wbz0505 commited on
Commit
d9c5878
·
verified ·
1 Parent(s): e5544e7

Upload 10 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,518 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</Motion Tokens>": 32613,
3
+ "<0>": 32100,
4
+ "<100>": 32200,
5
+ "<101>": 32201,
6
+ "<102>": 32202,
7
+ "<103>": 32203,
8
+ "<104>": 32204,
9
+ "<105>": 32205,
10
+ "<106>": 32206,
11
+ "<107>": 32207,
12
+ "<108>": 32208,
13
+ "<109>": 32209,
14
+ "<10>": 32110,
15
+ "<110>": 32210,
16
+ "<111>": 32211,
17
+ "<112>": 32212,
18
+ "<113>": 32213,
19
+ "<114>": 32214,
20
+ "<115>": 32215,
21
+ "<116>": 32216,
22
+ "<117>": 32217,
23
+ "<118>": 32218,
24
+ "<119>": 32219,
25
+ "<11>": 32111,
26
+ "<120>": 32220,
27
+ "<121>": 32221,
28
+ "<122>": 32222,
29
+ "<123>": 32223,
30
+ "<124>": 32224,
31
+ "<125>": 32225,
32
+ "<126>": 32226,
33
+ "<127>": 32227,
34
+ "<128>": 32228,
35
+ "<129>": 32229,
36
+ "<12>": 32112,
37
+ "<130>": 32230,
38
+ "<131>": 32231,
39
+ "<132>": 32232,
40
+ "<133>": 32233,
41
+ "<134>": 32234,
42
+ "<135>": 32235,
43
+ "<136>": 32236,
44
+ "<137>": 32237,
45
+ "<138>": 32238,
46
+ "<139>": 32239,
47
+ "<13>": 32113,
48
+ "<140>": 32240,
49
+ "<141>": 32241,
50
+ "<142>": 32242,
51
+ "<143>": 32243,
52
+ "<144>": 32244,
53
+ "<145>": 32245,
54
+ "<146>": 32246,
55
+ "<147>": 32247,
56
+ "<148>": 32248,
57
+ "<149>": 32249,
58
+ "<14>": 32114,
59
+ "<150>": 32250,
60
+ "<151>": 32251,
61
+ "<152>": 32252,
62
+ "<153>": 32253,
63
+ "<154>": 32254,
64
+ "<155>": 32255,
65
+ "<156>": 32256,
66
+ "<157>": 32257,
67
+ "<158>": 32258,
68
+ "<159>": 32259,
69
+ "<15>": 32115,
70
+ "<160>": 32260,
71
+ "<161>": 32261,
72
+ "<162>": 32262,
73
+ "<163>": 32263,
74
+ "<164>": 32264,
75
+ "<165>": 32265,
76
+ "<166>": 32266,
77
+ "<167>": 32267,
78
+ "<168>": 32268,
79
+ "<169>": 32269,
80
+ "<16>": 32116,
81
+ "<170>": 32270,
82
+ "<171>": 32271,
83
+ "<172>": 32272,
84
+ "<173>": 32273,
85
+ "<174>": 32274,
86
+ "<175>": 32275,
87
+ "<176>": 32276,
88
+ "<177>": 32277,
89
+ "<178>": 32278,
90
+ "<179>": 32279,
91
+ "<17>": 32117,
92
+ "<180>": 32280,
93
+ "<181>": 32281,
94
+ "<182>": 32282,
95
+ "<183>": 32283,
96
+ "<184>": 32284,
97
+ "<185>": 32285,
98
+ "<186>": 32286,
99
+ "<187>": 32287,
100
+ "<188>": 32288,
101
+ "<189>": 32289,
102
+ "<18>": 32118,
103
+ "<190>": 32290,
104
+ "<191>": 32291,
105
+ "<192>": 32292,
106
+ "<193>": 32293,
107
+ "<194>": 32294,
108
+ "<195>": 32295,
109
+ "<196>": 32296,
110
+ "<197>": 32297,
111
+ "<198>": 32298,
112
+ "<199>": 32299,
113
+ "<19>": 32119,
114
+ "<1>": 32101,
115
+ "<200>": 32300,
116
+ "<201>": 32301,
117
+ "<202>": 32302,
118
+ "<203>": 32303,
119
+ "<204>": 32304,
120
+ "<205>": 32305,
121
+ "<206>": 32306,
122
+ "<207>": 32307,
123
+ "<208>": 32308,
124
+ "<209>": 32309,
125
+ "<20>": 32120,
126
+ "<210>": 32310,
127
+ "<211>": 32311,
128
+ "<212>": 32312,
129
+ "<213>": 32313,
130
+ "<214>": 32314,
131
+ "<215>": 32315,
132
+ "<216>": 32316,
133
+ "<217>": 32317,
134
+ "<218>": 32318,
135
+ "<219>": 32319,
136
+ "<21>": 32121,
137
+ "<220>": 32320,
138
+ "<221>": 32321,
139
+ "<222>": 32322,
140
+ "<223>": 32323,
141
+ "<224>": 32324,
142
+ "<225>": 32325,
143
+ "<226>": 32326,
144
+ "<227>": 32327,
145
+ "<228>": 32328,
146
+ "<229>": 32329,
147
+ "<22>": 32122,
148
+ "<230>": 32330,
149
+ "<231>": 32331,
150
+ "<232>": 32332,
151
+ "<233>": 32333,
152
+ "<234>": 32334,
153
+ "<235>": 32335,
154
+ "<236>": 32336,
155
+ "<237>": 32337,
156
+ "<238>": 32338,
157
+ "<239>": 32339,
158
+ "<23>": 32123,
159
+ "<240>": 32340,
160
+ "<241>": 32341,
161
+ "<242>": 32342,
162
+ "<243>": 32343,
163
+ "<244>": 32344,
164
+ "<245>": 32345,
165
+ "<246>": 32346,
166
+ "<247>": 32347,
167
+ "<248>": 32348,
168
+ "<249>": 32349,
169
+ "<24>": 32124,
170
+ "<250>": 32350,
171
+ "<251>": 32351,
172
+ "<252>": 32352,
173
+ "<253>": 32353,
174
+ "<254>": 32354,
175
+ "<255>": 32355,
176
+ "<256>": 32356,
177
+ "<257>": 32357,
178
+ "<258>": 32358,
179
+ "<259>": 32359,
180
+ "<25>": 32125,
181
+ "<260>": 32360,
182
+ "<261>": 32361,
183
+ "<262>": 32362,
184
+ "<263>": 32363,
185
+ "<264>": 32364,
186
+ "<265>": 32365,
187
+ "<266>": 32366,
188
+ "<267>": 32367,
189
+ "<268>": 32368,
190
+ "<269>": 32369,
191
+ "<26>": 32126,
192
+ "<270>": 32370,
193
+ "<271>": 32371,
194
+ "<272>": 32372,
195
+ "<273>": 32373,
196
+ "<274>": 32374,
197
+ "<275>": 32375,
198
+ "<276>": 32376,
199
+ "<277>": 32377,
200
+ "<278>": 32378,
201
+ "<279>": 32379,
202
+ "<27>": 32127,
203
+ "<280>": 32380,
204
+ "<281>": 32381,
205
+ "<282>": 32382,
206
+ "<283>": 32383,
207
+ "<284>": 32384,
208
+ "<285>": 32385,
209
+ "<286>": 32386,
210
+ "<287>": 32387,
211
+ "<288>": 32388,
212
+ "<289>": 32389,
213
+ "<28>": 32128,
214
+ "<290>": 32390,
215
+ "<291>": 32391,
216
+ "<292>": 32392,
217
+ "<293>": 32393,
218
+ "<294>": 32394,
219
+ "<295>": 32395,
220
+ "<296>": 32396,
221
+ "<297>": 32397,
222
+ "<298>": 32398,
223
+ "<299>": 32399,
224
+ "<29>": 32129,
225
+ "<2>": 32102,
226
+ "<300>": 32400,
227
+ "<301>": 32401,
228
+ "<302>": 32402,
229
+ "<303>": 32403,
230
+ "<304>": 32404,
231
+ "<305>": 32405,
232
+ "<306>": 32406,
233
+ "<307>": 32407,
234
+ "<308>": 32408,
235
+ "<309>": 32409,
236
+ "<30>": 32130,
237
+ "<310>": 32410,
238
+ "<311>": 32411,
239
+ "<312>": 32412,
240
+ "<313>": 32413,
241
+ "<314>": 32414,
242
+ "<315>": 32415,
243
+ "<316>": 32416,
244
+ "<317>": 32417,
245
+ "<318>": 32418,
246
+ "<319>": 32419,
247
+ "<31>": 32131,
248
+ "<320>": 32420,
249
+ "<321>": 32421,
250
+ "<322>": 32422,
251
+ "<323>": 32423,
252
+ "<324>": 32424,
253
+ "<325>": 32425,
254
+ "<326>": 32426,
255
+ "<327>": 32427,
256
+ "<328>": 32428,
257
+ "<329>": 32429,
258
+ "<32>": 32132,
259
+ "<330>": 32430,
260
+ "<331>": 32431,
261
+ "<332>": 32432,
262
+ "<333>": 32433,
263
+ "<334>": 32434,
264
+ "<335>": 32435,
265
+ "<336>": 32436,
266
+ "<337>": 32437,
267
+ "<338>": 32438,
268
+ "<339>": 32439,
269
+ "<33>": 32133,
270
+ "<340>": 32440,
271
+ "<341>": 32441,
272
+ "<342>": 32442,
273
+ "<343>": 32443,
274
+ "<344>": 32444,
275
+ "<345>": 32445,
276
+ "<346>": 32446,
277
+ "<347>": 32447,
278
+ "<348>": 32448,
279
+ "<349>": 32449,
280
+ "<34>": 32134,
281
+ "<350>": 32450,
282
+ "<351>": 32451,
283
+ "<352>": 32452,
284
+ "<353>": 32453,
285
+ "<354>": 32454,
286
+ "<355>": 32455,
287
+ "<356>": 32456,
288
+ "<357>": 32457,
289
+ "<358>": 32458,
290
+ "<359>": 32459,
291
+ "<35>": 32135,
292
+ "<360>": 32460,
293
+ "<361>": 32461,
294
+ "<362>": 32462,
295
+ "<363>": 32463,
296
+ "<364>": 32464,
297
+ "<365>": 32465,
298
+ "<366>": 32466,
299
+ "<367>": 32467,
300
+ "<368>": 32468,
301
+ "<369>": 32469,
302
+ "<36>": 32136,
303
+ "<370>": 32470,
304
+ "<371>": 32471,
305
+ "<372>": 32472,
306
+ "<373>": 32473,
307
+ "<374>": 32474,
308
+ "<375>": 32475,
309
+ "<376>": 32476,
310
+ "<377>": 32477,
311
+ "<378>": 32478,
312
+ "<379>": 32479,
313
+ "<37>": 32137,
314
+ "<380>": 32480,
315
+ "<381>": 32481,
316
+ "<382>": 32482,
317
+ "<383>": 32483,
318
+ "<384>": 32484,
319
+ "<385>": 32485,
320
+ "<386>": 32486,
321
+ "<387>": 32487,
322
+ "<388>": 32488,
323
+ "<389>": 32489,
324
+ "<38>": 32138,
325
+ "<390>": 32490,
326
+ "<391>": 32491,
327
+ "<392>": 32492,
328
+ "<393>": 32493,
329
+ "<394>": 32494,
330
+ "<395>": 32495,
331
+ "<396>": 32496,
332
+ "<397>": 32497,
333
+ "<398>": 32498,
334
+ "<399>": 32499,
335
+ "<39>": 32139,
336
+ "<3>": 32103,
337
+ "<400>": 32500,
338
+ "<401>": 32501,
339
+ "<402>": 32502,
340
+ "<403>": 32503,
341
+ "<404>": 32504,
342
+ "<405>": 32505,
343
+ "<406>": 32506,
344
+ "<407>": 32507,
345
+ "<408>": 32508,
346
+ "<409>": 32509,
347
+ "<40>": 32140,
348
+ "<410>": 32510,
349
+ "<411>": 32511,
350
+ "<412>": 32512,
351
+ "<413>": 32513,
352
+ "<414>": 32514,
353
+ "<415>": 32515,
354
+ "<416>": 32516,
355
+ "<417>": 32517,
356
+ "<418>": 32518,
357
+ "<419>": 32519,
358
+ "<41>": 32141,
359
+ "<420>": 32520,
360
+ "<421>": 32521,
361
+ "<422>": 32522,
362
+ "<423>": 32523,
363
+ "<424>": 32524,
364
+ "<425>": 32525,
365
+ "<426>": 32526,
366
+ "<427>": 32527,
367
+ "<428>": 32528,
368
+ "<429>": 32529,
369
+ "<42>": 32142,
370
+ "<430>": 32530,
371
+ "<431>": 32531,
372
+ "<432>": 32532,
373
+ "<433>": 32533,
374
+ "<434>": 32534,
375
+ "<435>": 32535,
376
+ "<436>": 32536,
377
+ "<437>": 32537,
378
+ "<438>": 32538,
379
+ "<439>": 32539,
380
+ "<43>": 32143,
381
+ "<440>": 32540,
382
+ "<441>": 32541,
383
+ "<442>": 32542,
384
+ "<443>": 32543,
385
+ "<444>": 32544,
386
+ "<445>": 32545,
387
+ "<446>": 32546,
388
+ "<447>": 32547,
389
+ "<448>": 32548,
390
+ "<449>": 32549,
391
+ "<44>": 32144,
392
+ "<450>": 32550,
393
+ "<451>": 32551,
394
+ "<452>": 32552,
395
+ "<453>": 32553,
396
+ "<454>": 32554,
397
+ "<455>": 32555,
398
+ "<456>": 32556,
399
+ "<457>": 32557,
400
+ "<458>": 32558,
401
+ "<459>": 32559,
402
+ "<45>": 32145,
403
+ "<460>": 32560,
404
+ "<461>": 32561,
405
+ "<462>": 32562,
406
+ "<463>": 32563,
407
+ "<464>": 32564,
408
+ "<465>": 32565,
409
+ "<466>": 32566,
410
+ "<467>": 32567,
411
+ "<468>": 32568,
412
+ "<469>": 32569,
413
+ "<46>": 32146,
414
+ "<470>": 32570,
415
+ "<471>": 32571,
416
+ "<472>": 32572,
417
+ "<473>": 32573,
418
+ "<474>": 32574,
419
+ "<475>": 32575,
420
+ "<476>": 32576,
421
+ "<477>": 32577,
422
+ "<478>": 32578,
423
+ "<479>": 32579,
424
+ "<47>": 32147,
425
+ "<480>": 32580,
426
+ "<481>": 32581,
427
+ "<482>": 32582,
428
+ "<483>": 32583,
429
+ "<484>": 32584,
430
+ "<485>": 32585,
431
+ "<486>": 32586,
432
+ "<487>": 32587,
433
+ "<488>": 32588,
434
+ "<489>": 32589,
435
+ "<48>": 32148,
436
+ "<490>": 32590,
437
+ "<491>": 32591,
438
+ "<492>": 32592,
439
+ "<493>": 32593,
440
+ "<494>": 32594,
441
+ "<495>": 32595,
442
+ "<496>": 32596,
443
+ "<497>": 32597,
444
+ "<498>": 32598,
445
+ "<499>": 32599,
446
+ "<49>": 32149,
447
+ "<4>": 32104,
448
+ "<500>": 32600,
449
+ "<501>": 32601,
450
+ "<502>": 32602,
451
+ "<503>": 32603,
452
+ "<504>": 32604,
453
+ "<505>": 32605,
454
+ "<506>": 32606,
455
+ "<507>": 32607,
456
+ "<508>": 32608,
457
+ "<509>": 32609,
458
+ "<50>": 32150,
459
+ "<510>": 32610,
460
+ "<511>": 32611,
461
+ "<51>": 32151,
462
+ "<52>": 32152,
463
+ "<53>": 32153,
464
+ "<54>": 32154,
465
+ "<55>": 32155,
466
+ "<56>": 32156,
467
+ "<57>": 32157,
468
+ "<58>": 32158,
469
+ "<59>": 32159,
470
+ "<5>": 32105,
471
+ "<60>": 32160,
472
+ "<61>": 32161,
473
+ "<62>": 32162,
474
+ "<63>": 32163,
475
+ "<64>": 32164,
476
+ "<65>": 32165,
477
+ "<66>": 32166,
478
+ "<67>": 32167,
479
+ "<68>": 32168,
480
+ "<69>": 32169,
481
+ "<6>": 32106,
482
+ "<70>": 32170,
483
+ "<71>": 32171,
484
+ "<72>": 32172,
485
+ "<73>": 32173,
486
+ "<74>": 32174,
487
+ "<75>": 32175,
488
+ "<76>": 32176,
489
+ "<77>": 32177,
490
+ "<78>": 32178,
491
+ "<79>": 32179,
492
+ "<7>": 32107,
493
+ "<80>": 32180,
494
+ "<81>": 32181,
495
+ "<82>": 32182,
496
+ "<83>": 32183,
497
+ "<84>": 32184,
498
+ "<85>": 32185,
499
+ "<86>": 32186,
500
+ "<87>": 32187,
501
+ "<88>": 32188,
502
+ "<89>": 32189,
503
+ "<8>": 32108,
504
+ "<90>": 32190,
505
+ "<91>": 32191,
506
+ "<92>": 32192,
507
+ "<93>": 32193,
508
+ "<94>": 32194,
509
+ "<95>": 32195,
510
+ "<96>": 32196,
511
+ "<97>": 32197,
512
+ "<98>": 32198,
513
+ "<99>": 32199,
514
+ "<9>": 32109,
515
+ "<Motion Tokens>": 32612,
516
+ "<Motionless>": 32614,
517
+ "<SEP>": 32615
518
+ }
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "GSPretrained_base/checkpoint-300000/",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.28.1",
58
+ "use_cache": true,
59
+ "vocab_size": 32616
60
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.28.1"
7
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efdbd8f78657d7c4aa9cc2e222a02df360daceac9a6560125d26ce1b240778bf
3
+ size 14308
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d6c5b93324fa72d02f994789c5634c26492fde161591c5c12fb60aea61a68f2
3
+ size 1064
special_tokens_map.json ADDED
@@ -0,0 +1,107 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "eos_token": "</s>",
105
+ "pad_token": "<pad>",
106
+ "unk_token": "<unk>"
107
+ }
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
tokenizer_config.json ADDED
@@ -0,0 +1,112 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>"
103
+ ],
104
+ "clean_up_tokenization_spaces": true,
105
+ "eos_token": "</s>",
106
+ "extra_ids": 100,
107
+ "model_max_length": 1000000000000000019884624838656,
108
+ "pad_token": "<pad>",
109
+ "sp_model_kwargs": {},
110
+ "tokenizer_class": "T5Tokenizer",
111
+ "unk_token": "<unk>"
112
+ }
trainer_state.json ADDED
@@ -0,0 +1,1296 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 65.01950585175553,
5
+ "global_step": 100000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.33,
12
+ "learning_rate": 9.999383162408304e-05,
13
+ "loss": 1.7573,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.65,
18
+ "learning_rate": 9.997532801828658e-05,
19
+ "loss": 1.7515,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.98,
24
+ "learning_rate": 9.994449374809851e-05,
25
+ "loss": 1.7622,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 1.3,
30
+ "learning_rate": 9.990133642141359e-05,
31
+ "loss": 1.7166,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 1.63,
36
+ "learning_rate": 9.98458666866564e-05,
37
+ "loss": 1.7249,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 1.95,
42
+ "learning_rate": 9.977809823015401e-05,
43
+ "loss": 1.7108,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 2.28,
48
+ "learning_rate": 9.9698047772759e-05,
49
+ "loss": 1.6752,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 2.6,
54
+ "learning_rate": 9.96057350657239e-05,
55
+ "loss": 1.6668,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 2.93,
60
+ "learning_rate": 9.950118288582788e-05,
61
+ "loss": 1.6827,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 3.25,
66
+ "learning_rate": 9.938441702975689e-05,
67
+ "loss": 1.641,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 3.58,
72
+ "learning_rate": 9.92554663077387e-05,
73
+ "loss": 1.6448,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 3.9,
78
+ "learning_rate": 9.911436253643445e-05,
79
+ "loss": 1.6305,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 4.23,
84
+ "learning_rate": 9.896114053108829e-05,
85
+ "loss": 1.6062,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 4.55,
90
+ "learning_rate": 9.879583809693738e-05,
91
+ "loss": 1.5972,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 4.88,
96
+ "learning_rate": 9.861849601988383e-05,
97
+ "loss": 1.6113,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 5.2,
102
+ "learning_rate": 9.842915805643155e-05,
103
+ "loss": 1.5867,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 5.53,
108
+ "learning_rate": 9.822787092288991e-05,
109
+ "loss": 1.5802,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 5.85,
114
+ "learning_rate": 9.801468428384716e-05,
115
+ "loss": 1.5769,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 6.18,
120
+ "learning_rate": 9.778965073991651e-05,
121
+ "loss": 1.5605,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 6.5,
126
+ "learning_rate": 9.755282581475769e-05,
127
+ "loss": 1.5464,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 6.5,
132
+ "eval_loss": 1.9329419136047363,
133
+ "eval_runtime": 3.0765,
134
+ "eval_samples_per_second": 499.912,
135
+ "eval_steps_per_second": 31.529,
136
+ "step": 10000
137
+ },
138
+ {
139
+ "epoch": 6.83,
140
+ "learning_rate": 9.730426794137727e-05,
141
+ "loss": 1.527,
142
+ "step": 10500
143
+ },
144
+ {
145
+ "epoch": 7.15,
146
+ "learning_rate": 9.704403844771128e-05,
147
+ "loss": 1.5289,
148
+ "step": 11000
149
+ },
150
+ {
151
+ "epoch": 7.48,
152
+ "learning_rate": 9.677220154149337e-05,
153
+ "loss": 1.5033,
154
+ "step": 11500
155
+ },
156
+ {
157
+ "epoch": 7.8,
158
+ "learning_rate": 9.648882429441257e-05,
159
+ "loss": 1.5205,
160
+ "step": 12000
161
+ },
162
+ {
163
+ "epoch": 8.13,
164
+ "learning_rate": 9.619397662556435e-05,
165
+ "loss": 1.4979,
166
+ "step": 12500
167
+ },
168
+ {
169
+ "epoch": 8.45,
170
+ "learning_rate": 9.588773128419906e-05,
171
+ "loss": 1.4869,
172
+ "step": 13000
173
+ },
174
+ {
175
+ "epoch": 8.78,
176
+ "learning_rate": 9.557016383177227e-05,
177
+ "loss": 1.499,
178
+ "step": 13500
179
+ },
180
+ {
181
+ "epoch": 9.1,
182
+ "learning_rate": 9.524135262330098e-05,
183
+ "loss": 1.4814,
184
+ "step": 14000
185
+ },
186
+ {
187
+ "epoch": 9.43,
188
+ "learning_rate": 9.490137878803079e-05,
189
+ "loss": 1.4447,
190
+ "step": 14500
191
+ },
192
+ {
193
+ "epoch": 9.75,
194
+ "learning_rate": 9.45503262094184e-05,
195
+ "loss": 1.4686,
196
+ "step": 15000
197
+ },
198
+ {
199
+ "epoch": 10.08,
200
+ "learning_rate": 9.418828150443469e-05,
201
+ "loss": 1.4544,
202
+ "step": 15500
203
+ },
204
+ {
205
+ "epoch": 10.4,
206
+ "learning_rate": 9.381533400219318e-05,
207
+ "loss": 1.4202,
208
+ "step": 16000
209
+ },
210
+ {
211
+ "epoch": 10.73,
212
+ "learning_rate": 9.343157572190957e-05,
213
+ "loss": 1.4411,
214
+ "step": 16500
215
+ },
216
+ {
217
+ "epoch": 11.05,
218
+ "learning_rate": 9.30371013501972e-05,
219
+ "loss": 1.4283,
220
+ "step": 17000
221
+ },
222
+ {
223
+ "epoch": 11.38,
224
+ "learning_rate": 9.263200821770461e-05,
225
+ "loss": 1.4019,
226
+ "step": 17500
227
+ },
228
+ {
229
+ "epoch": 11.7,
230
+ "learning_rate": 9.221639627510076e-05,
231
+ "loss": 1.3973,
232
+ "step": 18000
233
+ },
234
+ {
235
+ "epoch": 12.03,
236
+ "learning_rate": 9.179036806841353e-05,
237
+ "loss": 1.4206,
238
+ "step": 18500
239
+ },
240
+ {
241
+ "epoch": 12.35,
242
+ "learning_rate": 9.135402871372808e-05,
243
+ "loss": 1.3621,
244
+ "step": 19000
245
+ },
246
+ {
247
+ "epoch": 12.68,
248
+ "learning_rate": 9.090748587125118e-05,
249
+ "loss": 1.3871,
250
+ "step": 19500
251
+ },
252
+ {
253
+ "epoch": 13.0,
254
+ "learning_rate": 9.045084971874738e-05,
255
+ "loss": 1.3845,
256
+ "step": 20000
257
+ },
258
+ {
259
+ "epoch": 13.0,
260
+ "eval_loss": 2.0226035118103027,
261
+ "eval_runtime": 2.951,
262
+ "eval_samples_per_second": 521.185,
263
+ "eval_steps_per_second": 32.871,
264
+ "step": 20000
265
+ },
266
+ {
267
+ "epoch": 13.33,
268
+ "learning_rate": 8.998423292435454e-05,
269
+ "loss": 1.3537,
270
+ "step": 20500
271
+ },
272
+ {
273
+ "epoch": 13.65,
274
+ "learning_rate": 8.950775061878453e-05,
275
+ "loss": 1.3523,
276
+ "step": 21000
277
+ },
278
+ {
279
+ "epoch": 13.98,
280
+ "learning_rate": 8.90215203669165e-05,
281
+ "loss": 1.3479,
282
+ "step": 21500
283
+ },
284
+ {
285
+ "epoch": 14.3,
286
+ "learning_rate": 8.852566213878947e-05,
287
+ "loss": 1.3374,
288
+ "step": 22000
289
+ },
290
+ {
291
+ "epoch": 14.63,
292
+ "learning_rate": 8.802029828000156e-05,
293
+ "loss": 1.333,
294
+ "step": 22500
295
+ },
296
+ {
297
+ "epoch": 14.95,
298
+ "learning_rate": 8.750555348152298e-05,
299
+ "loss": 1.3351,
300
+ "step": 23000
301
+ },
302
+ {
303
+ "epoch": 15.28,
304
+ "learning_rate": 8.69815547489305e-05,
305
+ "loss": 1.3062,
306
+ "step": 23500
307
+ },
308
+ {
309
+ "epoch": 15.6,
310
+ "learning_rate": 8.644843137107059e-05,
311
+ "loss": 1.3155,
312
+ "step": 24000
313
+ },
314
+ {
315
+ "epoch": 15.93,
316
+ "learning_rate": 8.590631488815944e-05,
317
+ "loss": 1.3185,
318
+ "step": 24500
319
+ },
320
+ {
321
+ "epoch": 16.25,
322
+ "learning_rate": 8.535533905932738e-05,
323
+ "loss": 1.2986,
324
+ "step": 25000
325
+ },
326
+ {
327
+ "epoch": 16.58,
328
+ "learning_rate": 8.479563982961571e-05,
329
+ "loss": 1.2964,
330
+ "step": 25500
331
+ },
332
+ {
333
+ "epoch": 16.91,
334
+ "learning_rate": 8.422735529643444e-05,
335
+ "loss": 1.2848,
336
+ "step": 26000
337
+ },
338
+ {
339
+ "epoch": 17.23,
340
+ "learning_rate": 8.365062567548867e-05,
341
+ "loss": 1.2745,
342
+ "step": 26500
343
+ },
344
+ {
345
+ "epoch": 17.56,
346
+ "learning_rate": 8.306559326618259e-05,
347
+ "loss": 1.2786,
348
+ "step": 27000
349
+ },
350
+ {
351
+ "epoch": 17.88,
352
+ "learning_rate": 8.247240241650918e-05,
353
+ "loss": 1.2679,
354
+ "step": 27500
355
+ },
356
+ {
357
+ "epoch": 18.21,
358
+ "learning_rate": 8.18711994874345e-05,
359
+ "loss": 1.2481,
360
+ "step": 28000
361
+ },
362
+ {
363
+ "epoch": 18.53,
364
+ "learning_rate": 8.126213281678528e-05,
365
+ "loss": 1.228,
366
+ "step": 28500
367
+ },
368
+ {
369
+ "epoch": 18.86,
370
+ "learning_rate": 8.064535268264883e-05,
371
+ "loss": 1.2452,
372
+ "step": 29000
373
+ },
374
+ {
375
+ "epoch": 19.18,
376
+ "learning_rate": 8.002101126629421e-05,
377
+ "loss": 1.2361,
378
+ "step": 29500
379
+ },
380
+ {
381
+ "epoch": 19.51,
382
+ "learning_rate": 7.938926261462366e-05,
383
+ "loss": 1.2254,
384
+ "step": 30000
385
+ },
386
+ {
387
+ "epoch": 19.51,
388
+ "eval_loss": 2.1522765159606934,
389
+ "eval_runtime": 2.9695,
390
+ "eval_samples_per_second": 517.928,
391
+ "eval_steps_per_second": 32.665,
392
+ "step": 30000
393
+ },
394
+ {
395
+ "epoch": 19.83,
396
+ "learning_rate": 7.875026260216393e-05,
397
+ "loss": 1.2298,
398
+ "step": 30500
399
+ },
400
+ {
401
+ "epoch": 20.16,
402
+ "learning_rate": 7.810416889260653e-05,
403
+ "loss": 1.2179,
404
+ "step": 31000
405
+ },
406
+ {
407
+ "epoch": 20.48,
408
+ "learning_rate": 7.74511408999066e-05,
409
+ "loss": 1.2067,
410
+ "step": 31500
411
+ },
412
+ {
413
+ "epoch": 20.81,
414
+ "learning_rate": 7.679133974894983e-05,
415
+ "loss": 1.2117,
416
+ "step": 32000
417
+ },
418
+ {
419
+ "epoch": 21.13,
420
+ "learning_rate": 7.612492823579745e-05,
421
+ "loss": 1.186,
422
+ "step": 32500
423
+ },
424
+ {
425
+ "epoch": 21.46,
426
+ "learning_rate": 7.545207078751857e-05,
427
+ "loss": 1.177,
428
+ "step": 33000
429
+ },
430
+ {
431
+ "epoch": 21.78,
432
+ "learning_rate": 7.477293342162039e-05,
433
+ "loss": 1.1876,
434
+ "step": 33500
435
+ },
436
+ {
437
+ "epoch": 22.11,
438
+ "learning_rate": 7.408768370508576e-05,
439
+ "loss": 1.1808,
440
+ "step": 34000
441
+ },
442
+ {
443
+ "epoch": 22.43,
444
+ "learning_rate": 7.339649071302867e-05,
445
+ "loss": 1.1701,
446
+ "step": 34500
447
+ },
448
+ {
449
+ "epoch": 22.76,
450
+ "learning_rate": 7.269952498697734e-05,
451
+ "loss": 1.1714,
452
+ "step": 35000
453
+ },
454
+ {
455
+ "epoch": 23.08,
456
+ "learning_rate": 7.199695849279576e-05,
457
+ "loss": 1.1605,
458
+ "step": 35500
459
+ },
460
+ {
461
+ "epoch": 23.41,
462
+ "learning_rate": 7.128896457825364e-05,
463
+ "loss": 1.155,
464
+ "step": 36000
465
+ },
466
+ {
467
+ "epoch": 23.73,
468
+ "learning_rate": 7.057571793025544e-05,
469
+ "loss": 1.148,
470
+ "step": 36500
471
+ },
472
+ {
473
+ "epoch": 24.06,
474
+ "learning_rate": 6.985739453173903e-05,
475
+ "loss": 1.142,
476
+ "step": 37000
477
+ },
478
+ {
479
+ "epoch": 24.38,
480
+ "learning_rate": 6.91341716182545e-05,
481
+ "loss": 1.1411,
482
+ "step": 37500
483
+ },
484
+ {
485
+ "epoch": 24.71,
486
+ "learning_rate": 6.840622763423391e-05,
487
+ "loss": 1.1299,
488
+ "step": 38000
489
+ },
490
+ {
491
+ "epoch": 25.03,
492
+ "learning_rate": 6.767374218896286e-05,
493
+ "loss": 1.1411,
494
+ "step": 38500
495
+ },
496
+ {
497
+ "epoch": 25.36,
498
+ "learning_rate": 6.693689601226458e-05,
499
+ "loss": 1.1077,
500
+ "step": 39000
501
+ },
502
+ {
503
+ "epoch": 25.68,
504
+ "learning_rate": 6.619587090990748e-05,
505
+ "loss": 1.12,
506
+ "step": 39500
507
+ },
508
+ {
509
+ "epoch": 26.01,
510
+ "learning_rate": 6.545084971874738e-05,
511
+ "loss": 1.1234,
512
+ "step": 40000
513
+ },
514
+ {
515
+ "epoch": 26.01,
516
+ "eval_loss": 2.1873600482940674,
517
+ "eval_runtime": 2.9334,
518
+ "eval_samples_per_second": 524.314,
519
+ "eval_steps_per_second": 33.068,
520
+ "step": 40000
521
+ },
522
+ {
523
+ "epoch": 26.33,
524
+ "learning_rate": 6.47020162616152e-05,
525
+ "loss": 1.0914,
526
+ "step": 40500
527
+ },
528
+ {
529
+ "epoch": 26.66,
530
+ "learning_rate": 6.394955530196147e-05,
531
+ "loss": 1.0948,
532
+ "step": 41000
533
+ },
534
+ {
535
+ "epoch": 26.98,
536
+ "learning_rate": 6.319365249826865e-05,
537
+ "loss": 1.104,
538
+ "step": 41500
539
+ },
540
+ {
541
+ "epoch": 27.31,
542
+ "learning_rate": 6.243449435824276e-05,
543
+ "loss": 1.0838,
544
+ "step": 42000
545
+ },
546
+ {
547
+ "epoch": 27.63,
548
+ "learning_rate": 6.167226819279528e-05,
549
+ "loss": 1.0952,
550
+ "step": 42500
551
+ },
552
+ {
553
+ "epoch": 27.96,
554
+ "learning_rate": 6.090716206982714e-05,
555
+ "loss": 1.0876,
556
+ "step": 43000
557
+ },
558
+ {
559
+ "epoch": 28.28,
560
+ "learning_rate": 6.0139364767825626e-05,
561
+ "loss": 1.0641,
562
+ "step": 43500
563
+ },
564
+ {
565
+ "epoch": 28.61,
566
+ "learning_rate": 5.9369065729286245e-05,
567
+ "loss": 1.0735,
568
+ "step": 44000
569
+ },
570
+ {
571
+ "epoch": 28.93,
572
+ "learning_rate": 5.859645501397048e-05,
573
+ "loss": 1.066,
574
+ "step": 44500
575
+ },
576
+ {
577
+ "epoch": 29.26,
578
+ "learning_rate": 5.782172325201155e-05,
579
+ "loss": 1.0598,
580
+ "step": 45000
581
+ },
582
+ {
583
+ "epoch": 29.58,
584
+ "learning_rate": 5.7045061596879134e-05,
585
+ "loss": 1.0457,
586
+ "step": 45500
587
+ },
588
+ {
589
+ "epoch": 29.91,
590
+ "learning_rate": 5.6266661678215216e-05,
591
+ "loss": 1.0597,
592
+ "step": 46000
593
+ },
594
+ {
595
+ "epoch": 30.23,
596
+ "learning_rate": 5.548671555455226e-05,
597
+ "loss": 1.041,
598
+ "step": 46500
599
+ },
600
+ {
601
+ "epoch": 30.56,
602
+ "learning_rate": 5.470541566592573e-05,
603
+ "loss": 1.0387,
604
+ "step": 47000
605
+ },
606
+ {
607
+ "epoch": 30.88,
608
+ "learning_rate": 5.392295478639225e-05,
609
+ "loss": 1.0336,
610
+ "step": 47500
611
+ },
612
+ {
613
+ "epoch": 31.21,
614
+ "learning_rate": 5.313952597646568e-05,
615
+ "loss": 1.0336,
616
+ "step": 48000
617
+ },
618
+ {
619
+ "epoch": 31.53,
620
+ "learning_rate": 5.235532253548213e-05,
621
+ "loss": 1.0209,
622
+ "step": 48500
623
+ },
624
+ {
625
+ "epoch": 31.86,
626
+ "learning_rate": 5.157053795390642e-05,
627
+ "loss": 1.0265,
628
+ "step": 49000
629
+ },
630
+ {
631
+ "epoch": 32.18,
632
+ "learning_rate": 5.078536586559104e-05,
633
+ "loss": 1.0256,
634
+ "step": 49500
635
+ },
636
+ {
637
+ "epoch": 32.51,
638
+ "learning_rate": 5e-05,
639
+ "loss": 1.0087,
640
+ "step": 50000
641
+ },
642
+ {
643
+ "epoch": 32.51,
644
+ "eval_loss": 2.3058323860168457,
645
+ "eval_runtime": 3.027,
646
+ "eval_samples_per_second": 508.087,
647
+ "eval_steps_per_second": 32.045,
648
+ "step": 50000
649
+ },
650
+ {
651
+ "epoch": 32.83,
652
+ "learning_rate": 4.921463413440898e-05,
653
+ "loss": 1.0137,
654
+ "step": 50500
655
+ },
656
+ {
657
+ "epoch": 33.16,
658
+ "learning_rate": 4.8429462046093585e-05,
659
+ "loss": 1.0106,
660
+ "step": 51000
661
+ },
662
+ {
663
+ "epoch": 33.49,
664
+ "learning_rate": 4.7644677464517874e-05,
665
+ "loss": 1.008,
666
+ "step": 51500
667
+ },
668
+ {
669
+ "epoch": 33.81,
670
+ "learning_rate": 4.6860474023534335e-05,
671
+ "loss": 1.0014,
672
+ "step": 52000
673
+ },
674
+ {
675
+ "epoch": 34.14,
676
+ "learning_rate": 4.607704521360776e-05,
677
+ "loss": 0.9956,
678
+ "step": 52500
679
+ },
680
+ {
681
+ "epoch": 34.46,
682
+ "learning_rate": 4.529458433407429e-05,
683
+ "loss": 0.9926,
684
+ "step": 53000
685
+ },
686
+ {
687
+ "epoch": 34.79,
688
+ "learning_rate": 4.451328444544774e-05,
689
+ "loss": 0.9865,
690
+ "step": 53500
691
+ },
692
+ {
693
+ "epoch": 35.11,
694
+ "learning_rate": 4.373333832178478e-05,
695
+ "loss": 0.9809,
696
+ "step": 54000
697
+ },
698
+ {
699
+ "epoch": 35.44,
700
+ "learning_rate": 4.295493840312087e-05,
701
+ "loss": 0.9809,
702
+ "step": 54500
703
+ },
704
+ {
705
+ "epoch": 35.76,
706
+ "learning_rate": 4.2178276747988446e-05,
707
+ "loss": 0.9785,
708
+ "step": 55000
709
+ },
710
+ {
711
+ "epoch": 36.09,
712
+ "learning_rate": 4.140354498602952e-05,
713
+ "loss": 0.9749,
714
+ "step": 55500
715
+ },
716
+ {
717
+ "epoch": 36.41,
718
+ "learning_rate": 4.063093427071376e-05,
719
+ "loss": 0.9737,
720
+ "step": 56000
721
+ },
722
+ {
723
+ "epoch": 36.74,
724
+ "learning_rate": 3.986063523217439e-05,
725
+ "loss": 0.9697,
726
+ "step": 56500
727
+ },
728
+ {
729
+ "epoch": 37.06,
730
+ "learning_rate": 3.9092837930172884e-05,
731
+ "loss": 0.9528,
732
+ "step": 57000
733
+ },
734
+ {
735
+ "epoch": 37.39,
736
+ "learning_rate": 3.832773180720475e-05,
737
+ "loss": 0.9622,
738
+ "step": 57500
739
+ },
740
+ {
741
+ "epoch": 37.71,
742
+ "learning_rate": 3.756550564175727e-05,
743
+ "loss": 0.9613,
744
+ "step": 58000
745
+ },
746
+ {
747
+ "epoch": 38.04,
748
+ "learning_rate": 3.680634750173137e-05,
749
+ "loss": 0.954,
750
+ "step": 58500
751
+ },
752
+ {
753
+ "epoch": 38.36,
754
+ "learning_rate": 3.605044469803854e-05,
755
+ "loss": 0.9494,
756
+ "step": 59000
757
+ },
758
+ {
759
+ "epoch": 38.69,
760
+ "learning_rate": 3.529798373838481e-05,
761
+ "loss": 0.9451,
762
+ "step": 59500
763
+ },
764
+ {
765
+ "epoch": 39.01,
766
+ "learning_rate": 3.4549150281252636e-05,
767
+ "loss": 0.946,
768
+ "step": 60000
769
+ },
770
+ {
771
+ "epoch": 39.01,
772
+ "eval_loss": 2.416886568069458,
773
+ "eval_runtime": 2.9777,
774
+ "eval_samples_per_second": 516.506,
775
+ "eval_steps_per_second": 32.575,
776
+ "step": 60000
777
+ },
778
+ {
779
+ "epoch": 39.34,
780
+ "learning_rate": 3.380412909009254e-05,
781
+ "loss": 0.939,
782
+ "step": 60500
783
+ },
784
+ {
785
+ "epoch": 39.66,
786
+ "learning_rate": 3.3063103987735433e-05,
787
+ "loss": 0.9388,
788
+ "step": 61000
789
+ },
790
+ {
791
+ "epoch": 39.99,
792
+ "learning_rate": 3.2326257811037155e-05,
793
+ "loss": 0.9315,
794
+ "step": 61500
795
+ },
796
+ {
797
+ "epoch": 40.31,
798
+ "learning_rate": 3.1593772365766105e-05,
799
+ "loss": 0.9318,
800
+ "step": 62000
801
+ },
802
+ {
803
+ "epoch": 40.64,
804
+ "learning_rate": 3.086582838174551e-05,
805
+ "loss": 0.9243,
806
+ "step": 62500
807
+ },
808
+ {
809
+ "epoch": 40.96,
810
+ "learning_rate": 3.0142605468260978e-05,
811
+ "loss": 0.9259,
812
+ "step": 63000
813
+ },
814
+ {
815
+ "epoch": 41.29,
816
+ "learning_rate": 2.9424282069744564e-05,
817
+ "loss": 0.9274,
818
+ "step": 63500
819
+ },
820
+ {
821
+ "epoch": 41.61,
822
+ "learning_rate": 2.8711035421746367e-05,
823
+ "loss": 0.9215,
824
+ "step": 64000
825
+ },
826
+ {
827
+ "epoch": 41.94,
828
+ "learning_rate": 2.8003041507204242e-05,
829
+ "loss": 0.9226,
830
+ "step": 64500
831
+ },
832
+ {
833
+ "epoch": 42.26,
834
+ "learning_rate": 2.7300475013022663e-05,
835
+ "loss": 0.9104,
836
+ "step": 65000
837
+ },
838
+ {
839
+ "epoch": 42.59,
840
+ "learning_rate": 2.660350928697134e-05,
841
+ "loss": 0.91,
842
+ "step": 65500
843
+ },
844
+ {
845
+ "epoch": 42.91,
846
+ "learning_rate": 2.591231629491423e-05,
847
+ "loss": 0.9108,
848
+ "step": 66000
849
+ },
850
+ {
851
+ "epoch": 43.24,
852
+ "learning_rate": 2.522706657837962e-05,
853
+ "loss": 0.9068,
854
+ "step": 66500
855
+ },
856
+ {
857
+ "epoch": 43.56,
858
+ "learning_rate": 2.4547929212481435e-05,
859
+ "loss": 0.9092,
860
+ "step": 67000
861
+ },
862
+ {
863
+ "epoch": 43.89,
864
+ "learning_rate": 2.3875071764202563e-05,
865
+ "loss": 0.9059,
866
+ "step": 67500
867
+ },
868
+ {
869
+ "epoch": 44.21,
870
+ "learning_rate": 2.3208660251050158e-05,
871
+ "loss": 0.8956,
872
+ "step": 68000
873
+ },
874
+ {
875
+ "epoch": 44.54,
876
+ "learning_rate": 2.2548859100093407e-05,
877
+ "loss": 0.9111,
878
+ "step": 68500
879
+ },
880
+ {
881
+ "epoch": 44.86,
882
+ "learning_rate": 2.1895831107393484e-05,
883
+ "loss": 0.9063,
884
+ "step": 69000
885
+ },
886
+ {
887
+ "epoch": 45.19,
888
+ "learning_rate": 2.124973739783609e-05,
889
+ "loss": 0.897,
890
+ "step": 69500
891
+ },
892
+ {
893
+ "epoch": 45.51,
894
+ "learning_rate": 2.061073738537635e-05,
895
+ "loss": 0.8985,
896
+ "step": 70000
897
+ },
898
+ {
899
+ "epoch": 45.51,
900
+ "eval_loss": 2.46510648727417,
901
+ "eval_runtime": 3.0226,
902
+ "eval_samples_per_second": 508.83,
903
+ "eval_steps_per_second": 32.091,
904
+ "step": 70000
905
+ },
906
+ {
907
+ "epoch": 45.84,
908
+ "learning_rate": 1.9978988733705807e-05,
909
+ "loss": 0.8948,
910
+ "step": 70500
911
+ },
912
+ {
913
+ "epoch": 46.16,
914
+ "learning_rate": 1.9354647317351188e-05,
915
+ "loss": 0.8855,
916
+ "step": 71000
917
+ },
918
+ {
919
+ "epoch": 46.49,
920
+ "learning_rate": 1.8737867183214757e-05,
921
+ "loss": 0.8908,
922
+ "step": 71500
923
+ },
924
+ {
925
+ "epoch": 46.81,
926
+ "learning_rate": 1.8128800512565513e-05,
927
+ "loss": 0.8945,
928
+ "step": 72000
929
+ },
930
+ {
931
+ "epoch": 47.14,
932
+ "learning_rate": 1.7527597583490822e-05,
933
+ "loss": 0.8896,
934
+ "step": 72500
935
+ },
936
+ {
937
+ "epoch": 47.46,
938
+ "learning_rate": 1.6934406733817414e-05,
939
+ "loss": 0.8817,
940
+ "step": 73000
941
+ },
942
+ {
943
+ "epoch": 47.79,
944
+ "learning_rate": 1.6349374324511345e-05,
945
+ "loss": 0.8842,
946
+ "step": 73500
947
+ },
948
+ {
949
+ "epoch": 48.11,
950
+ "learning_rate": 1.5772644703565565e-05,
951
+ "loss": 0.8852,
952
+ "step": 74000
953
+ },
954
+ {
955
+ "epoch": 48.44,
956
+ "learning_rate": 1.5204360170384286e-05,
957
+ "loss": 0.8803,
958
+ "step": 74500
959
+ },
960
+ {
961
+ "epoch": 48.76,
962
+ "learning_rate": 1.4644660940672627e-05,
963
+ "loss": 0.8763,
964
+ "step": 75000
965
+ },
966
+ {
967
+ "epoch": 49.09,
968
+ "learning_rate": 1.4093685111840566e-05,
969
+ "loss": 0.8765,
970
+ "step": 75500
971
+ },
972
+ {
973
+ "epoch": 49.41,
974
+ "learning_rate": 1.3551568628929434e-05,
975
+ "loss": 0.8757,
976
+ "step": 76000
977
+ },
978
+ {
979
+ "epoch": 49.74,
980
+ "learning_rate": 1.301844525106951e-05,
981
+ "loss": 0.8738,
982
+ "step": 76500
983
+ },
984
+ {
985
+ "epoch": 50.07,
986
+ "learning_rate": 1.2494446518477022e-05,
987
+ "loss": 0.8683,
988
+ "step": 77000
989
+ },
990
+ {
991
+ "epoch": 50.39,
992
+ "learning_rate": 1.1979701719998453e-05,
993
+ "loss": 0.8791,
994
+ "step": 77500
995
+ },
996
+ {
997
+ "epoch": 50.72,
998
+ "learning_rate": 1.1474337861210543e-05,
999
+ "loss": 0.867,
1000
+ "step": 78000
1001
+ },
1002
+ {
1003
+ "epoch": 51.04,
1004
+ "learning_rate": 1.097847963308351e-05,
1005
+ "loss": 0.8751,
1006
+ "step": 78500
1007
+ },
1008
+ {
1009
+ "epoch": 51.37,
1010
+ "learning_rate": 1.049224938121548e-05,
1011
+ "loss": 0.868,
1012
+ "step": 79000
1013
+ },
1014
+ {
1015
+ "epoch": 51.69,
1016
+ "learning_rate": 1.0015767075645471e-05,
1017
+ "loss": 0.8604,
1018
+ "step": 79500
1019
+ },
1020
+ {
1021
+ "epoch": 52.02,
1022
+ "learning_rate": 9.549150281252633e-06,
1023
+ "loss": 0.8658,
1024
+ "step": 80000
1025
+ },
1026
+ {
1027
+ "epoch": 52.02,
1028
+ "eval_loss": 2.4832496643066406,
1029
+ "eval_runtime": 2.9713,
1030
+ "eval_samples_per_second": 517.619,
1031
+ "eval_steps_per_second": 32.646,
1032
+ "step": 80000
1033
+ },
1034
+ {
1035
+ "epoch": 52.34,
1036
+ "learning_rate": 9.09251412874882e-06,
1037
+ "loss": 0.866,
1038
+ "step": 80500
1039
+ },
1040
+ {
1041
+ "epoch": 52.67,
1042
+ "learning_rate": 8.645971286271904e-06,
1043
+ "loss": 0.8648,
1044
+ "step": 81000
1045
+ },
1046
+ {
1047
+ "epoch": 52.99,
1048
+ "learning_rate": 8.209631931586498e-06,
1049
+ "loss": 0.8655,
1050
+ "step": 81500
1051
+ },
1052
+ {
1053
+ "epoch": 53.32,
1054
+ "learning_rate": 7.783603724899257e-06,
1055
+ "loss": 0.8679,
1056
+ "step": 82000
1057
+ },
1058
+ {
1059
+ "epoch": 53.64,
1060
+ "learning_rate": 7.367991782295391e-06,
1061
+ "loss": 0.8555,
1062
+ "step": 82500
1063
+ },
1064
+ {
1065
+ "epoch": 53.97,
1066
+ "learning_rate": 6.962898649802823e-06,
1067
+ "loss": 0.8593,
1068
+ "step": 83000
1069
+ },
1070
+ {
1071
+ "epoch": 54.29,
1072
+ "learning_rate": 6.568424278090446e-06,
1073
+ "loss": 0.863,
1074
+ "step": 83500
1075
+ },
1076
+ {
1077
+ "epoch": 54.62,
1078
+ "learning_rate": 6.184665997806832e-06,
1079
+ "loss": 0.8577,
1080
+ "step": 84000
1081
+ },
1082
+ {
1083
+ "epoch": 54.94,
1084
+ "learning_rate": 5.811718495565327e-06,
1085
+ "loss": 0.8646,
1086
+ "step": 84500
1087
+ },
1088
+ {
1089
+ "epoch": 55.27,
1090
+ "learning_rate": 5.449673790581611e-06,
1091
+ "loss": 0.8512,
1092
+ "step": 85000
1093
+ },
1094
+ {
1095
+ "epoch": 55.59,
1096
+ "learning_rate": 5.098621211969223e-06,
1097
+ "loss": 0.8623,
1098
+ "step": 85500
1099
+ },
1100
+ {
1101
+ "epoch": 55.92,
1102
+ "learning_rate": 4.758647376699032e-06,
1103
+ "loss": 0.8535,
1104
+ "step": 86000
1105
+ },
1106
+ {
1107
+ "epoch": 56.24,
1108
+ "learning_rate": 4.429836168227735e-06,
1109
+ "loss": 0.8414,
1110
+ "step": 86500
1111
+ },
1112
+ {
1113
+ "epoch": 56.57,
1114
+ "learning_rate": 4.112268715800943e-06,
1115
+ "loss": 0.8644,
1116
+ "step": 87000
1117
+ },
1118
+ {
1119
+ "epoch": 56.89,
1120
+ "learning_rate": 3.8060233744356633e-06,
1121
+ "loss": 0.8567,
1122
+ "step": 87500
1123
+ },
1124
+ {
1125
+ "epoch": 57.22,
1126
+ "learning_rate": 3.511175705587433e-06,
1127
+ "loss": 0.8577,
1128
+ "step": 88000
1129
+ },
1130
+ {
1131
+ "epoch": 57.54,
1132
+ "learning_rate": 3.2277984585066366e-06,
1133
+ "loss": 0.855,
1134
+ "step": 88500
1135
+ },
1136
+ {
1137
+ "epoch": 57.87,
1138
+ "learning_rate": 2.9559615522887273e-06,
1139
+ "loss": 0.8384,
1140
+ "step": 89000
1141
+ },
1142
+ {
1143
+ "epoch": 58.19,
1144
+ "learning_rate": 2.695732058622735e-06,
1145
+ "loss": 0.8549,
1146
+ "step": 89500
1147
+ },
1148
+ {
1149
+ "epoch": 58.52,
1150
+ "learning_rate": 2.4471741852423237e-06,
1151
+ "loss": 0.8479,
1152
+ "step": 90000
1153
+ },
1154
+ {
1155
+ "epoch": 58.52,
1156
+ "eval_loss": 2.5272536277770996,
1157
+ "eval_runtime": 2.9755,
1158
+ "eval_samples_per_second": 516.881,
1159
+ "eval_steps_per_second": 32.599,
1160
+ "step": 90000
1161
+ },
1162
+ {
1163
+ "epoch": 58.84,
1164
+ "learning_rate": 2.210349260083494e-06,
1165
+ "loss": 0.8509,
1166
+ "step": 90500
1167
+ },
1168
+ {
1169
+ "epoch": 59.17,
1170
+ "learning_rate": 1.985315716152847e-06,
1171
+ "loss": 0.8535,
1172
+ "step": 91000
1173
+ },
1174
+ {
1175
+ "epoch": 59.49,
1176
+ "learning_rate": 1.7721290771100961e-06,
1177
+ "loss": 0.8562,
1178
+ "step": 91500
1179
+ },
1180
+ {
1181
+ "epoch": 59.82,
1182
+ "learning_rate": 1.5708419435684462e-06,
1183
+ "loss": 0.8448,
1184
+ "step": 92000
1185
+ },
1186
+ {
1187
+ "epoch": 60.14,
1188
+ "learning_rate": 1.3815039801161721e-06,
1189
+ "loss": 0.8514,
1190
+ "step": 92500
1191
+ },
1192
+ {
1193
+ "epoch": 60.47,
1194
+ "learning_rate": 1.2041619030626284e-06,
1195
+ "loss": 0.8516,
1196
+ "step": 93000
1197
+ },
1198
+ {
1199
+ "epoch": 60.79,
1200
+ "learning_rate": 1.0388594689117071e-06,
1201
+ "loss": 0.8464,
1202
+ "step": 93500
1203
+ },
1204
+ {
1205
+ "epoch": 61.12,
1206
+ "learning_rate": 8.856374635655695e-07,
1207
+ "loss": 0.8573,
1208
+ "step": 94000
1209
+ },
1210
+ {
1211
+ "epoch": 61.44,
1212
+ "learning_rate": 7.445336922613067e-07,
1213
+ "loss": 0.8526,
1214
+ "step": 94500
1215
+ },
1216
+ {
1217
+ "epoch": 61.77,
1218
+ "learning_rate": 6.15582970243117e-07,
1219
+ "loss": 0.8404,
1220
+ "step": 95000
1221
+ },
1222
+ {
1223
+ "epoch": 62.09,
1224
+ "learning_rate": 4.988171141721232e-07,
1225
+ "loss": 0.8532,
1226
+ "step": 95500
1227
+ },
1228
+ {
1229
+ "epoch": 62.42,
1230
+ "learning_rate": 3.9426493427611177e-07,
1231
+ "loss": 0.852,
1232
+ "step": 96000
1233
+ },
1234
+ {
1235
+ "epoch": 62.74,
1236
+ "learning_rate": 3.019522272410202e-07,
1237
+ "loss": 0.8508,
1238
+ "step": 96500
1239
+ },
1240
+ {
1241
+ "epoch": 63.07,
1242
+ "learning_rate": 2.219017698460002e-07,
1243
+ "loss": 0.8475,
1244
+ "step": 97000
1245
+ },
1246
+ {
1247
+ "epoch": 63.39,
1248
+ "learning_rate": 1.5413331334360182e-07,
1249
+ "loss": 0.8491,
1250
+ "step": 97500
1251
+ },
1252
+ {
1253
+ "epoch": 63.72,
1254
+ "learning_rate": 9.866357858642205e-08,
1255
+ "loss": 0.8448,
1256
+ "step": 98000
1257
+ },
1258
+ {
1259
+ "epoch": 64.04,
1260
+ "learning_rate": 5.550625190150483e-08,
1261
+ "loss": 0.8596,
1262
+ "step": 98500
1263
+ },
1264
+ {
1265
+ "epoch": 64.37,
1266
+ "learning_rate": 2.467198171342e-08,
1267
+ "loss": 0.8478,
1268
+ "step": 99000
1269
+ },
1270
+ {
1271
+ "epoch": 64.69,
1272
+ "learning_rate": 6.1683759169706146e-09,
1273
+ "loss": 0.8531,
1274
+ "step": 99500
1275
+ },
1276
+ {
1277
+ "epoch": 65.02,
1278
+ "learning_rate": 0.0,
1279
+ "loss": 0.8508,
1280
+ "step": 100000
1281
+ },
1282
+ {
1283
+ "epoch": 65.02,
1284
+ "eval_loss": 2.5495502948760986,
1285
+ "eval_runtime": 3.0032,
1286
+ "eval_samples_per_second": 512.121,
1287
+ "eval_steps_per_second": 32.299,
1288
+ "step": 100000
1289
+ }
1290
+ ],
1291
+ "max_steps": 100000,
1292
+ "num_train_epochs": 66,
1293
+ "total_flos": 1.258672428813312e+17,
1294
+ "trial_name": null,
1295
+ "trial_params": null
1296
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f88fb219e94e0c8d612a300071c37a71bf56ac4a783af33d92c65d9437a94bfb
3
+ size 4152