Fernandosr85 commited on
Commit
088701f
·
verified ·
1 Parent(s): 22bc096

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +557 -0
trainer_state.json ADDED
@@ -0,0 +1,557 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 13,
7
+ "global_step": 69,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.043478260869565216,
14
+ "grad_norm": 0.7634550333023071,
15
+ "learning_rate": 0.0,
16
+ "loss": 1.4302,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.08695652173913043,
21
+ "grad_norm": 0.6962459683418274,
22
+ "learning_rate": 3.75e-05,
23
+ "loss": 1.4551,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.13043478260869565,
28
+ "grad_norm": 0.5518960952758789,
29
+ "learning_rate": 7.5e-05,
30
+ "loss": 1.1797,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.17391304347826086,
35
+ "grad_norm": 0.5505545735359192,
36
+ "learning_rate": 0.0001125,
37
+ "loss": 1.2842,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.21739130434782608,
42
+ "grad_norm": 0.22443649172782898,
43
+ "learning_rate": 0.00015,
44
+ "loss": 1.0566,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.2608695652173913,
49
+ "grad_norm": 0.1337110549211502,
50
+ "learning_rate": 0.00014992117531118204,
51
+ "loss": 1.0601,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.30434782608695654,
56
+ "grad_norm": 0.13790792226791382,
57
+ "learning_rate": 0.00014968488534344127,
58
+ "loss": 1.1489,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.34782608695652173,
63
+ "grad_norm": 0.11578402668237686,
64
+ "learning_rate": 0.00014929168196294593,
65
+ "loss": 0.9451,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.391304347826087,
70
+ "grad_norm": 0.11953484266996384,
71
+ "learning_rate": 0.0001487424835144104,
72
+ "loss": 0.9878,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.43478260869565216,
77
+ "grad_norm": 0.10787869244813919,
78
+ "learning_rate": 0.00014803857267625851,
79
+ "loss": 1.061,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.4782608695652174,
84
+ "grad_norm": 0.0799792930483818,
85
+ "learning_rate": 0.0001471815934648687,
86
+ "loss": 1.1069,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.5217391304347826,
91
+ "grad_norm": 0.0694027692079544,
92
+ "learning_rate": 0.00014617354739489765,
93
+ "loss": 1.0718,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.5652173913043478,
98
+ "grad_norm": 0.06417166441679001,
99
+ "learning_rate": 0.0001450167888046499,
100
+ "loss": 1.0039,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.6086956521739131,
105
+ "grad_norm": 0.06427249312400818,
106
+ "learning_rate": 0.0001437140193574118,
107
+ "loss": 0.9702,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.6521739130434783,
112
+ "grad_norm": 0.0544593371450901,
113
+ "learning_rate": 0.00014226828173159164,
114
+ "loss": 0.8267,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.6956521739130435,
119
+ "grad_norm": 0.07362852245569229,
120
+ "learning_rate": 0.0001406829525144033,
121
+ "loss": 0.9722,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.7391304347826086,
126
+ "grad_norm": 0.07204615324735641,
127
+ "learning_rate": 0.00013896173431569029,
128
+ "loss": 0.9663,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.7391304347826086,
133
+ "eval_loss": 0.90576171875,
134
+ "eval_runtime": 4.3927,
135
+ "eval_samples_per_second": 2.277,
136
+ "eval_steps_per_second": 0.455,
137
+ "step": 17
138
+ },
139
+ {
140
+ "epoch": 0.782608695652174,
141
+ "grad_norm": 0.06425914168357849,
142
+ "learning_rate": 0.00013710864712030893,
143
+ "loss": 0.9819,
144
+ "step": 18
145
+ },
146
+ {
147
+ "epoch": 0.8260869565217391,
148
+ "grad_norm": 0.036391615867614746,
149
+ "learning_rate": 0.0001351280188992675,
150
+ "loss": 0.8041,
151
+ "step": 19
152
+ },
153
+ {
154
+ "epoch": 0.8695652173913043,
155
+ "grad_norm": 0.04094429686665535,
156
+ "learning_rate": 0.0001330244755015493,
157
+ "loss": 1.0254,
158
+ "step": 20
159
+ },
160
+ {
161
+ "epoch": 0.9130434782608695,
162
+ "grad_norm": 0.04050302132964134,
163
+ "learning_rate": 0.00013080292985022816,
164
+ "loss": 1.0042,
165
+ "step": 21
166
+ },
167
+ {
168
+ "epoch": 0.9565217391304348,
169
+ "grad_norm": 0.038431670516729355,
170
+ "learning_rate": 0.0001284685704681088,
171
+ "loss": 0.9805,
172
+ "step": 22
173
+ },
174
+ {
175
+ "epoch": 1.0,
176
+ "grad_norm": 0.04781308397650719,
177
+ "learning_rate": 0.0001260268493596915,
178
+ "loss": 1.0522,
179
+ "step": 23
180
+ },
181
+ {
182
+ "epoch": 1.0434782608695652,
183
+ "grad_norm": 0.03875584527850151,
184
+ "learning_rate": 0.000123483469277763,
185
+ "loss": 0.9946,
186
+ "step": 24
187
+ },
188
+ {
189
+ "epoch": 1.0869565217391304,
190
+ "grad_norm": 0.038722969591617584,
191
+ "learning_rate": 0.000120844370404353,
192
+ "loss": 1.0386,
193
+ "step": 25
194
+ },
195
+ {
196
+ "epoch": 1.1304347826086956,
197
+ "grad_norm": 0.036717191338539124,
198
+ "learning_rate": 0.00011811571647716395,
199
+ "loss": 0.8828,
200
+ "step": 26
201
+ },
202
+ {
203
+ "epoch": 1.1739130434782608,
204
+ "grad_norm": 0.04138258099555969,
205
+ "learning_rate": 0.00011530388039387589,
206
+ "loss": 0.9724,
207
+ "step": 27
208
+ },
209
+ {
210
+ "epoch": 1.2173913043478262,
211
+ "grad_norm": 0.035979870706796646,
212
+ "learning_rate": 0.00011241542932794879,
213
+ "loss": 0.8765,
214
+ "step": 28
215
+ },
216
+ {
217
+ "epoch": 1.2608695652173914,
218
+ "grad_norm": 0.03545432165265083,
219
+ "learning_rate": 0.00010945710939068444,
220
+ "loss": 0.9277,
221
+ "step": 29
222
+ },
223
+ {
224
+ "epoch": 1.3043478260869565,
225
+ "grad_norm": 0.04141005128622055,
226
+ "learning_rate": 0.00010643582987537115,
227
+ "loss": 0.9895,
228
+ "step": 30
229
+ },
230
+ {
231
+ "epoch": 1.3043478260869565,
232
+ "eval_loss": 0.873291015625,
233
+ "eval_runtime": 4.3467,
234
+ "eval_samples_per_second": 2.301,
235
+ "eval_steps_per_second": 0.46,
236
+ "step": 30
237
+ },
238
+ {
239
+ "epoch": 1.3478260869565217,
240
+ "grad_norm": 0.03391469269990921,
241
+ "learning_rate": 0.00010335864712030894,
242
+ "loss": 0.824,
243
+ "step": 31
244
+ },
245
+ {
246
+ "epoch": 1.391304347826087,
247
+ "grad_norm": 0.032858580350875854,
248
+ "learning_rate": 0.00010023274802840461,
249
+ "loss": 0.8816,
250
+ "step": 32
251
+ },
252
+ {
253
+ "epoch": 1.434782608695652,
254
+ "grad_norm": 0.03364916890859604,
255
+ "learning_rate": 9.706543328182691e-05,
256
+ "loss": 0.9536,
257
+ "step": 33
258
+ },
259
+ {
260
+ "epoch": 1.4782608695652173,
261
+ "grad_norm": 0.034717388451099396,
262
+ "learning_rate": 9.386410029092507e-05,
263
+ "loss": 1.0142,
264
+ "step": 34
265
+ },
266
+ {
267
+ "epoch": 1.5217391304347827,
268
+ "grad_norm": 0.040148813277482986,
269
+ "learning_rate": 9.063622591723429e-05,
270
+ "loss": 0.9756,
271
+ "step": 35
272
+ },
273
+ {
274
+ "epoch": 1.5652173913043477,
275
+ "grad_norm": 0.03408735245466232,
276
+ "learning_rate": 8.738934901091897e-05,
277
+ "loss": 0.9312,
278
+ "step": 36
279
+ },
280
+ {
281
+ "epoch": 1.608695652173913,
282
+ "grad_norm": 0.0321255587041378,
283
+ "learning_rate": 8.413105280343893e-05,
284
+ "loss": 0.9043,
285
+ "step": 37
286
+ },
287
+ {
288
+ "epoch": 1.6521739130434783,
289
+ "grad_norm": 0.03103114850819111,
290
+ "learning_rate": 8.086894719656108e-05,
291
+ "loss": 0.7712,
292
+ "step": 38
293
+ },
294
+ {
295
+ "epoch": 1.6956521739130435,
296
+ "grad_norm": 0.03416038304567337,
297
+ "learning_rate": 7.761065098908105e-05,
298
+ "loss": 0.9062,
299
+ "step": 39
300
+ },
301
+ {
302
+ "epoch": 1.7391304347826086,
303
+ "grad_norm": 0.03245851770043373,
304
+ "learning_rate": 7.436377408276571e-05,
305
+ "loss": 0.9055,
306
+ "step": 40
307
+ },
308
+ {
309
+ "epoch": 1.7826086956521738,
310
+ "grad_norm": 0.03590395301580429,
311
+ "learning_rate": 7.113589970907489e-05,
312
+ "loss": 0.9229,
313
+ "step": 41
314
+ },
315
+ {
316
+ "epoch": 1.8260869565217392,
317
+ "grad_norm": 0.032267358154058456,
318
+ "learning_rate": 6.793456671817309e-05,
319
+ "loss": 0.7598,
320
+ "step": 42
321
+ },
322
+ {
323
+ "epoch": 1.8695652173913042,
324
+ "grad_norm": 0.03612128645181656,
325
+ "learning_rate": 6.476725197159537e-05,
326
+ "loss": 0.9722,
327
+ "step": 43
328
+ },
329
+ {
330
+ "epoch": 1.8695652173913042,
331
+ "eval_loss": 0.8590087890625,
332
+ "eval_runtime": 4.3562,
333
+ "eval_samples_per_second": 2.296,
334
+ "eval_steps_per_second": 0.459,
335
+ "step": 43
336
+ },
337
+ {
338
+ "epoch": 1.9130434782608696,
339
+ "grad_norm": 0.03612411394715309,
340
+ "learning_rate": 6.164135287969106e-05,
341
+ "loss": 0.9539,
342
+ "step": 44
343
+ },
344
+ {
345
+ "epoch": 1.9565217391304348,
346
+ "grad_norm": 0.03321658819913864,
347
+ "learning_rate": 5.856417012462883e-05,
348
+ "loss": 0.9355,
349
+ "step": 45
350
+ },
351
+ {
352
+ "epoch": 2.0,
353
+ "grad_norm": 0.03923092409968376,
354
+ "learning_rate": 5.554289060931557e-05,
355
+ "loss": 0.9976,
356
+ "step": 46
357
+ },
358
+ {
359
+ "epoch": 2.0434782608695654,
360
+ "grad_norm": 0.03545928746461868,
361
+ "learning_rate": 5.258457067205121e-05,
362
+ "loss": 0.9478,
363
+ "step": 47
364
+ },
365
+ {
366
+ "epoch": 2.0869565217391304,
367
+ "grad_norm": 0.033447980880737305,
368
+ "learning_rate": 4.969611960612409e-05,
369
+ "loss": 0.9971,
370
+ "step": 48
371
+ },
372
+ {
373
+ "epoch": 2.130434782608696,
374
+ "grad_norm": 0.030967000871896744,
375
+ "learning_rate": 4.688428352283603e-05,
376
+ "loss": 0.8506,
377
+ "step": 49
378
+ },
379
+ {
380
+ "epoch": 2.1739130434782608,
381
+ "grad_norm": 0.03545485809445381,
382
+ "learning_rate": 4.4155629595646986e-05,
383
+ "loss": 0.9329,
384
+ "step": 50
385
+ },
386
+ {
387
+ "epoch": 2.217391304347826,
388
+ "grad_norm": 0.03325090929865837,
389
+ "learning_rate": 4.151653072223699e-05,
390
+ "loss": 0.8462,
391
+ "step": 51
392
+ },
393
+ {
394
+ "epoch": 2.260869565217391,
395
+ "grad_norm": 0.029995175078511238,
396
+ "learning_rate": 3.8973150640308497e-05,
397
+ "loss": 0.897,
398
+ "step": 52
399
+ },
400
+ {
401
+ "epoch": 2.3043478260869565,
402
+ "grad_norm": 0.03587409108877182,
403
+ "learning_rate": 3.653142953189118e-05,
404
+ "loss": 0.9556,
405
+ "step": 53
406
+ },
407
+ {
408
+ "epoch": 2.3478260869565215,
409
+ "grad_norm": 0.03194161877036095,
410
+ "learning_rate": 3.419707014977183e-05,
411
+ "loss": 0.8,
412
+ "step": 54
413
+ },
414
+ {
415
+ "epoch": 2.391304347826087,
416
+ "grad_norm": 0.03137723729014397,
417
+ "learning_rate": 3.197552449845067e-05,
418
+ "loss": 0.8569,
419
+ "step": 55
420
+ },
421
+ {
422
+ "epoch": 2.4347826086956523,
423
+ "grad_norm": 0.032495368272066116,
424
+ "learning_rate": 2.987198110073249e-05,
425
+ "loss": 0.9253,
426
+ "step": 56
427
+ },
428
+ {
429
+ "epoch": 2.4347826086956523,
430
+ "eval_loss": 0.85205078125,
431
+ "eval_runtime": 4.3748,
432
+ "eval_samples_per_second": 2.286,
433
+ "eval_steps_per_second": 0.457,
434
+ "step": 56
435
+ },
436
+ {
437
+ "epoch": 2.4782608695652173,
438
+ "grad_norm": 0.03498929366469383,
439
+ "learning_rate": 2.7891352879691055e-05,
440
+ "loss": 0.9917,
441
+ "step": 57
442
+ },
443
+ {
444
+ "epoch": 2.5217391304347827,
445
+ "grad_norm": 0.03751672804355621,
446
+ "learning_rate": 2.6038265684309694e-05,
447
+ "loss": 0.9478,
448
+ "step": 58
449
+ },
450
+ {
451
+ "epoch": 2.5652173913043477,
452
+ "grad_norm": 0.03346884623169899,
453
+ "learning_rate": 2.4317047485596673e-05,
454
+ "loss": 0.9106,
455
+ "step": 59
456
+ },
457
+ {
458
+ "epoch": 2.608695652173913,
459
+ "grad_norm": 0.033841464668512344,
460
+ "learning_rate": 2.2731718268408336e-05,
461
+ "loss": 0.8862,
462
+ "step": 60
463
+ },
464
+ {
465
+ "epoch": 2.6521739130434785,
466
+ "grad_norm": 0.030843093991279602,
467
+ "learning_rate": 2.1285980642588184e-05,
468
+ "loss": 0.7553,
469
+ "step": 61
470
+ },
471
+ {
472
+ "epoch": 2.6956521739130435,
473
+ "grad_norm": 0.03349591791629791,
474
+ "learning_rate": 1.9983211195350096e-05,
475
+ "loss": 0.8867,
476
+ "step": 62
477
+ },
478
+ {
479
+ "epoch": 2.7391304347826084,
480
+ "grad_norm": 0.03341041877865791,
481
+ "learning_rate": 1.8826452605102335e-05,
482
+ "loss": 0.8867,
483
+ "step": 63
484
+ },
485
+ {
486
+ "epoch": 2.782608695652174,
487
+ "grad_norm": 0.03790866583585739,
488
+ "learning_rate": 1.781840653513128e-05,
489
+ "loss": 0.9058,
490
+ "step": 64
491
+ },
492
+ {
493
+ "epoch": 2.8260869565217392,
494
+ "grad_norm": 0.02903052605688572,
495
+ "learning_rate": 1.696142732374149e-05,
496
+ "loss": 0.7474,
497
+ "step": 65
498
+ },
499
+ {
500
+ "epoch": 2.869565217391304,
501
+ "grad_norm": 0.032797202467918396,
502
+ "learning_rate": 1.625751648558959e-05,
503
+ "loss": 0.9536,
504
+ "step": 66
505
+ },
506
+ {
507
+ "epoch": 2.9130434782608696,
508
+ "grad_norm": 0.032798297703266144,
509
+ "learning_rate": 1.570831803705405e-05,
510
+ "loss": 0.939,
511
+ "step": 67
512
+ },
513
+ {
514
+ "epoch": 2.9565217391304346,
515
+ "grad_norm": 0.03150875121355057,
516
+ "learning_rate": 1.5315114656558717e-05,
517
+ "loss": 0.9233,
518
+ "step": 68
519
+ },
520
+ {
521
+ "epoch": 3.0,
522
+ "grad_norm": 0.04027010127902031,
523
+ "learning_rate": 1.5078824688817951e-05,
524
+ "loss": 0.9819,
525
+ "step": 69
526
+ },
527
+ {
528
+ "epoch": 3.0,
529
+ "eval_loss": 0.849609375,
530
+ "eval_runtime": 4.3942,
531
+ "eval_samples_per_second": 2.276,
532
+ "eval_steps_per_second": 0.455,
533
+ "step": 69
534
+ }
535
+ ],
536
+ "logging_steps": 1.0,
537
+ "max_steps": 69,
538
+ "num_input_tokens_seen": 0,
539
+ "num_train_epochs": 3,
540
+ "save_steps": 0,
541
+ "stateful_callbacks": {
542
+ "TrainerControl": {
543
+ "args": {
544
+ "should_epoch_stop": false,
545
+ "should_evaluate": false,
546
+ "should_log": false,
547
+ "should_save": true,
548
+ "should_training_stop": true
549
+ },
550
+ "attributes": {}
551
+ }
552
+ },
553
+ "total_flos": 3.464101375758041e+18,
554
+ "train_batch_size": 1,
555
+ "trial_name": null,
556
+ "trial_params": null
557
+ }