
      "loss": 1.0517,
      "num_input_tokens_seen": 132431672,
      "step": 8230
    },
    {
      "epoch": 0.5765671105974239,
      "grad_norm": 4.438333988189697,
      "learning_rate": 4.2397415061295976e-05,
      "loss": 1.1601,
      "num_input_tokens_seen": 132447408,
      "step": 8231
    },
    {
      "epoch": 0.5766371588431533,
      "grad_norm": 3.4421169757843018,
      "learning_rate": 4.239041681260946e-05,
      "loss": 1.017,
      "num_input_tokens_seen": 132463600,
      "step": 8232
    },
    {
      "epoch": 0.5767072070888825,
      "grad_norm": 3.851832389831543,
      "learning_rate": 4.2383418563922946e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 132479984,
      "step": 8233
    },
    {
      "epoch": 0.5767772553346117,
      "grad_norm": 5.055593967437744,
      "learning_rate": 4.2376420315236435e-05,
      "loss": 1.0512,
      "num_input_tokens_seen": 132495872,
      "step": 8234
    },
    {
      "epoch": 0.576847303580341,
      "grad_norm": 3.958918809890747,
      "learning_rate": 4.236942206654991e-05,
      "loss": 1.0412,
      "num_input_tokens_seen": 132512256,
      "step": 8235
    },
    {
      "epoch": 0.5769173518260702,
      "grad_norm": 3.6936240196228027,
      "learning_rate": 4.2362423817863405e-05,
      "loss": 1.0251,
      "num_input_tokens_seen": 132528640,
      "step": 8236
    },
    {
      "epoch": 0.5769874000717995,
      "grad_norm": 4.020928859710693,
      "learning_rate": 4.235542556917688e-05,
      "loss": 0.979,
      "num_input_tokens_seen": 132543712,
      "step": 8237
    },
    {
      "epoch": 0.5770574483175287,
      "grad_norm": 3.5804123878479004,
      "learning_rate": 4.234842732049037e-05,
      "loss": 1.0581,
      "num_input_tokens_seen": 132559568,
      "step": 8238
    },
    {
      "epoch": 0.5771274965632579,
      "grad_norm": 5.450964450836182,
      "learning_rate": 4.234142907180386e-05,
      "loss": 1.0109,
      "num_input_tokens_seen": 132575952,
      "step": 8239
    },
    {
      "epoch": 0.5771975448089872,
      "grad_norm": 4.571927547454834,
      "learning_rate": 4.233443082311734e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 132591832,
      "step": 8240
    },
    {
      "epoch": 0.5772675930547164,
      "grad_norm": 5.09351921081543,
      "learning_rate": 4.232743257443083e-05,
      "loss": 0.9558,
      "num_input_tokens_seen": 132608216,
      "step": 8241
    },
    {
      "epoch": 0.5773376413004457,
      "grad_norm": 3.211693525314331,
      "learning_rate": 4.232043432574431e-05,
      "loss": 0.9804,
      "num_input_tokens_seen": 132624600,
      "step": 8242
    },
    {
      "epoch": 0.5774076895461749,
      "grad_norm": 3.822723865509033,
      "learning_rate": 4.23134360770578e-05,
      "loss": 1.0267,
      "num_input_tokens_seen": 132640984,
      "step": 8243
    },
    {
      "epoch": 0.5774777377919041,
      "grad_norm": 5.541271686553955,
      "learning_rate": 4.230643782837129e-05,
      "loss": 0.7341,
      "num_input_tokens_seen": 132657368,
      "step": 8244
    },
    {
      "epoch": 0.5775477860376335,
      "grad_norm": 6.682171821594238,
      "learning_rate": 4.229943957968476e-05,
      "loss": 0.9682,
      "num_input_tokens_seen": 132673752,
      "step": 8245
    },
    {
      "epoch": 0.5776178342833627,
      "grad_norm": 3.4161407947540283,
      "learning_rate": 4.229244133099826e-05,
      "loss": 0.9307,
      "num_input_tokens_seen": 132690136,
      "step": 8246
    },
    {
      "epoch": 0.5776878825290919,
      "grad_norm": 3.6526873111724854,
      "learning_rate": 4.228544308231173e-05,
      "loss": 1.0356,
      "num_input_tokens_seen": 132706520,
      "step": 8247
    },
    {
      "epoch": 0.5777579307748212,
      "grad_norm": 4.506532192230225,
      "learning_rate": 4.227844483362522e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 132721600,
      "step": 8248
    },
    {
      "epoch": 0.5778279790205504,
      "grad_norm": 3.9028801918029785,
      "learning_rate": 4.227144658493871e-05,
      "loss": 1.1071,
      "num_input_tokens_seen": 132737728,
      "step": 8249
    },
    {
      "epoch": 0.5778980272662797,
      "grad_norm": 4.508323669433594,
      "learning_rate": 4.226444833625219e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 132753240,
      "step": 8250
    },
    {
      "epoch": 0.5779680755120089,
      "grad_norm": 4.045495986938477,
      "learning_rate": 4.225745008756568e-05,
      "loss": 1.0133,
      "num_input_tokens_seen": 132769024,
      "step": 8251
    },
    {
      "epoch": 0.5780381237577381,
      "grad_norm": 3.3747479915618896,
      "learning_rate": 4.225045183887916e-05,
      "loss": 0.9641,
      "num_input_tokens_seen": 132785408,
      "step": 8252
    },
    {
      "epoch": 0.5781081720034674,
      "grad_norm": 3.527862787246704,
      "learning_rate": 4.224345359019265e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 132801792,
      "step": 8253
    },
    {
      "epoch": 0.5781782202491966,
      "grad_norm": 5.42443323135376,
      "learning_rate": 4.223645534150614e-05,
      "loss": 0.9197,
      "num_input_tokens_seen": 132817536,
      "step": 8254
    },
    {
      "epoch": 0.5782482684949258,
      "grad_norm": 3.450995922088623,
      "learning_rate": 4.2229457092819614e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 132833920,
      "step": 8255
    },
    {
      "epoch": 0.5783183167406551,
      "grad_norm": 3.824470043182373,
      "learning_rate": 4.222245884413311e-05,
      "loss": 1.0504,
      "num_input_tokens_seen": 132850304,
      "step": 8256
    },
    {
      "epoch": 0.5783883649863844,
      "grad_norm": 4.278646945953369,
      "learning_rate": 4.2215460595446584e-05,
      "loss": 0.766,
      "num_input_tokens_seen": 132865216,
      "step": 8257
    },
    {
      "epoch": 0.5784584132321137,
      "grad_norm": 3.582697629928589,
      "learning_rate": 4.220846234676007e-05,
      "loss": 0.9996,
      "num_input_tokens_seen": 132881600,
      "step": 8258
    },
    {
      "epoch": 0.5785284614778429,
      "grad_norm": 4.3087849617004395,
      "learning_rate": 4.2201464098073555e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 132897360,
      "step": 8259
    },
    {
      "epoch": 0.5785985097235721,
      "grad_norm": 3.9091784954071045,
      "learning_rate": 4.219446584938704e-05,
      "loss": 1.0683,
      "num_input_tokens_seen": 132913504,
      "step": 8260
    },
    {
      "epoch": 0.5786685579693014,
      "grad_norm": 3.899811267852783,
      "learning_rate": 4.218746760070053e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 132929184,
      "step": 8261
    },
    {
      "epoch": 0.5787386062150306,
      "grad_norm": 4.090141773223877,
      "learning_rate": 4.2180469352014014e-05,
      "loss": 1.1917,
      "num_input_tokens_seen": 132945016,
      "step": 8262
    },
    {
      "epoch": 0.5788086544607598,
      "grad_norm": 4.3014631271362305,
      "learning_rate": 4.21734711033275e-05,
      "loss": 0.8849,
      "num_input_tokens_seen": 132961224,
      "step": 8263
    },
    {
      "epoch": 0.5788787027064891,
      "grad_norm": 4.155791282653809,
      "learning_rate": 4.216647285464098e-05,
      "loss": 1.2749,
      "num_input_tokens_seen": 132977464,
      "step": 8264
    },
    {
      "epoch": 0.5789487509522183,
      "grad_norm": 3.6278793811798096,
      "learning_rate": 4.2159474605954466e-05,
      "loss": 0.9852,
      "num_input_tokens_seen": 132993848,
      "step": 8265
    },
    {
      "epoch": 0.5790187991979476,
      "grad_norm": 3.477964401245117,
      "learning_rate": 4.215247635726796e-05,
      "loss": 0.8369,
      "num_input_tokens_seen": 133010232,
      "step": 8266
    },
    {
      "epoch": 0.5790888474436768,
      "grad_norm": 5.3280439376831055,
      "learning_rate": 4.2145478108581436e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 133026616,
      "step": 8267
    },
    {
      "epoch": 0.579158895689406,
      "grad_norm": 3.5057740211486816,
      "learning_rate": 4.2138479859894925e-05,
      "loss": 0.9546,
      "num_input_tokens_seen": 133043000,
      "step": 8268
    },
    {
      "epoch": 0.5792289439351354,
      "grad_norm": 6.108116149902344,
      "learning_rate": 4.2131481611208407e-05,
      "loss": 0.9677,
      "num_input_tokens_seen": 133058584,
      "step": 8269
    },
    {
      "epoch": 0.5792989921808646,
      "grad_norm": 3.764878273010254,
      "learning_rate": 4.2124483362521895e-05,
      "loss": 0.9287,
      "num_input_tokens_seen": 133074176,
      "step": 8270
    },
    {
      "epoch": 0.5793690404265938,
      "grad_norm": 3.9870357513427734,
      "learning_rate": 4.2117485113835384e-05,
      "loss": 1.1149,
      "num_input_tokens_seen": 133090320,
      "step": 8271
    },
    {
      "epoch": 0.5794390886723231,
      "grad_norm": 3.3531510829925537,
      "learning_rate": 4.2110486865148866e-05,
      "loss": 0.7976,
      "num_input_tokens_seen": 133106312,
      "step": 8272
    },
    {
      "epoch": 0.5795091369180523,
      "grad_norm": 3.321676731109619,
      "learning_rate": 4.2103488616462354e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 133122696,
      "step": 8273
    },
    {
      "epoch": 0.5795791851637816,
      "grad_norm": 4.315324306488037,
      "learning_rate": 4.209649036777583e-05,
      "loss": 1.0502,
      "num_input_tokens_seen": 133139080,
      "step": 8274
    },
    {
      "epoch": 0.5796492334095108,
      "grad_norm": 3.9912524223327637,
      "learning_rate": 4.208949211908932e-05,
      "loss": 1.0919,
      "num_input_tokens_seen": 133155192,
      "step": 8275
    },
    {
      "epoch": 0.57971928165524,
      "grad_norm": 3.8014609813690186,
      "learning_rate": 4.208249387040281e-05,
      "loss": 1.0997,
      "num_input_tokens_seen": 133171576,
      "step": 8276
    },
    {
      "epoch": 0.5797893299009693,
      "grad_norm": 3.790175437927246,
      "learning_rate": 4.207549562171629e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 133187448,
      "step": 8277
    },
    {
      "epoch": 0.5798593781466985,
      "grad_norm": 3.7050399780273438,
      "learning_rate": 4.206849737302978e-05,
      "loss": 1.0661,
      "num_input_tokens_seen": 133203832,
      "step": 8278
    },
    {
      "epoch": 0.5799294263924278,
      "grad_norm": 3.9394073486328125,
      "learning_rate": 4.206149912434326e-05,
      "loss": 1.0586,
      "num_input_tokens_seen": 133219176,
      "step": 8279
    },
    {
      "epoch": 0.579999474638157,
      "grad_norm": 3.8162503242492676,
      "learning_rate": 4.205450087565675e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 133235272,
      "step": 8280
    },
    {
      "epoch": 0.5800695228838862,
      "grad_norm": 4.20829439163208,
      "learning_rate": 4.2047502626970236e-05,
      "loss": 1.0363,
      "num_input_tokens_seen": 133250872,
      "step": 8281
    },
    {
      "epoch": 0.5801395711296156,
      "grad_norm": 4.279886722564697,
      "learning_rate": 4.204050437828372e-05,
      "loss": 1.1982,
      "num_input_tokens_seen": 133266680,
      "step": 8282
    },
    {
      "epoch": 0.5802096193753448,
      "grad_norm": 4.4257354736328125,
      "learning_rate": 4.2033506129597206e-05,
      "loss": 1.1011,
      "num_input_tokens_seen": 133282928,
      "step": 8283
    },
    {
      "epoch": 0.580279667621074,
      "grad_norm": 4.393752098083496,
      "learning_rate": 4.202650788091068e-05,
      "loss": 1.373,
      "num_input_tokens_seen": 133299312,
      "step": 8284
    },
    {
      "epoch": 0.5803497158668033,
      "grad_norm": 3.4351353645324707,
      "learning_rate": 4.201950963222417e-05,
      "loss": 0.9685,
      "num_input_tokens_seen": 133315568,
      "step": 8285
    },
    {
      "epoch": 0.5804197641125325,
      "grad_norm": 5.595382213592529,
      "learning_rate": 4.201251138353765e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 133331952,
      "step": 8286
    },
    {
      "epoch": 0.5804898123582618,
      "grad_norm": 3.8465957641601562,
      "learning_rate": 4.200551313485114e-05,
      "loss": 1.083,
      "num_input_tokens_seen": 133347992,
      "step": 8287
    },
    {
      "epoch": 0.580559860603991,
      "grad_norm": 3.53275465965271,
      "learning_rate": 4.199851488616463e-05,
      "loss": 0.8685,
      "num_input_tokens_seen": 133364376,
      "step": 8288
    },
    {
      "epoch": 0.5806299088497202,
      "grad_norm": 3.821805000305176,
      "learning_rate": 4.199151663747811e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 133380760,
      "step": 8289
    },
    {
      "epoch": 0.5806999570954495,
      "grad_norm": 3.843419313430786,
      "learning_rate": 4.19845183887916e-05,
      "loss": 1.0135,
      "num_input_tokens_seen": 133397144,
      "step": 8290
    },
    {
      "epoch": 0.5807700053411787,
      "grad_norm": 3.5370657444000244,
      "learning_rate": 4.1977520140105074e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 133413528,
      "step": 8291
    },
    {
      "epoch": 0.5808400535869079,
      "grad_norm": 3.9939937591552734,
      "learning_rate": 4.197052189141857e-05,
      "loss": 1.1104,
      "num_input_tokens_seen": 133429912,
      "step": 8292
    },
    {
      "epoch": 0.5809101018326372,
      "grad_norm": 3.8171210289001465,
      "learning_rate": 4.196352364273206e-05,
      "loss": 1.0962,
      "num_input_tokens_seen": 133446280,
      "step": 8293
    },
    {
      "epoch": 0.5809801500783665,
      "grad_norm": 5.993227481842041,
      "learning_rate": 4.195652539404553e-05,
      "loss": 1.0393,
      "num_input_tokens_seen": 133462272,
      "step": 8294
    },
    {
      "epoch": 0.5810501983240958,
      "grad_norm": 3.480929136276245,
      "learning_rate": 4.194952714535902e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 133478656,
      "step": 8295
    },
    {
      "epoch": 0.581120246569825,
      "grad_norm": 4.371162414550781,
      "learning_rate": 4.1942528896672504e-05,
      "loss": 0.861,
      "num_input_tokens_seen": 133495040,
      "step": 8296
    },
    {
      "epoch": 0.5811902948155542,
      "grad_norm": 3.9835524559020996,
      "learning_rate": 4.193553064798599e-05,
      "loss": 1.025,
      "num_input_tokens_seen": 133510576,
      "step": 8297
    },
    {
      "epoch": 0.5812603430612835,
      "grad_norm": 3.935680866241455,
      "learning_rate": 4.192853239929948e-05,
      "loss": 1.2121,
      "num_input_tokens_seen": 133526960,
      "step": 8298
    },
    {
      "epoch": 0.5813303913070127,
      "grad_norm": 5.499049186706543,
      "learning_rate": 4.192153415061296e-05,
      "loss": 0.9905,
      "num_input_tokens_seen": 133542336,
      "step": 8299
    },
    {
      "epoch": 0.5814004395527419,
      "grad_norm": 4.98264217376709,
      "learning_rate": 4.191453590192645e-05,
      "loss": 1.0016,
      "num_input_tokens_seen": 133556464,
      "step": 8300
    },
    {
      "epoch": 0.5814704877984712,
      "grad_norm": 4.091787338256836,
      "learning_rate": 4.1907537653239926e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 133572280,
      "step": 8301
    },
    {
      "epoch": 0.5815405360442004,
      "grad_norm": 3.9688591957092285,
      "learning_rate": 4.190053940455342e-05,
      "loss": 1.0631,
      "num_input_tokens_seen": 133588224,
      "step": 8302
    },
    {
      "epoch": 0.5816105842899297,
      "grad_norm": 3.7555758953094482,
      "learning_rate": 4.189354115586691e-05,
      "loss": 1.0418,
      "num_input_tokens_seen": 133604016,
      "step": 8303
    },
    {
      "epoch": 0.5816806325356589,
      "grad_norm": 3.423367977142334,
      "learning_rate": 4.1886542907180385e-05,
      "loss": 1.0608,
      "num_input_tokens_seen": 133620400,
      "step": 8304
    },
    {
      "epoch": 0.5817506807813881,
      "grad_norm": 4.812343597412109,
      "learning_rate": 4.1879544658493874e-05,
      "loss": 1.0295,
      "num_input_tokens_seen": 133636008,
      "step": 8305
    },
    {
      "epoch": 0.5818207290271175,
      "grad_norm": 4.64470100402832,
      "learning_rate": 4.1872546409807356e-05,
      "loss": 1.1065,
      "num_input_tokens_seen": 133651944,
      "step": 8306
    },
    {
      "epoch": 0.5818907772728467,
      "grad_norm": 3.9178478717803955,
      "learning_rate": 4.1865548161120844e-05,
      "loss": 1.162,
      "num_input_tokens_seen": 133667336,
      "step": 8307
    },
    {
      "epoch": 0.5819608255185759,
      "grad_norm": 3.7328476905822754,
      "learning_rate": 4.185854991243433e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 133683720,
      "step": 8308
    },
    {
      "epoch": 0.5820308737643052,
      "grad_norm": 5.883306980133057,
      "learning_rate": 4.1851551663747815e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 133700104,
      "step": 8309
    },
    {
      "epoch": 0.5821009220100344,
      "grad_norm": 3.6011807918548584,
      "learning_rate": 4.18445534150613e-05,
      "loss": 0.8344,
      "num_input_tokens_seen": 133716184,
      "step": 8310
    },
    {
      "epoch": 0.5821709702557637,
      "grad_norm": 5.971445083618164,
      "learning_rate": 4.183755516637478e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 133732552,
      "step": 8311
    },
    {
      "epoch": 0.5822410185014929,
      "grad_norm": 3.7484395503997803,
      "learning_rate": 4.1830556917688274e-05,
      "loss": 0.9394,
      "num_input_tokens_seen": 133748400,
      "step": 8312
    },
    {
      "epoch": 0.5823110667472221,
      "grad_norm": 4.420420169830322,
      "learning_rate": 4.182355866900175e-05,
      "loss": 0.8677,
      "num_input_tokens_seen": 133764784,
      "step": 8313
    },
    {
      "epoch": 0.5823811149929514,
      "grad_norm": 4.270792484283447,
      "learning_rate": 4.181656042031524e-05,
      "loss": 1.0086,
      "num_input_tokens_seen": 133780472,
      "step": 8314
    },
    {
      "epoch": 0.5824511632386806,
      "grad_norm": 3.99307918548584,
      "learning_rate": 4.1809562171628726e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 133796856,
      "step": 8315
    },
    {
      "epoch": 0.5825212114844099,
      "grad_norm": 3.3790438175201416,
      "learning_rate": 4.180256392294221e-05,
      "loss": 0.8995,
      "num_input_tokens_seen": 133813240,
      "step": 8316
    },
    {
      "epoch": 0.5825912597301391,
      "grad_norm": 3.9880869388580322,
      "learning_rate": 4.1795565674255696e-05,
      "loss": 1.0347,
      "num_input_tokens_seen": 133829624,
      "step": 8317
    },
    {
      "epoch": 0.5826613079758683,
      "grad_norm": 4.359625339508057,
      "learning_rate": 4.178856742556918e-05,
      "loss": 1.0979,
      "num_input_tokens_seen": 133845048,
      "step": 8318
    },
    {
      "epoch": 0.5827313562215977,
      "grad_norm": 3.856341600418091,
      "learning_rate": 4.178156917688267e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 133861432,
      "step": 8319
    },
    {
      "epoch": 0.5828014044673269,
      "grad_norm": 4.141717433929443,
      "learning_rate": 4.1774570928196155e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 133877176,
      "step": 8320
    },
    {
      "epoch": 0.5828714527130561,
      "grad_norm": 4.0302605628967285,
      "learning_rate": 4.176757267950963e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 133892840,
      "step": 8321
    },
    {
      "epoch": 0.5829415009587854,
      "grad_norm": 4.464367389678955,
      "learning_rate": 4.1760574430823126e-05,
      "loss": 1.2155,
      "num_input_tokens_seen": 133909224,
      "step": 8322
    },
    {
      "epoch": 0.5830115492045146,
      "grad_norm": 5.884219169616699,
      "learning_rate": 4.17535761821366e-05,
      "loss": 1.2055,
      "num_input_tokens_seen": 133925432,
      "step": 8323
    },
    {
      "epoch": 0.5830815974502439,
      "grad_norm": 4.006690979003906,
      "learning_rate": 4.174657793345009e-05,
      "loss": 1.0853,
      "num_input_tokens_seen": 133941816,
      "step": 8324
    },
    {
      "epoch": 0.5831516456959731,
      "grad_norm": 4.723453998565674,
      "learning_rate": 4.173957968476358e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 133958200,
      "step": 8325
    },
    {
      "epoch": 0.5832216939417023,
      "grad_norm": 4.328615665435791,
      "learning_rate": 4.173258143607706e-05,
      "loss": 1.3134,
      "num_input_tokens_seen": 133974584,
      "step": 8326
    },
    {
      "epoch": 0.5832917421874316,
      "grad_norm": 3.6810529232025146,
      "learning_rate": 4.172558318739055e-05,
      "loss": 1.0627,
      "num_input_tokens_seen": 133990936,
      "step": 8327
    },
    {
      "epoch": 0.5833617904331608,
      "grad_norm": 4.4679741859436035,
      "learning_rate": 4.171858493870403e-05,
      "loss": 1.0715,
      "num_input_tokens_seen": 134006488,
      "step": 8328
    },
    {
      "epoch": 0.58343183867889,
      "grad_norm": 3.5542099475860596,
      "learning_rate": 4.171158669001752e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 134022128,
      "step": 8329
    },
    {
      "epoch": 0.5835018869246194,
      "grad_norm": 5.318003177642822,
      "learning_rate": 4.170458844133101e-05,
      "loss": 1.0429,
      "num_input_tokens_seen": 134038512,
      "step": 8330
    },
    {
      "epoch": 0.5835719351703486,
      "grad_norm": 3.7052905559539795,
      "learning_rate": 4.169759019264448e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 134054816,
      "step": 8331
    },
    {
      "epoch": 0.5836419834160779,
      "grad_norm": 3.937094211578369,
      "learning_rate": 4.169059194395798e-05,
      "loss": 0.8523,
      "num_input_tokens_seen": 134071200,
      "step": 8332
    },
    {
      "epoch": 0.5837120316618071,
      "grad_norm": 4.033092498779297,
      "learning_rate": 4.168359369527145e-05,
      "loss": 1.0589,
      "num_input_tokens_seen": 134087416,
      "step": 8333
    },
    {
      "epoch": 0.5837820799075363,
      "grad_norm": 3.4819228649139404,
      "learning_rate": 4.167659544658494e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 134103800,
      "step": 8334
    },
    {
      "epoch": 0.5838521281532656,
      "grad_norm": 4.7478861808776855,
      "learning_rate": 4.166959719789843e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 134119856,
      "step": 8335
    },
    {
      "epoch": 0.5839221763989948,
      "grad_norm": 3.6348135471343994,
      "learning_rate": 4.166259894921191e-05,
      "loss": 0.9789,
      "num_input_tokens_seen": 134135992,
      "step": 8336
    },
    {
      "epoch": 0.583992224644724,
      "grad_norm": 3.5731043815612793,
      "learning_rate": 4.16556007005254e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 134152376,
      "step": 8337
    },
    {
      "epoch": 0.5840622728904533,
      "grad_norm": 4.3569254875183105,
      "learning_rate": 4.164860245183888e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 134168760,
      "step": 8338
    },
    {
      "epoch": 0.5841323211361825,
      "grad_norm": 3.468846082687378,
      "learning_rate": 4.164160420315237e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 134185144,
      "step": 8339
    },
    {
      "epoch": 0.5842023693819118,
      "grad_norm": 6.284951210021973,
      "learning_rate": 4.1634605954465846e-05,
      "loss": 0.9518,
      "num_input_tokens_seen": 134200680,
      "step": 8340
    },
    {
      "epoch": 0.584272417627641,
      "grad_norm": 4.579503536224365,
      "learning_rate": 4.1627607705779334e-05,
      "loss": 0.9929,
      "num_input_tokens_seen": 134217064,
      "step": 8341
    },
    {
      "epoch": 0.5843424658733702,
      "grad_norm": 3.716926336288452,
      "learning_rate": 4.162060945709283e-05,
      "loss": 1.0092,
      "num_input_tokens_seen": 134233448,
      "step": 8342
    },
    {
      "epoch": 0.5844125141190996,
      "grad_norm": 4.120275974273682,
      "learning_rate": 4.1613611208406305e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 134249832,
      "step": 8343
    },
    {
      "epoch": 0.5844825623648288,
      "grad_norm": 5.749308109283447,
      "learning_rate": 4.160661295971979e-05,
      "loss": 1.0097,
      "num_input_tokens_seen": 134266216,
      "step": 8344
    },
    {
      "epoch": 0.5845526106105581,
      "grad_norm": 3.9694550037384033,
      "learning_rate": 4.1599614711033275e-05,
      "loss": 0.9815,
      "num_input_tokens_seen": 134282600,
      "step": 8345
    },
    {
      "epoch": 0.5846226588562873,
      "grad_norm": 4.208174705505371,
      "learning_rate": 4.1592616462346764e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 134298088,
      "step": 8346
    },
    {
      "epoch": 0.5846927071020165,
      "grad_norm": 3.957308530807495,
      "learning_rate": 4.158561821366025e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 134314144,
      "step": 8347
    },
    {
      "epoch": 0.5847627553477458,
      "grad_norm": 4.887307643890381,
      "learning_rate": 4.1578619964973734e-05,
      "loss": 0.9538,
      "num_input_tokens_seen": 134330528,
      "step": 8348
    },
    {
      "epoch": 0.584832803593475,
      "grad_norm": 4.353172302246094,
      "learning_rate": 4.157162171628722e-05,
      "loss": 1.1329,
      "num_input_tokens_seen": 134346912,
      "step": 8349
    },
    {
      "epoch": 0.5849028518392042,
      "grad_norm": 3.504237174987793,
      "learning_rate": 4.15646234676007e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 134363296,
      "step": 8350
    },
    {
      "epoch": 0.5849729000849335,
      "grad_norm": 4.646234512329102,
      "learning_rate": 4.1557625218914186e-05,
      "loss": 1.0612,
      "num_input_tokens_seen": 134379376,
      "step": 8351
    },
    {
      "epoch": 0.5850429483306627,
      "grad_norm": 4.285154819488525,
      "learning_rate": 4.155062697022768e-05,
      "loss": 0.9446,
      "num_input_tokens_seen": 134395760,
      "step": 8352
    },
    {
      "epoch": 0.585112996576392,
      "grad_norm": 3.5311827659606934,
      "learning_rate": 4.154362872154116e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 134412144,
      "step": 8353
    },
    {
      "epoch": 0.5851830448221212,
      "grad_norm": 4.413166522979736,
      "learning_rate": 4.1536630472854645e-05,
      "loss": 0.9335,
      "num_input_tokens_seen": 134428136,
      "step": 8354
    },
    {
      "epoch": 0.5852530930678504,
      "grad_norm": 4.07955265045166,
      "learning_rate": 4.152963222416813e-05,
      "loss": 0.8923,
      "num_input_tokens_seen": 134444512,
      "step": 8355
    },
    {
      "epoch": 0.5853231413135798,
      "grad_norm": 3.809666633605957,
      "learning_rate": 4.1522633975481616e-05,
      "loss": 0.8746,
      "num_input_tokens_seen": 134460896,
      "step": 8356
    },
    {
      "epoch": 0.585393189559309,
      "grad_norm": 3.916811466217041,
      "learning_rate": 4.1515635726795104e-05,
      "loss": 0.9548,
      "num_input_tokens_seen": 134477128,
      "step": 8357
    },
    {
      "epoch": 0.5854632378050382,
      "grad_norm": 4.052529811859131,
      "learning_rate": 4.1508637478108586e-05,
      "loss": 1.0494,
      "num_input_tokens_seen": 134493512,
      "step": 8358
    },
    {
      "epoch": 0.5855332860507675,
      "grad_norm": 4.45082950592041,
      "learning_rate": 4.1501639229422075e-05,
      "loss": 1.2332,
      "num_input_tokens_seen": 134509192,
      "step": 8359
    },
    {
      "epoch": 0.5856033342964967,
      "grad_norm": 4.140877723693848,
      "learning_rate": 4.149464098073555e-05,
      "loss": 1.1588,
      "num_input_tokens_seen": 134524848,
      "step": 8360
    },
    {
      "epoch": 0.585673382542226,
      "grad_norm": 3.7743544578552246,
      "learning_rate": 4.148764273204904e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 134541232,
      "step": 8361
    },
    {
      "epoch": 0.5857434307879552,
      "grad_norm": 3.654794692993164,
      "learning_rate": 4.1480644483362534e-05,
      "loss": 1.2079,
      "num_input_tokens_seen": 134557616,
      "step": 8362
    },
    {
      "epoch": 0.5858134790336844,
      "grad_norm": 3.4448959827423096,
      "learning_rate": 4.147364623467601e-05,
      "loss": 0.8832,
      "num_input_tokens_seen": 134574000,
      "step": 8363
    },
    {
      "epoch": 0.5858835272794137,
      "grad_norm": 4.829925537109375,
      "learning_rate": 4.14666479859895e-05,
      "loss": 1.1541,
      "num_input_tokens_seen": 134590384,
      "step": 8364
    },
    {
      "epoch": 0.5859535755251429,
      "grad_norm": 3.5955686569213867,
      "learning_rate": 4.145964973730298e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 134606768,
      "step": 8365
    },
    {
      "epoch": 0.5860236237708721,
      "grad_norm": 5.0735368728637695,
      "learning_rate": 4.145265148861647e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 134622752,
      "step": 8366
    },
    {
      "epoch": 0.5860936720166015,
      "grad_norm": 3.8610787391662598,
      "learning_rate": 4.144565323992994e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 134639112,
      "step": 8367
    },
    {
      "epoch": 0.5861637202623307,
      "grad_norm": 3.630153179168701,
      "learning_rate": 4.143865499124344e-05,
      "loss": 0.9734,
      "num_input_tokens_seen": 134655496,
      "step": 8368
    },
    {
      "epoch": 0.58623376850806,
      "grad_norm": 4.367414951324463,
      "learning_rate": 4.143165674255693e-05,
      "loss": 1.0743,
      "num_input_tokens_seen": 134671880,
      "step": 8369
    },
    {
      "epoch": 0.5863038167537892,
      "grad_norm": 3.709831953048706,
      "learning_rate": 4.14246584938704e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 134688168,
      "step": 8370
    },
    {
      "epoch": 0.5863738649995184,
      "grad_norm": 4.649940490722656,
      "learning_rate": 4.141766024518389e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 134704552,
      "step": 8371
    },
    {
      "epoch": 0.5864439132452477,
      "grad_norm": 3.7334702014923096,
      "learning_rate": 4.141066199649737e-05,
      "loss": 0.7565,
      "num_input_tokens_seen": 134719896,
      "step": 8372
    },
    {
      "epoch": 0.5865139614909769,
      "grad_norm": 4.841366291046143,
      "learning_rate": 4.140366374781086e-05,
      "loss": 1.337,
      "num_input_tokens_seen": 134735168,
      "step": 8373
    },
    {
      "epoch": 0.5865840097367061,
      "grad_norm": 4.331602096557617,
      "learning_rate": 4.139666549912435e-05,
      "loss": 1.1276,
      "num_input_tokens_seen": 134751552,
      "step": 8374
    },
    {
      "epoch": 0.5866540579824354,
      "grad_norm": 4.3667497634887695,
      "learning_rate": 4.138966725043783e-05,
      "loss": 0.9588,
      "num_input_tokens_seen": 134767272,
      "step": 8375
    },
    {
      "epoch": 0.5867241062281646,
      "grad_norm": 3.6581032276153564,
      "learning_rate": 4.138266900175132e-05,
      "loss": 1.0778,
      "num_input_tokens_seen": 134783656,
      "step": 8376
    },
    {
      "epoch": 0.5867941544738939,
      "grad_norm": 3.8137452602386475,
      "learning_rate": 4.1375670753064795e-05,
      "loss": 1.1091,
      "num_input_tokens_seen": 134799760,
      "step": 8377
    },
    {
      "epoch": 0.5868642027196231,
      "grad_norm": 4.273350238800049,
      "learning_rate": 4.136867250437829e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 134816144,
      "step": 8378
    },
    {
      "epoch": 0.5869342509653523,
      "grad_norm": 4.527581214904785,
      "learning_rate": 4.136167425569178e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 134832528,
      "step": 8379
    },
    {
      "epoch": 0.5870042992110817,
      "grad_norm": 4.647453784942627,
      "learning_rate": 4.1354676007005254e-05,
      "loss": 1.0938,
      "num_input_tokens_seen": 134847984,
      "step": 8380
    },
    {
      "epoch": 0.5870743474568109,
      "grad_norm": 5.219020843505859,
      "learning_rate": 4.134767775831874e-05,
      "loss": 1.2273,
      "num_input_tokens_seen": 134864368,
      "step": 8381
    },
    {
      "epoch": 0.5871443957025402,
      "grad_norm": 4.104679107666016,
      "learning_rate": 4.1340679509632224e-05,
      "loss": 1.0368,
      "num_input_tokens_seen": 134879776,
      "step": 8382
    },
    {
      "epoch": 0.5872144439482694,
      "grad_norm": 4.659088134765625,
      "learning_rate": 4.133368126094571e-05,
      "loss": 1.258,
      "num_input_tokens_seen": 134894656,
      "step": 8383
    },
    {
      "epoch": 0.5872844921939986,
      "grad_norm": 5.709257125854492,
      "learning_rate": 4.13266830122592e-05,
      "loss": 1.0678,
      "num_input_tokens_seen": 134910952,
      "step": 8384
    },
    {
      "epoch": 0.5873545404397279,
      "grad_norm": 3.874393939971924,
      "learning_rate": 4.131968476357268e-05,
      "loss": 1.2776,
      "num_input_tokens_seen": 134927320,
      "step": 8385
    },
    {
      "epoch": 0.5874245886854571,
      "grad_norm": 3.5335848331451416,
      "learning_rate": 4.131268651488617e-05,
      "loss": 0.9319,
      "num_input_tokens_seen": 134943704,
      "step": 8386
    },
    {
      "epoch": 0.5874946369311863,
      "grad_norm": 7.110137462615967,
      "learning_rate": 4.130568826619965e-05,
      "loss": 1.0478,
      "num_input_tokens_seen": 134960088,
      "step": 8387
    },
    {
      "epoch": 0.5875646851769156,
      "grad_norm": 5.622186660766602,
      "learning_rate": 4.129869001751314e-05,
      "loss": 1.3012,
      "num_input_tokens_seen": 134976232,
      "step": 8388
    },
    {
      "epoch": 0.5876347334226448,
      "grad_norm": 4.596433162689209,
      "learning_rate": 4.129169176882663e-05,
      "loss": 1.1115,
      "num_input_tokens_seen": 134992616,
      "step": 8389
    },
    {
      "epoch": 0.5877047816683741,
      "grad_norm": 4.493381023406982,
      "learning_rate": 4.1284693520140106e-05,
      "loss": 0.9248,
      "num_input_tokens_seen": 135008704,
      "step": 8390
    },
    {
      "epoch": 0.5877748299141033,
      "grad_norm": 3.4309275150299072,
      "learning_rate": 4.1277695271453594e-05,
      "loss": 0.9195,
      "num_input_tokens_seen": 135024360,
      "step": 8391
    },
    {
      "epoch": 0.5878448781598326,
      "grad_norm": 3.7281200885772705,
      "learning_rate": 4.1270697022767076e-05,
      "loss": 1.1526,
      "num_input_tokens_seen": 135040744,
      "step": 8392
    },
    {
      "epoch": 0.5879149264055619,
      "grad_norm": 4.484415054321289,
      "learning_rate": 4.1263698774080565e-05,
      "loss": 0.9914,
      "num_input_tokens_seen": 135057128,
      "step": 8393
    },
    {
      "epoch": 0.5879849746512911,
      "grad_norm": 4.102346897125244,
      "learning_rate": 4.125670052539405e-05,
      "loss": 1.0439,
      "num_input_tokens_seen": 135073248,
      "step": 8394
    },
    {
      "epoch": 0.5880550228970203,
      "grad_norm": 7.703208923339844,
      "learning_rate": 4.1249702276707535e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 135088344,
      "step": 8395
    },
    {
      "epoch": 0.5881250711427496,
      "grad_norm": 6.113401889801025,
      "learning_rate": 4.1242704028021024e-05,
      "loss": 1.0845,
      "num_input_tokens_seen": 135104728,
      "step": 8396
    },
    {
      "epoch": 0.5881951193884788,
      "grad_norm": 3.8614649772644043,
      "learning_rate": 4.12357057793345e-05,
      "loss": 0.9821,
      "num_input_tokens_seen": 135120272,
      "step": 8397
    },
    {
      "epoch": 0.5882651676342081,
      "grad_norm": 3.9187567234039307,
      "learning_rate": 4.1228707530647994e-05,
      "loss": 0.9296,
      "num_input_tokens_seen": 135136152,
      "step": 8398
    },
    {
      "epoch": 0.5883352158799373,
      "grad_norm": 3.274703025817871,
      "learning_rate": 4.122170928196147e-05,
      "loss": 0.9896,
      "num_input_tokens_seen": 135152536,
      "step": 8399
    },
    {
      "epoch": 0.5884052641256665,
      "grad_norm": 3.8558948040008545,
      "learning_rate": 4.121471103327496e-05,
      "loss": 1.0359,
      "num_input_tokens_seen": 135168920,
      "step": 8400
    },
    {
      "epoch": 0.5884052641256665,
      "eval_loss": 1.1207953691482544,
      "eval_runtime": 0.5047,
      "eval_samples_per_second": 1.981,
      "eval_steps_per_second": 1.981,
      "num_input_tokens_seen": 135168920,
      "step": 8400
    }
  ],
  "logging_steps": 1,
  "max_steps": 14275,
  "num_input_tokens_seen": 135168920,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.9026020761447424e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}