{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 89.6328125,
      "epoch": 0.25,
      "grad_norm": 2.560847282409668,
      "kl": 0.0,
      "learning_rate": 9.99375e-07,
      "loss": 0.0,
      "reward": 1.43950754404068,
      "reward_std": 0.4176497310400009,
      "rewards/accuracy_reward": 0.5098200291395187,
      "rewards/format_reward": 0.9296875,
      "step": 1
    },
    {
      "completion_length": 87.2734375,
      "epoch": 0.5,
      "grad_norm": 2.703657865524292,
      "kl": 0.00046539306640625,
      "learning_rate": 9.9875e-07,
      "loss": 0.0,
      "reward": 1.2813043594360352,
      "reward_std": 0.47014792263507843,
      "rewards/accuracy_reward": 0.42192937433719635,
      "rewards/format_reward": 0.859375,
      "step": 2
    },
    {
      "completion_length": 80.0078125,
      "epoch": 0.75,
      "grad_norm": 5.132789134979248,
      "kl": 0.0010738372802734375,
      "learning_rate": 9.98125e-07,
      "loss": 0.0,
      "reward": 1.3763126730918884,
      "reward_std": 0.3502582609653473,
      "rewards/accuracy_reward": 0.4075627326965332,
      "rewards/format_reward": 0.96875,
      "step": 3
    },
    {
      "completion_length": 94.28571701049805,
      "epoch": 1.0,
      "grad_norm": 2.7119903564453125,
      "kl": 0.00128173828125,
      "learning_rate": 9.975e-07,
      "loss": 0.0001,
      "reward": 1.5021255612373352,
      "reward_std": 0.3990684002637863,
      "rewards/accuracy_reward": 0.502125546336174,
      "rewards/format_reward": 1.0,
      "step": 4
    },
    {
      "completion_length": 84.453125,
      "epoch": 1.25,
      "grad_norm": 2.7416577339172363,
      "kl": 0.0021209716796875,
      "learning_rate": 9.968749999999999e-07,
      "loss": 0.0001,
      "reward": 1.3453806638717651,
      "reward_std": 0.4074166566133499,
      "rewards/accuracy_reward": 0.4000682085752487,
      "rewards/format_reward": 0.9453125,
      "step": 5
    },
    {
      "completion_length": 78.625,
      "epoch": 1.5,
      "grad_norm": 2.8765275478363037,
      "kl": 0.006072998046875,
      "learning_rate": 9.9625e-07,
      "loss": 0.0002,
      "reward": 1.4404960870742798,
      "reward_std": 0.3917630910873413,
      "rewards/accuracy_reward": 0.44830864667892456,
      "rewards/format_reward": 0.9921875,
      "step": 6
    },
    {
      "completion_length": 70.7578125,
      "epoch": 1.75,
      "grad_norm": 4.0928168296813965,
      "kl": 0.0057220458984375,
      "learning_rate": 9.956249999999999e-07,
      "loss": 0.0002,
      "reward": 1.5452297925949097,
      "reward_std": 0.2363404855132103,
      "rewards/accuracy_reward": 0.5608547776937485,
      "rewards/format_reward": 0.984375,
      "step": 7
    },
    {
      "completion_length": 69.57143211364746,
      "epoch": 2.0,
      "grad_norm": 2.1500496864318848,
      "kl": 0.00787353515625,
      "learning_rate": 9.95e-07,
      "loss": 0.0003,
      "reward": 1.540364921092987,
      "reward_std": 0.03179515106603503,
      "rewards/accuracy_reward": 0.5403649136424065,
      "rewards/format_reward": 1.0,
      "step": 8
    },
    {
      "completion_length": 74.5234375,
      "epoch": 2.25,
      "grad_norm": 2.890000343322754,
      "kl": 0.009307861328125,
      "learning_rate": 9.94375e-07,
      "loss": 0.0004,
      "reward": 1.4441289901733398,
      "reward_std": 0.2048807591199875,
      "rewards/accuracy_reward": 0.44412901997566223,
      "rewards/format_reward": 1.0,
      "step": 9
    },
    {
      "completion_length": 73.109375,
      "epoch": 2.5,
      "grad_norm": 2.234746217727661,
      "kl": 0.00933837890625,
      "learning_rate": 9.9375e-07,
      "loss": 0.0004,
      "reward": 1.4635842442512512,
      "reward_std": 0.2971457466483116,
      "rewards/accuracy_reward": 0.46358419954776764,
      "rewards/format_reward": 1.0,
      "step": 10
    },
    {
      "completion_length": 67.8203125,
      "epoch": 2.75,
      "grad_norm": 2.0221376419067383,
      "kl": 0.018035888671875,
      "learning_rate": 9.93125e-07,
      "loss": 0.0007,
      "reward": 1.6205239295959473,
      "reward_std": 0.18765632808208466,
      "rewards/accuracy_reward": 0.6205238401889801,
      "rewards/format_reward": 1.0,
      "step": 11
    },
    {
      "completion_length": 75.00000381469727,
      "epoch": 3.0,
      "grad_norm": 2.337059259414673,
      "kl": 0.014556884765625,
      "learning_rate": 9.925e-07,
      "loss": 0.0006,
      "reward": 1.785714328289032,
      "reward_std": 0.26726123690605164,
      "rewards/accuracy_reward": 0.7857142984867096,
      "rewards/format_reward": 1.0,
      "step": 12
    },
    {
      "completion_length": 67.9765625,
      "epoch": 3.25,
      "grad_norm": 2.605905294418335,
      "kl": 0.0230712890625,
      "learning_rate": 9.91875e-07,
      "loss": 0.0009,
      "reward": 1.5252612233161926,
      "reward_std": 0.2860804498195648,
      "rewards/accuracy_reward": 0.5330736935138702,
      "rewards/format_reward": 0.9921875,
      "step": 13
    },
    {
      "completion_length": 71.0703125,
      "epoch": 3.5,
      "grad_norm": 2.2355077266693115,
      "kl": 0.02154541015625,
      "learning_rate": 9.912499999999998e-07,
      "loss": 0.0009,
      "reward": 1.5774829387664795,
      "reward_std": 0.1989663988351822,
      "rewards/accuracy_reward": 0.5774829983711243,
      "rewards/format_reward": 1.0,
      "step": 14
    },
    {
      "completion_length": 65.875,
      "epoch": 3.75,
      "grad_norm": 2.508030652999878,
      "kl": 0.0369873046875,
      "learning_rate": 9.90625e-07,
      "loss": 0.0015,
      "reward": 1.6362086534500122,
      "reward_std": 0.2645450085401535,
      "rewards/accuracy_reward": 0.6362085938453674,
      "rewards/format_reward": 1.0,
      "step": 15
    },
    {
      "completion_length": 58.00000190734863,
      "epoch": 4.0,
      "grad_norm": 2.397307872772217,
      "kl": 0.02752685546875,
      "learning_rate": 9.9e-07,
      "loss": 0.001,
      "reward": 1.837504506111145,
      "reward_std": 0.34773190319538116,
      "rewards/accuracy_reward": 0.8375044763088226,
      "rewards/format_reward": 1.0,
      "step": 16
    },
    {
      "completion_length": 62.1328125,
      "epoch": 4.25,
      "grad_norm": 1.985022783279419,
      "kl": 0.02044677734375,
      "learning_rate": 9.89375e-07,
      "loss": 0.0008,
      "reward": 1.8274397253990173,
      "reward_std": 0.15590714663267136,
      "rewards/accuracy_reward": 0.8274396657943726,
      "rewards/format_reward": 1.0,
      "step": 17
    },
    {
      "completion_length": 66.078125,
      "epoch": 4.5,
      "grad_norm": 2.398780584335327,
      "kl": 0.0401611328125,
      "learning_rate": 9.8875e-07,
      "loss": 0.0016,
      "reward": 1.642715036869049,
      "reward_std": 0.19355066865682602,
      "rewards/accuracy_reward": 0.6427150070667267,
      "rewards/format_reward": 1.0,
      "step": 18
    },
    {
      "completion_length": 66.0625,
      "epoch": 4.75,
      "grad_norm": 2.3296172618865967,
      "kl": 0.048095703125,
      "learning_rate": 9.88125e-07,
      "loss": 0.0019,
      "reward": 1.6349385380744934,
      "reward_std": 0.30714260041713715,
      "rewards/accuracy_reward": 0.6427510678768158,
      "rewards/format_reward": 0.9921875,
      "step": 19
    },
    {
      "completion_length": 73.50000381469727,
      "epoch": 5.0,
      "grad_norm": 2.703334331512451,
      "kl": 0.03143310546875,
      "learning_rate": 9.875e-07,
      "loss": 0.0012,
      "reward": 1.521121323108673,
      "reward_std": 0.409614622592926,
      "rewards/accuracy_reward": 0.5211213529109955,
      "rewards/format_reward": 1.0,
      "step": 20
    },
    {
      "completion_length": 63.859375,
      "epoch": 5.25,
      "grad_norm": 3.7955989837646484,
      "kl": 0.0419921875,
      "learning_rate": 9.86875e-07,
      "loss": 0.0017,
      "reward": 1.6154157519340515,
      "reward_std": 0.17696820944547653,
      "rewards/accuracy_reward": 0.6154157221317291,
      "rewards/format_reward": 1.0,
      "step": 21
    },
    {
      "completion_length": 62.3359375,
      "epoch": 5.5,
      "grad_norm": 2.0224432945251465,
      "kl": 0.04486083984375,
      "learning_rate": 9.862499999999999e-07,
      "loss": 0.0018,
      "reward": 1.7184607982635498,
      "reward_std": 0.16713028401136398,
      "rewards/accuracy_reward": 0.7184608280658722,
      "rewards/format_reward": 1.0,
      "step": 22
    },
    {
      "completion_length": 65.8515625,
      "epoch": 5.75,
      "grad_norm": 2.972362756729126,
      "kl": 0.0313720703125,
      "learning_rate": 9.85625e-07,
      "loss": 0.0013,
      "reward": 1.7493125200271606,
      "reward_std": 0.184370219707489,
      "rewards/accuracy_reward": 0.7493124902248383,
      "rewards/format_reward": 1.0,
      "step": 23
    },
    {
      "completion_length": 74.21429061889648,
      "epoch": 6.0,
      "grad_norm": 1.9288957118988037,
      "kl": 0.0411376953125,
      "learning_rate": 9.849999999999999e-07,
      "loss": 0.0019,
      "reward": 1.6142857074737549,
      "reward_std": 0.26726123690605164,
      "rewards/accuracy_reward": 0.6142857670783997,
      "rewards/format_reward": 1.0,
      "step": 24
    },
    {
      "completion_length": 61.2265625,
      "epoch": 6.25,
      "grad_norm": 7.291033744812012,
      "kl": 0.260986328125,
      "learning_rate": 9.84375e-07,
      "loss": 0.0104,
      "reward": 1.7374743819236755,
      "reward_std": 0.2810707241296768,
      "rewards/accuracy_reward": 0.7374744713306427,
      "rewards/format_reward": 1.0,
      "step": 25
    },
    {
      "completion_length": 60.390625,
      "epoch": 6.5,
      "grad_norm": 2.046499013900757,
      "kl": 0.03759765625,
      "learning_rate": 9.8375e-07,
      "loss": 0.0015,
      "reward": 1.663317084312439,
      "reward_std": 0.216216042637825,
      "rewards/accuracy_reward": 0.6633170545101166,
      "rewards/format_reward": 1.0,
      "step": 26
    },
    {
      "completion_length": 60.2421875,
      "epoch": 6.75,
      "grad_norm": 2.2611141204833984,
      "kl": 0.046875,
      "learning_rate": 9.83125e-07,
      "loss": 0.0019,
      "reward": 1.5771763920783997,
      "reward_std": 0.2472986802458763,
      "rewards/accuracy_reward": 0.5849888920783997,
      "rewards/format_reward": 0.9921875,
      "step": 27
    },
    {
      "completion_length": 64.28571891784668,
      "epoch": 7.0,
      "grad_norm": 2.10233736038208,
      "kl": 0.086181640625,
      "learning_rate": 9.825e-07,
      "loss": 0.0036,
      "reward": 1.82956200838089,
      "reward_std": 0.3444478511810303,
      "rewards/accuracy_reward": 0.8295620679855347,
      "rewards/format_reward": 1.0,
      "step": 28
    },
    {
      "completion_length": 63.015625,
      "epoch": 7.25,
      "grad_norm": 2.0856142044067383,
      "kl": 0.0609130859375,
      "learning_rate": 9.81875e-07,
      "loss": 0.0024,
      "reward": 1.6384294033050537,
      "reward_std": 0.20048531889915466,
      "rewards/accuracy_reward": 0.6384294033050537,
      "rewards/format_reward": 1.0,
      "step": 29
    },
    {
      "completion_length": 56.7109375,
      "epoch": 7.5,
      "grad_norm": 1.9775277376174927,
      "kl": 0.06494140625,
      "learning_rate": 9.8125e-07,
      "loss": 0.0026,
      "reward": 1.726142942905426,
      "reward_std": 0.183742456138134,
      "rewards/accuracy_reward": 0.733955442905426,
      "rewards/format_reward": 0.9921875,
      "step": 30
    },
    {
      "completion_length": 62.640625,
      "epoch": 7.75,
      "grad_norm": 4.462751388549805,
      "kl": 0.0439453125,
      "learning_rate": 9.806249999999998e-07,
      "loss": 0.0018,
      "reward": 1.7479652166366577,
      "reward_std": 0.23261219263076782,
      "rewards/accuracy_reward": 0.7479651868343353,
      "rewards/format_reward": 1.0,
      "step": 31
    },
    {
      "completion_length": 56.64285850524902,
      "epoch": 8.0,
      "grad_norm": 1.7369378805160522,
      "kl": 0.067138671875,
      "learning_rate": 9.8e-07,
      "loss": 0.0026,
      "reward": 1.9810991883277893,
      "reward_std": 0.024930346757173538,
      "rewards/accuracy_reward": 0.9810990691184998,
      "rewards/format_reward": 1.0,
      "step": 32
    },
    {
      "completion_length": 54.9609375,
      "epoch": 8.25,
      "grad_norm": 2.8938117027282715,
      "kl": 0.0703125,
      "learning_rate": 9.79375e-07,
      "loss": 0.0028,
      "reward": 1.7204629778862,
      "reward_std": 0.16244513541460037,
      "rewards/accuracy_reward": 0.7204630076885223,
      "rewards/format_reward": 1.0,
      "step": 33
    },
    {
      "completion_length": 56.1640625,
      "epoch": 8.5,
      "grad_norm": 1.817608118057251,
      "kl": 0.0526123046875,
      "learning_rate": 9.7875e-07,
      "loss": 0.0021,
      "reward": 1.8434885740280151,
      "reward_std": 0.13239304721355438,
      "rewards/accuracy_reward": 0.8434885144233704,
      "rewards/format_reward": 1.0,
      "step": 34
    },
    {
      "completion_length": 61.7421875,
      "epoch": 8.75,
      "grad_norm": 2.240640878677368,
      "kl": 0.0614013671875,
      "learning_rate": 9.78125e-07,
      "loss": 0.0025,
      "reward": 1.6918946504592896,
      "reward_std": 0.23886261880397797,
      "rewards/accuracy_reward": 0.6918946206569672,
      "rewards/format_reward": 1.0,
      "step": 35
    },
    {
      "completion_length": 67.07143020629883,
      "epoch": 9.0,
      "grad_norm": 2.1461708545684814,
      "kl": 0.0511474609375,
      "learning_rate": 9.775e-07,
      "loss": 0.0023,
      "reward": 1.5855827927589417,
      "reward_std": 0.4357292503118515,
      "rewards/accuracy_reward": 0.6570113599300385,
      "rewards/format_reward": 0.9285714626312256,
      "step": 36
    },
    {
      "completion_length": 59.109375,
      "epoch": 9.25,
      "grad_norm": 1.676393747329712,
      "kl": 0.07080078125,
      "learning_rate": 9.76875e-07,
      "loss": 0.0028,
      "reward": 1.7769874930381775,
      "reward_std": 0.18539611995220184,
      "rewards/accuracy_reward": 0.7769874632358551,
      "rewards/format_reward": 1.0,
      "step": 37
    },
    {
      "completion_length": 59.2109375,
      "epoch": 9.5,
      "grad_norm": 2.115464925765991,
      "kl": 0.0623779296875,
      "learning_rate": 9.7625e-07,
      "loss": 0.0025,
      "reward": 1.6574002504348755,
      "reward_std": 0.23898707330226898,
      "rewards/accuracy_reward": 0.6574002504348755,
      "rewards/format_reward": 1.0,
      "step": 38
    },
    {
      "completion_length": 58.7734375,
      "epoch": 9.75,
      "grad_norm": 2.0188374519348145,
      "kl": 0.0640869140625,
      "learning_rate": 9.756249999999999e-07,
      "loss": 0.0026,
      "reward": 1.8230915069580078,
      "reward_std": 0.07023201137781143,
      "rewards/accuracy_reward": 0.8230914771556854,
      "rewards/format_reward": 1.0,
      "step": 39
    },
    {
      "completion_length": 50.57143020629883,
      "epoch": 10.0,
      "grad_norm": 1.818213701248169,
      "kl": 0.069580078125,
      "learning_rate": 9.75e-07,
      "loss": 0.0027,
      "reward": 1.898565948009491,
      "reward_std": 0.1666813576593995,
      "rewards/accuracy_reward": 0.8985659182071686,
      "rewards/format_reward": 1.0,
      "step": 40
    },
    {
      "completion_length": 61.828125,
      "epoch": 10.25,
      "grad_norm": 2.0514211654663086,
      "kl": 0.054443359375,
      "learning_rate": 9.743749999999999e-07,
      "loss": 0.0022,
      "reward": 1.770473837852478,
      "reward_std": 0.26378537714481354,
      "rewards/accuracy_reward": 0.7860988080501556,
      "rewards/format_reward": 0.984375,
      "step": 41
    },
    {
      "completion_length": 62.3203125,
      "epoch": 10.5,
      "grad_norm": 1.9126454591751099,
      "kl": 0.0628662109375,
      "learning_rate": 9.7375e-07,
      "loss": 0.0025,
      "reward": 1.8098745346069336,
      "reward_std": 0.158866249024868,
      "rewards/accuracy_reward": 0.8176870048046112,
      "rewards/format_reward": 0.9921875,
      "step": 42
    },
    {
      "completion_length": 58.640625,
      "epoch": 10.75,
      "grad_norm": 1.7069599628448486,
      "kl": 0.0595703125,
      "learning_rate": 9.73125e-07,
      "loss": 0.0024,
      "reward": 1.7534176111221313,
      "reward_std": 0.15397731214761734,
      "rewards/accuracy_reward": 0.7534177303314209,
      "rewards/format_reward": 1.0,
      "step": 43
    },
    {
      "completion_length": 51.21428871154785,
      "epoch": 11.0,
      "grad_norm": 2.326500415802002,
      "kl": 0.0687255859375,
      "learning_rate": 9.725e-07,
      "loss": 0.0027,
      "reward": 1.612587034702301,
      "reward_std": 0.19589456543326378,
      "rewards/accuracy_reward": 0.6125869750976562,
      "rewards/format_reward": 1.0,
      "step": 44
    },
    {
      "completion_length": 63.1953125,
      "epoch": 11.25,
      "grad_norm": 2.318408966064453,
      "kl": 0.05615234375,
      "learning_rate": 9.71875e-07,
      "loss": 0.0022,
      "reward": 1.731492519378662,
      "reward_std": 0.1684369444847107,
      "rewards/accuracy_reward": 0.7314925193786621,
      "rewards/format_reward": 1.0,
      "step": 45
    },
    {
      "completion_length": 62.1796875,
      "epoch": 11.5,
      "grad_norm": 2.0479931831359863,
      "kl": 0.067138671875,
      "learning_rate": 9.712499999999998e-07,
      "loss": 0.0027,
      "reward": 1.8079904317855835,
      "reward_std": 0.14444740116596222,
      "rewards/accuracy_reward": 0.8079904615879059,
      "rewards/format_reward": 1.0,
      "step": 46
    },
    {
      "completion_length": 58.5234375,
      "epoch": 11.75,
      "grad_norm": 2.1100575923919678,
      "kl": 0.0506591796875,
      "learning_rate": 9.70625e-07,
      "loss": 0.002,
      "reward": 1.849604606628418,
      "reward_std": 0.1187722496688366,
      "rewards/accuracy_reward": 0.8496046662330627,
      "rewards/format_reward": 1.0,
      "step": 47
    },
    {
      "completion_length": 78.5714340209961,
      "epoch": 12.0,
      "grad_norm": 2.278730630874634,
      "kl": 0.0589599609375,
      "learning_rate": 9.7e-07,
      "loss": 0.0023,
      "reward": 1.650295615196228,
      "reward_std": 0.3869354873895645,
      "rewards/accuracy_reward": 0.6502955406904221,
      "rewards/format_reward": 1.0,
      "step": 48
    },
    {
      "completion_length": 62.8203125,
      "epoch": 12.25,
      "grad_norm": 1.7048614025115967,
      "kl": 0.0550537109375,
      "learning_rate": 9.69375e-07,
      "loss": 0.0022,
      "reward": 1.8256508708000183,
      "reward_std": 0.12432926893234253,
      "rewards/accuracy_reward": 0.8256509006023407,
      "rewards/format_reward": 1.0,
      "step": 49
    },
    {
      "completion_length": 67.4375,
      "epoch": 12.5,
      "grad_norm": 1.5910967588424683,
      "kl": 0.0621337890625,
      "learning_rate": 9.6875e-07,
      "loss": 0.0025,
      "reward": 1.801437258720398,
      "reward_std": 0.12012555077672005,
      "rewards/accuracy_reward": 0.8014372885227203,
      "rewards/format_reward": 1.0,
      "step": 50
    },
    {
      "completion_length": 60.3515625,
      "epoch": 12.75,
      "grad_norm": 1.7540335655212402,
      "kl": 0.07177734375,
      "learning_rate": 9.68125e-07,
      "loss": 0.0029,
      "reward": 1.7970203161239624,
      "reward_std": 0.09028816036880016,
      "rewards/accuracy_reward": 0.79702028632164,
      "rewards/format_reward": 1.0,
      "step": 51
    },
    {
      "completion_length": 68.85714721679688,
      "epoch": 13.0,
      "grad_norm": 1.9271409511566162,
      "kl": 0.057861328125,
      "learning_rate": 9.675e-07,
      "loss": 0.0022,
      "reward": 1.725570797920227,
      "reward_std": 0.2023605689755641,
      "rewards/accuracy_reward": 0.725570797920227,
      "rewards/format_reward": 1.0,
      "step": 52
    },
    {
      "completion_length": 64.7890625,
      "epoch": 13.25,
      "grad_norm": 1.6656194925308228,
      "kl": 0.057861328125,
      "learning_rate": 9.66875e-07,
      "loss": 0.0023,
      "reward": 1.762970507144928,
      "reward_std": 0.1429205760359764,
      "rewards/accuracy_reward": 0.7629704773426056,
      "rewards/format_reward": 1.0,
      "step": 53
    },
    {
      "completion_length": 69.1796875,
      "epoch": 13.5,
      "grad_norm": 1.7760093212127686,
      "kl": 0.084228515625,
      "learning_rate": 9.6625e-07,
      "loss": 0.0034,
      "reward": 1.7609952092170715,
      "reward_std": 0.11301954090595245,
      "rewards/accuracy_reward": 0.7609952092170715,
      "rewards/format_reward": 1.0,
      "step": 54
    },
    {
      "completion_length": 65.984375,
      "epoch": 13.75,
      "grad_norm": 1.944703459739685,
      "kl": 0.0472412109375,
      "learning_rate": 9.65625e-07,
      "loss": 0.0019,
      "reward": 1.8564435839653015,
      "reward_std": 0.14264069870114326,
      "rewards/accuracy_reward": 0.8564436435699463,
      "rewards/format_reward": 1.0,
      "step": 55
    },
    {
      "completion_length": 70.85714340209961,
      "epoch": 14.0,
      "grad_norm": 1.7048606872558594,
      "kl": 0.059326171875,
      "learning_rate": 9.649999999999999e-07,
      "loss": 0.0023,
      "reward": 1.8061460256576538,
      "reward_std": 0.2514180298894644,
      "rewards/accuracy_reward": 0.8061459064483643,
      "rewards/format_reward": 1.0,
      "step": 56
    },
    {
      "completion_length": 64.2421875,
      "epoch": 14.25,
      "grad_norm": 1.8009275197982788,
      "kl": 0.083251953125,
      "learning_rate": 9.64375e-07,
      "loss": 0.0033,
      "reward": 1.7855259776115417,
      "reward_std": 0.08876464702188969,
      "rewards/accuracy_reward": 0.785525918006897,
      "rewards/format_reward": 1.0,
      "step": 57
    },
    {
      "completion_length": 68.515625,
      "epoch": 14.5,
      "grad_norm": 1.522387981414795,
      "kl": 0.074951171875,
      "learning_rate": 9.637499999999999e-07,
      "loss": 0.003,
      "reward": 1.8352751731872559,
      "reward_std": 0.12848591804504395,
      "rewards/accuracy_reward": 0.8352752029895782,
      "rewards/format_reward": 1.0,
      "step": 58
    },
    {
      "completion_length": 67.0390625,
      "epoch": 14.75,
      "grad_norm": 2.3351709842681885,
      "kl": 0.065673828125,
      "learning_rate": 9.63125e-07,
      "loss": 0.0026,
      "reward": 1.8882685899734497,
      "reward_std": 0.11743934452533722,
      "rewards/accuracy_reward": 0.8882685601711273,
      "rewards/format_reward": 1.0,
      "step": 59
    },
    {
      "completion_length": 58.71428871154785,
      "epoch": 15.0,
      "grad_norm": 1.921271800994873,
      "kl": 0.05419921875,
      "learning_rate": 9.624999999999999e-07,
      "loss": 0.0022,
      "reward": 1.53987056016922,
      "reward_std": 0.16531461104750633,
      "rewards/accuracy_reward": 0.5398704707622528,
      "rewards/format_reward": 1.0,
      "step": 60
    },
    {
      "completion_length": 68.25,
      "epoch": 15.25,
      "grad_norm": 1.8104687929153442,
      "kl": 0.060302734375,
      "learning_rate": 9.61875e-07,
      "loss": 0.0024,
      "reward": 1.8509221076965332,
      "reward_std": 0.11642135679721832,
      "rewards/accuracy_reward": 0.8509220480918884,
      "rewards/format_reward": 1.0,
      "step": 61
    },
    {
      "completion_length": 67.453125,
      "epoch": 15.5,
      "grad_norm": 2.454641819000244,
      "kl": 0.0623779296875,
      "learning_rate": 9.6125e-07,
      "loss": 0.0025,
      "reward": 1.7732171416282654,
      "reward_std": 0.1827603131532669,
      "rewards/accuracy_reward": 0.7732171416282654,
      "rewards/format_reward": 1.0,
      "step": 62
    },
    {
      "completion_length": 70.375,
      "epoch": 15.75,
      "grad_norm": 3.006380558013916,
      "kl": 0.123046875,
      "learning_rate": 9.606249999999998e-07,
      "loss": 0.0049,
      "reward": 1.774861454963684,
      "reward_std": 0.1715347319841385,
      "rewards/accuracy_reward": 0.7748615145683289,
      "rewards/format_reward": 1.0,
      "step": 63
    },
    {
      "completion_length": 56.142860412597656,
      "epoch": 16.0,
      "grad_norm": 3.028219223022461,
      "kl": 0.0634765625,
      "learning_rate": 9.6e-07,
      "loss": 0.0025,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 64
    },
    {
      "completion_length": 74.2890625,
      "epoch": 16.25,
      "grad_norm": 1.6325531005859375,
      "kl": 0.06103515625,
      "learning_rate": 9.59375e-07,
      "loss": 0.0024,
      "reward": 1.7759021520614624,
      "reward_std": 0.15456650406122208,
      "rewards/accuracy_reward": 0.7759020924568176,
      "rewards/format_reward": 1.0,
      "step": 65
    },
    {
      "completion_length": 69.484375,
      "epoch": 16.5,
      "grad_norm": 1.501849889755249,
      "kl": 0.068115234375,
      "learning_rate": 9.5875e-07,
      "loss": 0.0027,
      "reward": 1.7950308322906494,
      "reward_std": 0.12456715479493141,
      "rewards/accuracy_reward": 0.7950307726860046,
      "rewards/format_reward": 1.0,
      "step": 66
    },
    {
      "completion_length": 64.953125,
      "epoch": 16.75,
      "grad_norm": 2.5453922748565674,
      "kl": 0.0596923828125,
      "learning_rate": 9.58125e-07,
      "loss": 0.0024,
      "reward": 1.8739762902259827,
      "reward_std": 0.11443666741251945,
      "rewards/accuracy_reward": 0.8817887902259827,
      "rewards/format_reward": 0.9921875,
      "step": 67
    },
    {
      "completion_length": 61.42857551574707,
      "epoch": 17.0,
      "grad_norm": 1.8238996267318726,
      "kl": 0.04443359375,
      "learning_rate": 9.575e-07,
      "loss": 0.0017,
      "reward": 1.8428571224212646,
      "reward_std": 0.1414213627576828,
      "rewards/accuracy_reward": 0.8428571820259094,
      "rewards/format_reward": 1.0,
      "step": 68
    },
    {
      "completion_length": 72.0234375,
      "epoch": 17.25,
      "grad_norm": 1.7668324708938599,
      "kl": 0.07373046875,
      "learning_rate": 9.56875e-07,
      "loss": 0.003,
      "reward": 1.8650294542312622,
      "reward_std": 0.13415485620498657,
      "rewards/accuracy_reward": 0.8650294542312622,
      "rewards/format_reward": 1.0,
      "step": 69
    },
    {
      "completion_length": 64.9453125,
      "epoch": 17.5,
      "grad_norm": 1.4389671087265015,
      "kl": 0.047607421875,
      "learning_rate": 9.5625e-07,
      "loss": 0.0019,
      "reward": 1.927801787853241,
      "reward_std": 0.06346799433231354,
      "rewards/accuracy_reward": 0.927801787853241,
      "rewards/format_reward": 1.0,
      "step": 70
    },
    {
      "completion_length": 67.9296875,
      "epoch": 17.75,
      "grad_norm": 1.4634050130844116,
      "kl": 0.072509765625,
      "learning_rate": 9.556249999999999e-07,
      "loss": 0.0029,
      "reward": 1.8554179668426514,
      "reward_std": 0.13238264620304108,
      "rewards/accuracy_reward": 0.8554179966449738,
      "rewards/format_reward": 1.0,
      "step": 71
    },
    {
      "completion_length": 57.07143020629883,
      "epoch": 18.0,
      "grad_norm": 4.220071792602539,
      "kl": 0.17724609375,
      "learning_rate": 9.55e-07,
      "loss": 0.0064,
      "reward": 1.9889448285102844,
      "reward_std": 0.006739838980138302,
      "rewards/accuracy_reward": 0.988944798707962,
      "rewards/format_reward": 1.0,
      "step": 72
    },
    {
      "completion_length": 68.5546875,
      "epoch": 18.25,
      "grad_norm": 1.6068540811538696,
      "kl": 0.0767822265625,
      "learning_rate": 9.543749999999999e-07,
      "loss": 0.0031,
      "reward": 1.8943632245063782,
      "reward_std": 0.07841086108237505,
      "rewards/accuracy_reward": 0.8943631649017334,
      "rewards/format_reward": 1.0,
      "step": 73
    },
    {
      "completion_length": 68.2421875,
      "epoch": 18.5,
      "grad_norm": 1.6987547874450684,
      "kl": 0.07958984375,
      "learning_rate": 9.5375e-07,
      "loss": 0.0032,
      "reward": 1.8070343136787415,
      "reward_std": 0.14604970812797546,
      "rewards/accuracy_reward": 0.8304717838764191,
      "rewards/format_reward": 0.9765625,
      "step": 74
    },
    {
      "completion_length": 65.3046875,
      "epoch": 18.75,
      "grad_norm": 1.6191295385360718,
      "kl": 0.04541015625,
      "learning_rate": 9.53125e-07,
      "loss": 0.0018,
      "reward": 1.7468606233596802,
      "reward_std": 0.16931980848312378,
      "rewards/accuracy_reward": 0.7468606233596802,
      "rewards/format_reward": 1.0,
      "step": 75
    },
    {
      "completion_length": 50.78571891784668,
      "epoch": 19.0,
      "grad_norm": 1.6109150648117065,
      "kl": 0.066162109375,
      "learning_rate": 9.525e-07,
      "loss": 0.0027,
      "reward": 1.9664621353149414,
      "reward_std": 0.00548175536096096,
      "rewards/accuracy_reward": 0.9664620459079742,
      "rewards/format_reward": 1.0,
      "step": 76
    },
    {
      "completion_length": 66.2109375,
      "epoch": 19.25,
      "grad_norm": 1.385863184928894,
      "kl": 0.059326171875,
      "learning_rate": 9.51875e-07,
      "loss": 0.0024,
      "reward": 1.849595844745636,
      "reward_std": 0.0718111265450716,
      "rewards/accuracy_reward": 0.849595844745636,
      "rewards/format_reward": 1.0,
      "step": 77
    },
    {
      "completion_length": 75.578125,
      "epoch": 19.5,
      "grad_norm": 1.782371163368225,
      "kl": 0.073974609375,
      "learning_rate": 9.5125e-07,
      "loss": 0.003,
      "reward": 1.8164063096046448,
      "reward_std": 0.16241852939128876,
      "rewards/accuracy_reward": 0.8320313096046448,
      "rewards/format_reward": 0.984375,
      "step": 78
    },
    {
      "completion_length": 69.609375,
      "epoch": 19.75,
      "grad_norm": 1.2934050559997559,
      "kl": 0.0748291015625,
      "learning_rate": 9.50625e-07,
      "loss": 0.003,
      "reward": 1.8654966354370117,
      "reward_std": 0.09261503256857395,
      "rewards/accuracy_reward": 0.8654966354370117,
      "rewards/format_reward": 1.0,
      "step": 79
    },
    {
      "completion_length": 58.00000190734863,
      "epoch": 20.0,
      "grad_norm": 1.742002248764038,
      "kl": 0.082275390625,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.0033,
      "reward": 1.8342429995536804,
      "reward_std": 0.2157389521598816,
      "rewards/accuracy_reward": 0.8342429995536804,
      "rewards/format_reward": 1.0,
      "step": 80
    },
    {
      "completion_length": 67.65625,
      "epoch": 20.25,
      "grad_norm": 1.1428124904632568,
      "kl": 0.0526123046875,
      "learning_rate": 9.493749999999999e-07,
      "loss": 0.0021,
      "reward": 1.9487730264663696,
      "reward_std": 0.07040097191929817,
      "rewards/accuracy_reward": 0.9565855264663696,
      "rewards/format_reward": 0.9921875,
      "step": 81
    },
    {
      "completion_length": 73.515625,
      "epoch": 20.5,
      "grad_norm": 2.601008653640747,
      "kl": 0.0626220703125,
      "learning_rate": 9.487499999999999e-07,
      "loss": 0.0025,
      "reward": 1.770107924938202,
      "reward_std": 0.11608634144067764,
      "rewards/accuracy_reward": 0.7701078951358795,
      "rewards/format_reward": 1.0,
      "step": 82
    },
    {
      "completion_length": 70.1640625,
      "epoch": 20.75,
      "grad_norm": 1.5553102493286133,
      "kl": 0.07421875,
      "learning_rate": 9.481249999999999e-07,
      "loss": 0.003,
      "reward": 1.8716879487037659,
      "reward_std": 0.13853080570697784,
      "rewards/accuracy_reward": 0.8716880083084106,
      "rewards/format_reward": 1.0,
      "step": 83
    },
    {
      "completion_length": 100.9285774230957,
      "epoch": 21.0,
      "grad_norm": 1.4347081184387207,
      "kl": 0.081298828125,
      "learning_rate": 9.474999999999999e-07,
      "loss": 0.0032,
      "reward": 1.7912707328796387,
      "reward_std": 0.22874768637120724,
      "rewards/accuracy_reward": 0.7912707030773163,
      "rewards/format_reward": 1.0,
      "step": 84
    },
    {
      "completion_length": 75.5078125,
      "epoch": 21.25,
      "grad_norm": 1.4661049842834473,
      "kl": 0.05859375,
      "learning_rate": 9.468749999999999e-07,
      "loss": 0.0023,
      "reward": 1.7916218042373657,
      "reward_std": 0.07297072582878172,
      "rewards/accuracy_reward": 0.7916218042373657,
      "rewards/format_reward": 1.0,
      "step": 85
    },
    {
      "completion_length": 67.9296875,
      "epoch": 21.5,
      "grad_norm": 1.3486883640289307,
      "kl": 0.0849609375,
      "learning_rate": 9.462499999999999e-07,
      "loss": 0.0034,
      "reward": 1.9112024307250977,
      "reward_std": 0.07055860431864858,
      "rewards/accuracy_reward": 0.9112024307250977,
      "rewards/format_reward": 1.0,
      "step": 86
    },
    {
      "completion_length": 83.8359375,
      "epoch": 21.75,
      "grad_norm": 1.7671387195587158,
      "kl": 0.062744140625,
      "learning_rate": 9.45625e-07,
      "loss": 0.0025,
      "reward": 1.8552258610725403,
      "reward_std": 0.14477503299713135,
      "rewards/accuracy_reward": 0.8630383908748627,
      "rewards/format_reward": 0.9921875,
      "step": 87
    },
    {
      "completion_length": 68.42857360839844,
      "epoch": 22.0,
      "grad_norm": 1.4051504135131836,
      "kl": 0.0599365234375,
      "learning_rate": 9.45e-07,
      "loss": 0.0023,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 88
    },
    {
      "completion_length": 67.8203125,
      "epoch": 22.25,
      "grad_norm": 1.5293912887573242,
      "kl": 0.083251953125,
      "learning_rate": 9.44375e-07,
      "loss": 0.0033,
      "reward": 1.9049308896064758,
      "reward_std": 0.10063770413398743,
      "rewards/accuracy_reward": 0.9049308598041534,
      "rewards/format_reward": 1.0,
      "step": 89
    },
    {
      "completion_length": 73.7890625,
      "epoch": 22.5,
      "grad_norm": 1.6518833637237549,
      "kl": 0.07080078125,
      "learning_rate": 9.4375e-07,
      "loss": 0.0028,
      "reward": 1.9123117327690125,
      "reward_std": 0.10563771054148674,
      "rewards/accuracy_reward": 0.9201242327690125,
      "rewards/format_reward": 0.9921875,
      "step": 90
    },
    {
      "completion_length": 76.1484375,
      "epoch": 22.75,
      "grad_norm": 1.4721721410751343,
      "kl": 0.0540771484375,
      "learning_rate": 9.43125e-07,
      "loss": 0.0022,
      "reward": 1.7826260328292847,
      "reward_std": 0.09974323213100433,
      "rewards/accuracy_reward": 0.7904385328292847,
      "rewards/format_reward": 0.9921875,
      "step": 91
    },
    {
      "completion_length": 76.28571701049805,
      "epoch": 23.0,
      "grad_norm": 1.7413424253463745,
      "kl": 0.0732421875,
      "learning_rate": 9.425e-07,
      "loss": 0.003,
      "reward": 1.7768830060958862,
      "reward_std": 0.27347957249730825,
      "rewards/accuracy_reward": 0.7768829166889191,
      "rewards/format_reward": 1.0,
      "step": 92
    },
    {
      "completion_length": 74.9296875,
      "epoch": 23.25,
      "grad_norm": 1.5143275260925293,
      "kl": 0.08154296875,
      "learning_rate": 9.41875e-07,
      "loss": 0.0033,
      "reward": 1.8272383213043213,
      "reward_std": 0.14210523292422295,
      "rewards/accuracy_reward": 0.8350508213043213,
      "rewards/format_reward": 0.9921875,
      "step": 93
    },
    {
      "completion_length": 69.9453125,
      "epoch": 23.5,
      "grad_norm": 1.544258713722229,
      "kl": 0.08349609375,
      "learning_rate": 9.4125e-07,
      "loss": 0.0033,
      "reward": 1.937786877155304,
      "reward_std": 0.07279435358941555,
      "rewards/accuracy_reward": 0.9377869367599487,
      "rewards/format_reward": 1.0,
      "step": 94
    },
    {
      "completion_length": 74.7109375,
      "epoch": 23.75,
      "grad_norm": 1.8090089559555054,
      "kl": 0.0645751953125,
      "learning_rate": 9.40625e-07,
      "loss": 0.0026,
      "reward": 1.8037108778953552,
      "reward_std": 0.16137579828500748,
      "rewards/accuracy_reward": 0.8037109076976776,
      "rewards/format_reward": 1.0,
      "step": 95
    },
    {
      "completion_length": 58.07143020629883,
      "epoch": 24.0,
      "grad_norm": 1.5920876264572144,
      "kl": 0.0655517578125,
      "learning_rate": 9.399999999999999e-07,
      "loss": 0.0028,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 96
    },
    {
      "completion_length": 71.515625,
      "epoch": 24.25,
      "grad_norm": 1.088055968284607,
      "kl": 0.0806884765625,
      "learning_rate": 9.393749999999999e-07,
      "loss": 0.0032,
      "reward": 1.924683690071106,
      "reward_std": 0.054580602794885635,
      "rewards/accuracy_reward": 0.924683690071106,
      "rewards/format_reward": 1.0,
      "step": 97
    },
    {
      "completion_length": 67.359375,
      "epoch": 24.5,
      "grad_norm": 1.886084794998169,
      "kl": 0.062255859375,
      "learning_rate": 9.387499999999999e-07,
      "loss": 0.0025,
      "reward": 1.8458982706069946,
      "reward_std": 0.08389683440327644,
      "rewards/accuracy_reward": 0.8458982408046722,
      "rewards/format_reward": 1.0,
      "step": 98
    },
    {
      "completion_length": 67.765625,
      "epoch": 24.75,
      "grad_norm": 1.9300857782363892,
      "kl": 0.086181640625,
      "learning_rate": 9.381249999999999e-07,
      "loss": 0.0035,
      "reward": 1.8570204973220825,
      "reward_std": 0.1493590921163559,
      "rewards/accuracy_reward": 0.8570204675197601,
      "rewards/format_reward": 1.0,
      "step": 99
    },
    {
      "completion_length": 69.14286041259766,
      "epoch": 25.0,
      "grad_norm": 1.8485654592514038,
      "kl": 0.08447265625,
      "learning_rate": 9.374999999999999e-07,
      "loss": 0.0035,
      "reward": 1.9482696056365967,
      "reward_std": 0.042133186012506485,
      "rewards/accuracy_reward": 0.9482696056365967,
      "rewards/format_reward": 1.0,
      "step": 100
    },
    {
      "completion_length": 69.8046875,
      "epoch": 25.25,
      "grad_norm": 1.7303087711334229,
      "kl": 0.0626220703125,
      "learning_rate": 9.368749999999999e-07,
      "loss": 0.0025,
      "reward": 1.772914171218872,
      "reward_std": 0.13923951238393784,
      "rewards/accuracy_reward": 0.7807266712188721,
      "rewards/format_reward": 0.9921875,
      "step": 101
    },
    {
      "completion_length": 64.5234375,
      "epoch": 25.5,
      "grad_norm": 1.4180161952972412,
      "kl": 0.092041015625,
      "learning_rate": 9.3625e-07,
      "loss": 0.0037,
      "reward": 1.9157771468162537,
      "reward_std": 0.09159575775265694,
      "rewards/accuracy_reward": 0.9157771468162537,
      "rewards/format_reward": 1.0,
      "step": 102
    },
    {
      "completion_length": 66.8515625,
      "epoch": 25.75,
      "grad_norm": 1.7584847211837769,
      "kl": 0.098388671875,
      "learning_rate": 9.35625e-07,
      "loss": 0.0039,
      "reward": 1.9192783832550049,
      "reward_std": 0.05962797999382019,
      "rewards/accuracy_reward": 0.9192784130573273,
      "rewards/format_reward": 1.0,
      "step": 103
    },
    {
      "completion_length": 80.28571701049805,
      "epoch": 26.0,
      "grad_norm": 1.4904193878173828,
      "kl": 0.079345703125,
      "learning_rate": 9.35e-07,
      "loss": 0.003,
      "reward": 1.8510922193527222,
      "reward_std": 0.12523864209651947,
      "rewards/accuracy_reward": 0.8510921597480774,
      "rewards/format_reward": 1.0,
      "step": 104
    },
    {
      "completion_length": 61.109375,
      "epoch": 26.25,
      "grad_norm": 1.324255108833313,
      "kl": 0.0677490234375,
      "learning_rate": 9.34375e-07,
      "loss": 0.0027,
      "reward": 1.9567571878433228,
      "reward_std": 0.0685323141515255,
      "rewards/accuracy_reward": 0.9567572176456451,
      "rewards/format_reward": 1.0,
      "step": 105
    },
    {
      "completion_length": 71.4296875,
      "epoch": 26.5,
      "grad_norm": 2.364198684692383,
      "kl": 0.07373046875,
      "learning_rate": 9.3375e-07,
      "loss": 0.0029,
      "reward": 1.808899700641632,
      "reward_std": 0.1115667074918747,
      "rewards/accuracy_reward": 0.8167122006416321,
      "rewards/format_reward": 0.9921875,
      "step": 106
    },
    {
      "completion_length": 73.1171875,
      "epoch": 26.75,
      "grad_norm": 2.9395134449005127,
      "kl": 0.0947265625,
      "learning_rate": 9.33125e-07,
      "loss": 0.0038,
      "reward": 1.7833083868026733,
      "reward_std": 0.13094724714756012,
      "rewards/accuracy_reward": 0.7911209166049957,
      "rewards/format_reward": 0.9921875,
      "step": 107
    },
    {
      "completion_length": 58.42857551574707,
      "epoch": 27.0,
      "grad_norm": 1.540488839149475,
      "kl": 0.093994140625,
      "learning_rate": 9.325e-07,
      "loss": 0.0036,
      "reward": 1.9905372262001038,
      "reward_std": 0.008901512250304222,
      "rewards/accuracy_reward": 0.9905371963977814,
      "rewards/format_reward": 1.0,
      "step": 108
    },
    {
      "completion_length": 74.328125,
      "epoch": 27.25,
      "grad_norm": 1.7257088422775269,
      "kl": 0.068115234375,
      "learning_rate": 9.31875e-07,
      "loss": 0.0027,
      "reward": 1.8329947590827942,
      "reward_std": 0.1075306311249733,
      "rewards/accuracy_reward": 0.8329947590827942,
      "rewards/format_reward": 1.0,
      "step": 109
    },
    {
      "completion_length": 65.4375,
      "epoch": 27.5,
      "grad_norm": 1.3653507232666016,
      "kl": 0.072265625,
      "learning_rate": 9.3125e-07,
      "loss": 0.0029,
      "reward": 1.9556488394737244,
      "reward_std": 0.022309845313429832,
      "rewards/accuracy_reward": 0.9556488394737244,
      "rewards/format_reward": 1.0,
      "step": 110
    },
    {
      "completion_length": 66.34375,
      "epoch": 27.75,
      "grad_norm": 2.26039981842041,
      "kl": 0.07080078125,
      "learning_rate": 9.30625e-07,
      "loss": 0.0028,
      "reward": 1.8479012846946716,
      "reward_std": 0.1186746098101139,
      "rewards/accuracy_reward": 0.8479012846946716,
      "rewards/format_reward": 1.0,
      "step": 111
    },
    {
      "completion_length": 73.64286231994629,
      "epoch": 28.0,
      "grad_norm": 1.920817494392395,
      "kl": 0.089599609375,
      "learning_rate": 9.3e-07,
      "loss": 0.0036,
      "reward": 1.8746840953826904,
      "reward_std": 0.15773950517177582,
      "rewards/accuracy_reward": 0.874684065580368,
      "rewards/format_reward": 1.0,
      "step": 112
    },
    {
      "completion_length": 70.9375,
      "epoch": 28.25,
      "grad_norm": 2.101787805557251,
      "kl": 0.070068359375,
      "learning_rate": 9.293749999999999e-07,
      "loss": 0.0028,
      "reward": 1.9114864468574524,
      "reward_std": 0.08409742452204227,
      "rewards/accuracy_reward": 0.9114864468574524,
      "rewards/format_reward": 1.0,
      "step": 113
    },
    {
      "completion_length": 68.3671875,
      "epoch": 28.5,
      "grad_norm": 1.2061454057693481,
      "kl": 0.071044921875,
      "learning_rate": 9.287499999999999e-07,
      "loss": 0.0028,
      "reward": 1.8754997849464417,
      "reward_std": 0.1343640312552452,
      "rewards/accuracy_reward": 0.883312314748764,
      "rewards/format_reward": 0.9921875,
      "step": 114
    },
    {
      "completion_length": 75.0390625,
      "epoch": 28.75,
      "grad_norm": 1.5401500463485718,
      "kl": 0.102294921875,
      "learning_rate": 9.281249999999999e-07,
      "loss": 0.0041,
      "reward": 1.7917361855506897,
      "reward_std": 0.11317018419504166,
      "rewards/accuracy_reward": 0.7917361855506897,
      "rewards/format_reward": 1.0,
      "step": 115
    },
    {
      "completion_length": 73.21429061889648,
      "epoch": 29.0,
      "grad_norm": 1.6947107315063477,
      "kl": 0.06884765625,
      "learning_rate": 9.274999999999999e-07,
      "loss": 0.0027,
      "reward": 1.8370923399925232,
      "reward_std": 0.21845543384552002,
      "rewards/accuracy_reward": 0.8370923399925232,
      "rewards/format_reward": 1.0,
      "step": 116
    },
    {
      "completion_length": 75.4765625,
      "epoch": 29.25,
      "grad_norm": 1.9241127967834473,
      "kl": 0.085693359375,
      "learning_rate": 9.268749999999999e-07,
      "loss": 0.0034,
      "reward": 1.8733400702476501,
      "reward_std": 0.15294026210904121,
      "rewards/accuracy_reward": 0.8811526000499725,
      "rewards/format_reward": 0.9921875,
      "step": 117
    },
    {
      "completion_length": 73.2890625,
      "epoch": 29.5,
      "grad_norm": 1.4431734085083008,
      "kl": 0.076171875,
      "learning_rate": 9.2625e-07,
      "loss": 0.0031,
      "reward": 1.829237937927246,
      "reward_std": 0.12272904813289642,
      "rewards/accuracy_reward": 0.8370503783226013,
      "rewards/format_reward": 0.9921875,
      "step": 118
    },
    {
      "completion_length": 63.21875,
      "epoch": 29.75,
      "grad_norm": 0.8593989014625549,
      "kl": 0.092529296875,
      "learning_rate": 9.25625e-07,
      "loss": 0.0037,
      "reward": 1.89995938539505,
      "reward_std": 0.026171773206442595,
      "rewards/accuracy_reward": 0.8999594449996948,
      "rewards/format_reward": 1.0,
      "step": 119
    },
    {
      "completion_length": 61.92857360839844,
      "epoch": 30.0,
      "grad_norm": 1.6377516984939575,
      "kl": 0.094970703125,
      "learning_rate": 9.25e-07,
      "loss": 0.0037,
      "reward": 1.8513376116752625,
      "reward_std": 0.1493394821882248,
      "rewards/accuracy_reward": 0.8513375520706177,
      "rewards/format_reward": 1.0,
      "step": 120
    },
    {
      "completion_length": 76.2578125,
      "epoch": 30.25,
      "grad_norm": 1.9491173028945923,
      "kl": 0.07861328125,
      "learning_rate": 9.243749999999999e-07,
      "loss": 0.0031,
      "reward": 1.874415099620819,
      "reward_std": 0.10396317765116692,
      "rewards/accuracy_reward": 0.8744150996208191,
      "rewards/format_reward": 1.0,
      "step": 121
    },
    {
      "completion_length": 71.0703125,
      "epoch": 30.5,
      "grad_norm": 1.4187074899673462,
      "kl": 0.0557861328125,
      "learning_rate": 9.237499999999999e-07,
      "loss": 0.0022,
      "reward": 1.8533148169517517,
      "reward_std": 0.1055279728025198,
      "rewards/accuracy_reward": 0.8533147573471069,
      "rewards/format_reward": 1.0,
      "step": 122
    },
    {
      "completion_length": 68.4921875,
      "epoch": 30.75,
      "grad_norm": 1.2265938520431519,
      "kl": 0.08837890625,
      "learning_rate": 9.23125e-07,
      "loss": 0.0035,
      "reward": 1.8870754837989807,
      "reward_std": 0.05487770680338144,
      "rewards/accuracy_reward": 0.8870754837989807,
      "rewards/format_reward": 1.0,
      "step": 123
    },
    {
      "completion_length": 68.42857551574707,
      "epoch": 31.0,
      "grad_norm": 1.4902782440185547,
      "kl": 0.141357421875,
      "learning_rate": 9.225e-07,
      "loss": 0.0054,
      "reward": 1.8678642511367798,
      "reward_std": 0.08099648356437683,
      "rewards/accuracy_reward": 0.8678641617298126,
      "rewards/format_reward": 1.0,
      "step": 124
    },
    {
      "completion_length": 65.359375,
      "epoch": 31.25,
      "grad_norm": 1.6962063312530518,
      "kl": 0.115478515625,
      "learning_rate": 9.21875e-07,
      "loss": 0.0046,
      "reward": 1.9279668927192688,
      "reward_std": 0.08291410095989704,
      "rewards/accuracy_reward": 0.9279668629169464,
      "rewards/format_reward": 1.0,
      "step": 125
    },
    {
      "completion_length": 66.3671875,
      "epoch": 31.5,
      "grad_norm": 1.7165579795837402,
      "kl": 0.072265625,
      "learning_rate": 9.2125e-07,
      "loss": 0.0029,
      "reward": 1.8470426201820374,
      "reward_std": 0.11905767396092415,
      "rewards/accuracy_reward": 0.8626675605773926,
      "rewards/format_reward": 0.984375,
      "step": 126
    },
    {
      "completion_length": 67.90625,
      "epoch": 31.75,
      "grad_norm": 1.3967280387878418,
      "kl": 0.070068359375,
      "learning_rate": 9.20625e-07,
      "loss": 0.0028,
      "reward": 1.8610433340072632,
      "reward_std": 0.07865873631089926,
      "rewards/accuracy_reward": 0.868855893611908,
      "rewards/format_reward": 0.9921875,
      "step": 127
    },
    {
      "completion_length": 89.50000381469727,
      "epoch": 32.0,
      "grad_norm": 1.5761455297470093,
      "kl": 0.08544921875,
      "learning_rate": 9.2e-07,
      "loss": 0.0036,
      "reward": 1.8485036492347717,
      "reward_std": 0.1893613114953041,
      "rewards/accuracy_reward": 0.848503589630127,
      "rewards/format_reward": 1.0,
      "step": 128
    },
    {
      "completion_length": 61.921875,
      "epoch": 32.25,
      "grad_norm": 2.392521619796753,
      "kl": 0.07666015625,
      "learning_rate": 9.19375e-07,
      "loss": 0.0031,
      "reward": 1.9423640966415405,
      "reward_std": 0.07001920230686665,
      "rewards/accuracy_reward": 0.9423640370368958,
      "rewards/format_reward": 1.0,
      "step": 129
    },
    {
      "completion_length": 63.15625,
      "epoch": 32.5,
      "grad_norm": 1.7595425844192505,
      "kl": 0.067138671875,
      "learning_rate": 9.187499999999999e-07,
      "loss": 0.0027,
      "reward": 1.7660340666770935,
      "reward_std": 0.13782843947410583,
      "rewards/accuracy_reward": 0.7738466858863831,
      "rewards/format_reward": 0.9921875,
      "step": 130
    },
    {
      "completion_length": 66.59375,
      "epoch": 32.75,
      "grad_norm": 1.5350968837738037,
      "kl": 0.07666015625,
      "learning_rate": 9.181249999999999e-07,
      "loss": 0.0031,
      "reward": 1.9150999784469604,
      "reward_std": 0.08709576074033976,
      "rewards/accuracy_reward": 0.9150999784469604,
      "rewards/format_reward": 1.0,
      "step": 131
    },
    {
      "completion_length": 76.21428680419922,
      "epoch": 33.0,
      "grad_norm": 1.6754910945892334,
      "kl": 0.10986328125,
      "learning_rate": 9.174999999999999e-07,
      "loss": 0.0044,
      "reward": 1.748177945613861,
      "reward_std": 0.24180777370929718,
      "rewards/accuracy_reward": 0.7481780052185059,
      "rewards/format_reward": 1.0,
      "step": 132
    },
    {
      "completion_length": 60.9921875,
      "epoch": 33.25,
      "grad_norm": 1.6315778493881226,
      "kl": 0.102294921875,
      "learning_rate": 9.168749999999999e-07,
      "loss": 0.0041,
      "reward": 1.9423342943191528,
      "reward_std": 0.08625898323953152,
      "rewards/accuracy_reward": 0.94233438372612,
      "rewards/format_reward": 1.0,
      "step": 133
    },
    {
      "completion_length": 65.8515625,
      "epoch": 33.5,
      "grad_norm": 1.549924373626709,
      "kl": 0.08837890625,
      "learning_rate": 9.1625e-07,
      "loss": 0.0035,
      "reward": 1.8791784048080444,
      "reward_std": 0.07251664437353611,
      "rewards/accuracy_reward": 0.8791784048080444,
      "rewards/format_reward": 1.0,
      "step": 134
    },
    {
      "completion_length": 64.0,
      "epoch": 33.75,
      "grad_norm": 1.5533236265182495,
      "kl": 0.1025390625,
      "learning_rate": 9.15625e-07,
      "loss": 0.0041,
      "reward": 1.8645389080047607,
      "reward_std": 0.06934745609760284,
      "rewards/accuracy_reward": 0.8645389974117279,
      "rewards/format_reward": 1.0,
      "step": 135
    },
    {
      "completion_length": 53.50000190734863,
      "epoch": 34.0,
      "grad_norm": 1.9051095247268677,
      "kl": 0.08349609375,
      "learning_rate": 9.15e-07,
      "loss": 0.0035,
      "reward": 1.9944872856140137,
      "reward_std": 0.002273906720802188,
      "rewards/accuracy_reward": 0.9944872260093689,
      "rewards/format_reward": 1.0,
      "step": 136
    },
    {
      "completion_length": 64.984375,
      "epoch": 34.25,
      "grad_norm": 1.688677191734314,
      "kl": 0.09130859375,
      "learning_rate": 9.14375e-07,
      "loss": 0.0037,
      "reward": 1.877675175666809,
      "reward_std": 0.07563214749097824,
      "rewards/accuracy_reward": 0.8776752054691315,
      "rewards/format_reward": 1.0,
      "step": 137
    },
    {
      "completion_length": 69.9375,
      "epoch": 34.5,
      "grad_norm": 2.37134051322937,
      "kl": 0.08447265625,
      "learning_rate": 9.137499999999999e-07,
      "loss": 0.0034,
      "reward": 1.8840071558952332,
      "reward_std": 0.0991634838283062,
      "rewards/accuracy_reward": 0.8840071260929108,
      "rewards/format_reward": 1.0,
      "step": 138
    },
    {
      "completion_length": 62.5078125,
      "epoch": 34.75,
      "grad_norm": 3.3183658123016357,
      "kl": 0.091064453125,
      "learning_rate": 9.131249999999999e-07,
      "loss": 0.0036,
      "reward": 1.9298002123832703,
      "reward_std": 0.04489972349256277,
      "rewards/accuracy_reward": 0.929800271987915,
      "rewards/format_reward": 1.0,
      "step": 139
    },
    {
      "completion_length": 43.64285850524902,
      "epoch": 35.0,
      "grad_norm": 1.5096288919448853,
      "kl": 0.11328125,
      "learning_rate": 9.124999999999999e-07,
      "loss": 0.0046,
      "reward": 2.000000238418579,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0000001788139343,
      "rewards/format_reward": 1.0,
      "step": 140
    },
    {
      "completion_length": 66.15625,
      "epoch": 35.25,
      "grad_norm": 1.5723927021026611,
      "kl": 0.125244140625,
      "learning_rate": 9.11875e-07,
      "loss": 0.005,
      "reward": 1.883579969406128,
      "reward_std": 0.11640417203307152,
      "rewards/accuracy_reward": 0.8835799992084503,
      "rewards/format_reward": 1.0,
      "step": 141
    },
    {
      "completion_length": 63.1875,
      "epoch": 35.5,
      "grad_norm": 1.9653661251068115,
      "kl": 0.112060546875,
      "learning_rate": 9.1125e-07,
      "loss": 0.0045,
      "reward": 1.9033833146095276,
      "reward_std": 0.1032666489481926,
      "rewards/accuracy_reward": 0.9033832252025604,
      "rewards/format_reward": 1.0,
      "step": 142
    },
    {
      "completion_length": 68.21875,
      "epoch": 35.75,
      "grad_norm": 1.6285479068756104,
      "kl": 0.098876953125,
      "learning_rate": 9.10625e-07,
      "loss": 0.004,
      "reward": 1.9241021275520325,
      "reward_std": 0.048172490671277046,
      "rewards/accuracy_reward": 0.9241021573543549,
      "rewards/format_reward": 1.0,
      "step": 143
    },
    {
      "completion_length": 66.00000381469727,
      "epoch": 36.0,
      "grad_norm": 1.6228117942810059,
      "kl": 0.086669921875,
      "learning_rate": 9.1e-07,
      "loss": 0.0036,
      "reward": 1.9067211747169495,
      "reward_std": 0.1559794396162033,
      "rewards/accuracy_reward": 0.9067211747169495,
      "rewards/format_reward": 1.0,
      "step": 144
    },
    {
      "completion_length": 66.4140625,
      "epoch": 36.25,
      "grad_norm": 1.8398008346557617,
      "kl": 0.11865234375,
      "learning_rate": 9.09375e-07,
      "loss": 0.0047,
      "reward": 1.8845882415771484,
      "reward_std": 0.09509449079632759,
      "rewards/accuracy_reward": 0.8845882713794708,
      "rewards/format_reward": 1.0,
      "step": 145
    },
    {
      "completion_length": 62.125,
      "epoch": 36.5,
      "grad_norm": 1.786875605583191,
      "kl": 0.133544921875,
      "learning_rate": 9.087499999999999e-07,
      "loss": 0.0053,
      "reward": 1.9288102984428406,
      "reward_std": 0.10526704788208008,
      "rewards/accuracy_reward": 0.928810328245163,
      "rewards/format_reward": 1.0,
      "step": 146
    },
    {
      "completion_length": 66.4609375,
      "epoch": 36.75,
      "grad_norm": 1.0654469728469849,
      "kl": 0.085693359375,
      "learning_rate": 9.081249999999999e-07,
      "loss": 0.0034,
      "reward": 1.9138463139533997,
      "reward_std": 0.05325407162308693,
      "rewards/accuracy_reward": 0.9138462841510773,
      "rewards/format_reward": 1.0,
      "step": 147
    },
    {
      "completion_length": 73.92857360839844,
      "epoch": 37.0,
      "grad_norm": 3.4973747730255127,
      "kl": 0.08349609375,
      "learning_rate": 9.074999999999999e-07,
      "loss": 0.0033,
      "reward": 1.9407565593719482,
      "reward_std": 0.02056967036332935,
      "rewards/accuracy_reward": 0.9407565593719482,
      "rewards/format_reward": 1.0,
      "step": 148
    },
    {
      "completion_length": 64.2421875,
      "epoch": 37.25,
      "grad_norm": 1.0742721557617188,
      "kl": 0.0662841796875,
      "learning_rate": 9.068749999999999e-07,
      "loss": 0.0027,
      "reward": 1.95524662733078,
      "reward_std": 0.03273457381874323,
      "rewards/accuracy_reward": 0.9552466571331024,
      "rewards/format_reward": 1.0,
      "step": 149
    },
    {
      "completion_length": 65.5234375,
      "epoch": 37.5,
      "grad_norm": 1.77584707736969,
      "kl": 0.13134765625,
      "learning_rate": 9.0625e-07,
      "loss": 0.0053,
      "reward": 1.9271560907363892,
      "reward_std": 0.09471606463193893,
      "rewards/accuracy_reward": 0.9271561205387115,
      "rewards/format_reward": 1.0,
      "step": 150
    },
    {
      "completion_length": 64.6484375,
      "epoch": 37.75,
      "grad_norm": 1.662251353263855,
      "kl": 0.10595703125,
      "learning_rate": 9.05625e-07,
      "loss": 0.0042,
      "reward": 1.886656403541565,
      "reward_std": 0.061278367415070534,
      "rewards/accuracy_reward": 0.8866565227508545,
      "rewards/format_reward": 1.0,
      "step": 151
    },
    {
      "completion_length": 62.14285850524902,
      "epoch": 38.0,
      "grad_norm": 1.0349845886230469,
      "kl": 0.084716796875,
      "learning_rate": 9.05e-07,
      "loss": 0.0033,
      "reward": 1.9759380221366882,
      "reward_std": 0.006050224881619215,
      "rewards/accuracy_reward": 0.9759379625320435,
      "rewards/format_reward": 1.0,
      "step": 152
    },
    {
      "completion_length": 68.9453125,
      "epoch": 38.25,
      "grad_norm": 1.4474000930786133,
      "kl": 0.081298828125,
      "learning_rate": 9.04375e-07,
      "loss": 0.0033,
      "reward": 1.899282455444336,
      "reward_std": 0.07941721752285957,
      "rewards/accuracy_reward": 0.8992824554443359,
      "rewards/format_reward": 1.0,
      "step": 153
    },
    {
      "completion_length": 67.90625,
      "epoch": 38.5,
      "grad_norm": 1.698777437210083,
      "kl": 0.091552734375,
      "learning_rate": 9.0375e-07,
      "loss": 0.0037,
      "reward": 1.8697898387908936,
      "reward_std": 0.096246431581676,
      "rewards/accuracy_reward": 0.8697898387908936,
      "rewards/format_reward": 1.0,
      "step": 154
    },
    {
      "completion_length": 62.40625,
      "epoch": 38.75,
      "grad_norm": 2.1754443645477295,
      "kl": 0.142578125,
      "learning_rate": 9.031249999999999e-07,
      "loss": 0.0057,
      "reward": 1.938372015953064,
      "reward_std": 0.05481863580644131,
      "rewards/accuracy_reward": 0.9383720755577087,
      "rewards/format_reward": 1.0,
      "step": 155
    },
    {
      "completion_length": 61.42857551574707,
      "epoch": 39.0,
      "grad_norm": 1.1363853216171265,
      "kl": 0.10986328125,
      "learning_rate": 9.024999999999999e-07,
      "loss": 0.0041,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 0.9999999403953552,
      "rewards/format_reward": 1.0,
      "step": 156
    },
    {
      "completion_length": 63.625,
      "epoch": 39.25,
      "grad_norm": 2.1956725120544434,
      "kl": 0.09814453125,
      "learning_rate": 9.018749999999999e-07,
      "loss": 0.0039,
      "reward": 1.9699036478996277,
      "reward_std": 0.028281668201088905,
      "rewards/accuracy_reward": 0.9699036478996277,
      "rewards/format_reward": 1.0,
      "step": 157
    },
    {
      "completion_length": 68.1953125,
      "epoch": 39.5,
      "grad_norm": 12.062308311462402,
      "kl": 0.089111328125,
      "learning_rate": 9.0125e-07,
      "loss": 0.0036,
      "reward": 1.9097211360931396,
      "reward_std": 0.028848751448094845,
      "rewards/accuracy_reward": 0.9097210764884949,
      "rewards/format_reward": 1.0,
      "step": 158
    },
    {
      "completion_length": 60.328125,
      "epoch": 39.75,
      "grad_norm": 1.7780170440673828,
      "kl": 0.117431640625,
      "learning_rate": 9.00625e-07,
      "loss": 0.0047,
      "reward": 1.9212573766708374,
      "reward_std": 0.05963377561420202,
      "rewards/accuracy_reward": 0.9212574064731598,
      "rewards/format_reward": 1.0,
      "step": 159
    },
    {
      "completion_length": 74.35714721679688,
      "epoch": 40.0,
      "grad_norm": 1.2525172233581543,
      "kl": 0.092041015625,
      "learning_rate": 9e-07,
      "loss": 0.0034,
      "reward": 1.60818749666214,
      "reward_std": 0.27048664540052414,
      "rewards/accuracy_reward": 0.6081875115633011,
      "rewards/format_reward": 1.0,
      "step": 160
    },
    {
      "completion_length": 64.375,
      "epoch": 40.25,
      "grad_norm": 1.7184271812438965,
      "kl": 0.124267578125,
      "learning_rate": 8.99375e-07,
      "loss": 0.005,
      "reward": 1.8835479021072388,
      "reward_std": 0.12309728935360909,
      "rewards/accuracy_reward": 0.8835478723049164,
      "rewards/format_reward": 1.0,
      "step": 161
    },
    {
      "completion_length": 64.40625,
      "epoch": 40.5,
      "grad_norm": 1.8725802898406982,
      "kl": 0.093017578125,
      "learning_rate": 8.9875e-07,
      "loss": 0.0037,
      "reward": 1.8645429015159607,
      "reward_std": 0.11185415461659431,
      "rewards/accuracy_reward": 0.8645428419113159,
      "rewards/format_reward": 1.0,
      "step": 162
    },
    {
      "completion_length": 63.2890625,
      "epoch": 40.75,
      "grad_norm": 1.6588366031646729,
      "kl": 0.103759765625,
      "learning_rate": 8.981249999999999e-07,
      "loss": 0.0042,
      "reward": 1.9586772918701172,
      "reward_std": 0.0604820279404521,
      "rewards/accuracy_reward": 0.9586772918701172,
      "rewards/format_reward": 1.0,
      "step": 163
    },
    {
      "completion_length": 57.07143020629883,
      "epoch": 41.0,
      "grad_norm": 1.847510576248169,
      "kl": 0.0975341796875,
      "learning_rate": 8.974999999999999e-07,
      "loss": 0.004,
      "reward": 1.996271789073944,
      "reward_std": 0.009226946160197258,
      "rewards/accuracy_reward": 0.9962717592716217,
      "rewards/format_reward": 1.0,
      "step": 164
    },
    {
      "completion_length": 65.078125,
      "epoch": 41.25,
      "grad_norm": 38.17945861816406,
      "kl": 0.104736328125,
      "learning_rate": 8.96875e-07,
      "loss": 0.0042,
      "reward": 1.972551941871643,
      "reward_std": 0.029624830232933164,
      "rewards/accuracy_reward": 0.9725519418716431,
      "rewards/format_reward": 1.0,
      "step": 165
    },
    {
      "completion_length": 61.625,
      "epoch": 41.5,
      "grad_norm": 1.6226741075515747,
      "kl": 0.10888671875,
      "learning_rate": 8.9625e-07,
      "loss": 0.0044,
      "reward": 1.9604188203811646,
      "reward_std": 0.032178135588765144,
      "rewards/accuracy_reward": 0.9604187309741974,
      "rewards/format_reward": 1.0,
      "step": 166
    },
    {
      "completion_length": 62.6953125,
      "epoch": 41.75,
      "grad_norm": 1.6520531177520752,
      "kl": 0.10791015625,
      "learning_rate": 8.95625e-07,
      "loss": 0.0043,
      "reward": 1.927883267402649,
      "reward_std": 0.13132158294320107,
      "rewards/accuracy_reward": 0.9278833270072937,
      "rewards/format_reward": 1.0,
      "step": 167
    },
    {
      "completion_length": 59.500003814697266,
      "epoch": 42.0,
      "grad_norm": 3.64481520652771,
      "kl": 0.224365234375,
      "learning_rate": 8.95e-07,
      "loss": 0.0083,
      "reward": 1.990772783756256,
      "reward_std": 0.014955918304622173,
      "rewards/accuracy_reward": 0.9907727539539337,
      "rewards/format_reward": 1.0,
      "step": 168
    },
    {
      "completion_length": 64.328125,
      "epoch": 42.25,
      "grad_norm": 1.8193657398223877,
      "kl": 0.139404296875,
      "learning_rate": 8.94375e-07,
      "loss": 0.0056,
      "reward": 1.8847713470458984,
      "reward_std": 0.07002338580787182,
      "rewards/accuracy_reward": 0.8847712576389313,
      "rewards/format_reward": 1.0,
      "step": 169
    },
    {
      "completion_length": 64.3203125,
      "epoch": 42.5,
      "grad_norm": 1.530835509300232,
      "kl": 0.081298828125,
      "learning_rate": 8.9375e-07,
      "loss": 0.0032,
      "reward": 1.932590126991272,
      "reward_std": 0.08853724412620068,
      "rewards/accuracy_reward": 0.932590126991272,
      "rewards/format_reward": 1.0,
      "step": 170
    },
    {
      "completion_length": 64.421875,
      "epoch": 42.75,
      "grad_norm": 1.8089220523834229,
      "kl": 0.10009765625,
      "learning_rate": 8.931249999999999e-07,
      "loss": 0.004,
      "reward": 1.9313350915908813,
      "reward_std": 0.10991925746202469,
      "rewards/accuracy_reward": 0.9313351213932037,
      "rewards/format_reward": 1.0,
      "step": 171
    },
    {
      "completion_length": 57.78571891784668,
      "epoch": 43.0,
      "grad_norm": 1.598233938217163,
      "kl": 0.1025390625,
      "learning_rate": 8.924999999999999e-07,
      "loss": 0.004,
      "reward": 1.9523810148239136,
      "reward_std": 0.117851123213768,
      "rewards/accuracy_reward": 0.9523809850215912,
      "rewards/format_reward": 1.0,
      "step": 172
    },
    {
      "completion_length": 71.3046875,
      "epoch": 43.25,
      "grad_norm": 1.3065526485443115,
      "kl": 0.10498046875,
      "learning_rate": 8.918749999999999e-07,
      "loss": 0.0042,
      "reward": 1.9195694327354431,
      "reward_std": 0.09789480268955231,
      "rewards/accuracy_reward": 0.9195694029331207,
      "rewards/format_reward": 1.0,
      "step": 173
    },
    {
      "completion_length": 65.5234375,
      "epoch": 43.5,
      "grad_norm": 2.1177382469177246,
      "kl": 0.077880859375,
      "learning_rate": 8.912499999999999e-07,
      "loss": 0.0031,
      "reward": 1.9476656317710876,
      "reward_std": 0.1074841171503067,
      "rewards/accuracy_reward": 0.9476656317710876,
      "rewards/format_reward": 1.0,
      "step": 174
    },
    {
      "completion_length": 65.203125,
      "epoch": 43.75,
      "grad_norm": 1.707092523574829,
      "kl": 0.092529296875,
      "learning_rate": 8.906249999999999e-07,
      "loss": 0.0037,
      "reward": 1.957956075668335,
      "reward_std": 0.08192763105034828,
      "rewards/accuracy_reward": 0.9579560160636902,
      "rewards/format_reward": 1.0,
      "step": 175
    },
    {
      "completion_length": 63.92857360839844,
      "epoch": 44.0,
      "grad_norm": 2.0081725120544434,
      "kl": 0.112060546875,
      "learning_rate": 8.9e-07,
      "loss": 0.0042,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 176
    },
    {
      "completion_length": 70.5234375,
      "epoch": 44.25,
      "grad_norm": 1.1276594400405884,
      "kl": 0.078125,
      "learning_rate": 8.89375e-07,
      "loss": 0.0031,
      "reward": 1.9139772653579712,
      "reward_std": 0.04851855710148811,
      "rewards/accuracy_reward": 0.9139772355556488,
      "rewards/format_reward": 1.0,
      "step": 177
    },
    {
      "completion_length": 66.9609375,
      "epoch": 44.5,
      "grad_norm": 1.5237661600112915,
      "kl": 0.087646484375,
      "learning_rate": 8.8875e-07,
      "loss": 0.0035,
      "reward": 1.9119119048118591,
      "reward_std": 0.11337531358003616,
      "rewards/accuracy_reward": 0.9119119048118591,
      "rewards/format_reward": 1.0,
      "step": 178
    },
    {
      "completion_length": 67.40625,
      "epoch": 44.75,
      "grad_norm": 1.9239343404769897,
      "kl": 0.118408203125,
      "learning_rate": 8.88125e-07,
      "loss": 0.0047,
      "reward": 1.950349748134613,
      "reward_std": 0.08595369663089514,
      "rewards/accuracy_reward": 0.950349748134613,
      "rewards/format_reward": 1.0,
      "step": 179
    },
    {
      "completion_length": 65.92857360839844,
      "epoch": 45.0,
      "grad_norm": 1.3003308773040771,
      "kl": 0.089111328125,
      "learning_rate": 8.874999999999999e-07,
      "loss": 0.0033,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 180
    },
    {
      "completion_length": 62.4296875,
      "epoch": 45.25,
      "grad_norm": 1.5781491994857788,
      "kl": 0.1025390625,
      "learning_rate": 8.86875e-07,
      "loss": 0.0041,
      "reward": 1.9496909379959106,
      "reward_std": 0.09814758412539959,
      "rewards/accuracy_reward": 0.957503467798233,
      "rewards/format_reward": 0.9921875,
      "step": 181
    },
    {
      "completion_length": 68.4453125,
      "epoch": 45.5,
      "grad_norm": 1.3683621883392334,
      "kl": 0.066162109375,
      "learning_rate": 8.8625e-07,
      "loss": 0.0027,
      "reward": 1.9691051840782166,
      "reward_std": 0.05337556218728423,
      "rewards/accuracy_reward": 0.9691051244735718,
      "rewards/format_reward": 1.0,
      "step": 182
    },
    {
      "completion_length": 68.578125,
      "epoch": 45.75,
      "grad_norm": 2.0152835845947266,
      "kl": 0.1162109375,
      "learning_rate": 8.85625e-07,
      "loss": 0.0047,
      "reward": 1.8733105659484863,
      "reward_std": 0.10198326967656612,
      "rewards/accuracy_reward": 0.8733105063438416,
      "rewards/format_reward": 1.0,
      "step": 183
    },
    {
      "completion_length": 67.00000190734863,
      "epoch": 46.0,
      "grad_norm": 1.3855398893356323,
      "kl": 0.091796875,
      "learning_rate": 8.85e-07,
      "loss": 0.0041,
      "reward": 2.0,
      "reward_std": 0.0,
      "rewards/accuracy_reward": 1.0,
      "rewards/format_reward": 1.0,
      "step": 184
    },
    {
      "completion_length": 64.0859375,
      "epoch": 46.25,
      "grad_norm": 1.4427531957626343,
      "kl": 0.12060546875,
      "learning_rate": 8.84375e-07,
      "loss": 0.0048,
      "reward": 1.9910829067230225,
      "reward_std": 0.013608792796730995,
      "rewards/accuracy_reward": 0.9910828173160553,
      "rewards/format_reward": 1.0,
      "step": 185
    },
    {
      "completion_length": 68.8984375,
      "epoch": 46.5,
      "grad_norm": 1.636596918106079,
      "kl": 0.11083984375,
      "learning_rate": 8.8375e-07,
      "loss": 0.0044,
      "reward": 1.8811240196228027,
      "reward_std": 0.08701632916927338,
      "rewards/accuracy_reward": 0.8811240196228027,
      "rewards/format_reward": 1.0,
      "step": 186
    },
    {
      "completion_length": 62.34375,
      "epoch": 46.75,
      "grad_norm": 1.5652858018875122,
      "kl": 0.095947265625,
      "learning_rate": 8.83125e-07,
      "loss": 0.0038,
      "reward": 1.9629307389259338,
      "reward_std": 0.04571997746825218,
      "rewards/accuracy_reward": 0.9629306495189667,
      "rewards/format_reward": 1.0,
      "step": 187
    },
    {
      "completion_length": 52.85714530944824,
      "epoch": 47.0,
      "grad_norm": 1.4924471378326416,
      "kl": 0.0849609375,
      "learning_rate": 8.824999999999999e-07,
      "loss": 0.0035,
      "reward": 1.8819976449012756,
      "reward_std": 0.043679721653461456,
      "rewards/accuracy_reward": 0.8819977045059204,
      "rewards/format_reward": 1.0,
      "step": 188
    },
    {
      "completion_length": 60.375,
      "epoch": 47.25,
      "grad_norm": 1.2424156665802002,
      "kl": 0.11572265625,
      "learning_rate": 8.818749999999999e-07,
      "loss": 0.0046,
      "reward": 1.9622814059257507,
      "reward_std": 0.03379644639790058,
      "rewards/accuracy_reward": 0.9622813165187836,
      "rewards/format_reward": 1.0,
      "step": 189
    },
    {
      "completion_length": 64.8125,
      "epoch": 47.5,
      "grad_norm": 1.445113182067871,
      "kl": 0.107421875,
      "learning_rate": 8.812499999999999e-07,
      "loss": 0.0043,
      "reward": 1.969697892665863,
      "reward_std": 0.025794532150030136,
      "rewards/accuracy_reward": 0.9696978330612183,
      "rewards/format_reward": 1.0,
      "step": 190
    },
    {
      "completion_length": 62.25,
      "epoch": 47.75,
      "grad_norm": 1.8363816738128662,
      "kl": 0.161865234375,
      "learning_rate": 8.806249999999999e-07,
      "loss": 0.0065,
      "reward": 1.900344431400299,
      "reward_std": 0.05351579561829567,
      "rewards/accuracy_reward": 0.9003444910049438,
      "rewards/format_reward": 1.0,
      "step": 191
    },
    {
      "completion_length": 65.71428680419922,
      "epoch": 48.0,
      "grad_norm": 1.4372133016586304,
      "kl": 0.103271484375,
      "learning_rate": 8.799999999999999e-07,
      "loss": 0.0038,
      "reward": 1.9987189769744873,
      "reward_std": 0.0020755312871187925,
      "rewards/accuracy_reward": 0.9987190961837769,
      "rewards/format_reward": 1.0,
      "step": 192
    },
    {
      "completion_length": 60.0859375,
      "epoch": 48.25,
      "grad_norm": 1.662278175354004,
      "kl": 0.087646484375,
      "learning_rate": 8.793749999999999e-07,
      "loss": 0.0035,
      "reward": 1.9600813388824463,
      "reward_std": 0.07481173612177372,
      "rewards/accuracy_reward": 0.9600813686847687,
      "rewards/format_reward": 1.0,
      "step": 193
    },
    {
      "completion_length": 60.2890625,
      "epoch": 48.5,
      "grad_norm": 1.3797621726989746,
      "kl": 0.10595703125,
      "learning_rate": 8.7875e-07,
      "loss": 0.0042,
      "reward": 1.9719964265823364,
      "reward_std": 0.02817021100781858,
      "rewards/accuracy_reward": 0.9719964861869812,
      "rewards/format_reward": 1.0,
      "step": 194
    },
    {
      "completion_length": 60.7578125,
      "epoch": 48.75,
      "grad_norm": 1.4646984338760376,
      "kl": 0.13037109375,
      "learning_rate": 8.78125e-07,
      "loss": 0.0052,
      "reward": 1.9132311940193176,
      "reward_std": 0.04202069714665413,
      "rewards/accuracy_reward": 0.9132311940193176,
      "rewards/format_reward": 1.0,
      "step": 195
    },
    {
      "completion_length": 64.21428871154785,
      "epoch": 49.0,
      "grad_norm": 1.513399600982666,
      "kl": 0.1103515625,
      "learning_rate": 8.774999999999999e-07,
      "loss": 0.0041,
      "reward": 1.9454046487808228,
      "reward_std": 0.055479995906353,
      "rewards/accuracy_reward": 0.9454046189785004,
      "rewards/format_reward": 1.0,
      "step": 196
    },
    {
      "completion_length": 61.5,
      "epoch": 49.25,
      "grad_norm": 1.0642904043197632,
      "kl": 0.094482421875,
      "learning_rate": 8.76875e-07,
      "loss": 0.0038,
      "reward": 1.9798645377159119,
      "reward_std": 0.031060860259458423,
      "rewards/accuracy_reward": 0.9798645973205566,
      "rewards/format_reward": 1.0,
      "step": 197
    },
    {
      "completion_length": 61.5,
      "epoch": 49.5,
      "grad_norm": 2.066206455230713,
      "kl": 0.1689453125,
      "learning_rate": 8.7625e-07,
      "loss": 0.0067,
      "reward": 1.917717456817627,
      "reward_std": 0.027819208800792694,
      "rewards/accuracy_reward": 0.9177174866199493,
      "rewards/format_reward": 1.0,
      "step": 198
    },
    {
      "completion_length": 59.9296875,
      "epoch": 49.75,
      "grad_norm": 1.2293668985366821,
      "kl": 0.114990234375,
      "learning_rate": 8.75625e-07,
      "loss": 0.0046,
      "reward": 1.9745354652404785,
      "reward_std": 0.048739076184574515,
      "rewards/accuracy_reward": 0.9823479354381561,
      "rewards/format_reward": 0.9921875,
      "step": 199
    },
    {
      "completion_length": 75.85714721679688,
      "epoch": 50.0,
      "grad_norm": 1.5204596519470215,
      "kl": 0.095703125,
      "learning_rate": 8.75e-07,
      "loss": 0.0041,
      "reward": 1.9712833166122437,
      "reward_std": 0.02327703475020826,
      "rewards/accuracy_reward": 0.9712833166122437,
      "rewards/format_reward": 1.0,
      "step": 200
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 400,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}