aadityap commited on
Commit
bbf4c49
·
verified ·
1 Parent(s): cdaa32e

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +602 -0
trainer_state.json ADDED
@@ -0,0 +1,602 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 80,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.05,
13
+ "grad_norm": 0.03321758509380139,
14
+ "learning_rate": 6.25e-06,
15
+ "loss": 0.1785,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.1,
20
+ "grad_norm": 0.03270132297361398,
21
+ "learning_rate": 1.25e-05,
22
+ "loss": 0.1808,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.15,
27
+ "grad_norm": 0.031504129734254496,
28
+ "learning_rate": 1.8750000000000002e-05,
29
+ "loss": 0.1782,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.2,
34
+ "grad_norm": 0.036360864431180286,
35
+ "learning_rate": 2.5e-05,
36
+ "loss": 0.1894,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.25,
41
+ "grad_norm": 0.03433059360510697,
42
+ "learning_rate": 3.125e-05,
43
+ "loss": 0.1797,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.3,
48
+ "grad_norm": 0.035603554562599456,
49
+ "learning_rate": 3.7500000000000003e-05,
50
+ "loss": 0.1749,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.35,
55
+ "grad_norm": 0.039772993741593606,
56
+ "learning_rate": 4.375e-05,
57
+ "loss": 0.1717,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.4,
62
+ "grad_norm": 0.03937828545083326,
63
+ "learning_rate": 5e-05,
64
+ "loss": 0.1695,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.45,
69
+ "grad_norm": 0.04137387140973187,
70
+ "learning_rate": 4.997620553954645e-05,
71
+ "loss": 0.1581,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.5,
76
+ "grad_norm": 0.0366225618627523,
77
+ "learning_rate": 4.990486745229364e-05,
78
+ "loss": 0.1537,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.55,
83
+ "grad_norm": 0.031129456262764633,
84
+ "learning_rate": 4.9786121534345265e-05,
85
+ "loss": 0.1501,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.6,
90
+ "grad_norm": 0.038046792742615784,
91
+ "learning_rate": 4.962019382530521e-05,
92
+ "loss": 0.1404,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.65,
97
+ "grad_norm": 0.0406727748673051,
98
+ "learning_rate": 4.940740017799833e-05,
99
+ "loss": 0.1347,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.7,
104
+ "grad_norm": 0.03940085410684804,
105
+ "learning_rate": 4.914814565722671e-05,
106
+ "loss": 0.1321,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.75,
111
+ "grad_norm": 0.037688137420001884,
112
+ "learning_rate": 4.884292376870567e-05,
113
+ "loss": 0.1221,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.8,
118
+ "grad_norm": 0.03429282891214494,
119
+ "learning_rate": 4.849231551964771e-05,
120
+ "loss": 0.1203,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.85,
125
+ "grad_norm": 0.03184344902827262,
126
+ "learning_rate": 4.8096988312782174e-05,
127
+ "loss": 0.1126,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.9,
132
+ "grad_norm": 0.03618849185621362,
133
+ "learning_rate": 4.765769467591625e-05,
134
+ "loss": 0.1078,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.95,
139
+ "grad_norm": 0.034298104293600626,
140
+ "learning_rate": 4.717527082945554e-05,
141
+ "loss": 0.1048,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 1.0,
146
+ "grad_norm": 0.034935183074113044,
147
+ "learning_rate": 4.665063509461097e-05,
148
+ "loss": 0.0946,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 1.05,
153
+ "grad_norm": 0.031815811600406016,
154
+ "learning_rate": 4.608478614532215e-05,
155
+ "loss": 0.0914,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 1.1,
160
+ "grad_norm": 0.03215063512841389,
161
+ "learning_rate": 4.54788011072248e-05,
162
+ "loss": 0.0842,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 1.15,
167
+ "grad_norm": 0.030140183862344357,
168
+ "learning_rate": 4.4833833507280884e-05,
169
+ "loss": 0.0771,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 1.2,
174
+ "grad_norm": 0.0293337861992031,
175
+ "learning_rate": 4.415111107797445e-05,
176
+ "loss": 0.0719,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 1.25,
181
+ "grad_norm": 0.030203745320595544,
182
+ "learning_rate": 4.34319334202531e-05,
183
+ "loss": 0.0724,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 1.3,
188
+ "grad_norm": 0.028088817651504092,
189
+ "learning_rate": 4.267766952966369e-05,
190
+ "loss": 0.0723,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 1.35,
195
+ "grad_norm": 0.027148000968074536,
196
+ "learning_rate": 4.188975519039151e-05,
197
+ "loss": 0.0633,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 1.4,
202
+ "grad_norm": 0.02964713724375281,
203
+ "learning_rate": 4.1069690242163484e-05,
204
+ "loss": 0.0556,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 1.45,
209
+ "grad_norm": 0.027421116188822536,
210
+ "learning_rate": 4.021903572521802e-05,
211
+ "loss": 0.0532,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 1.5,
216
+ "grad_norm": 0.02716026363115862,
217
+ "learning_rate": 3.933941090877615e-05,
218
+ "loss": 0.0511,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 1.55,
223
+ "grad_norm": 0.026329098311513383,
224
+ "learning_rate": 3.84324902086706e-05,
225
+ "loss": 0.0563,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 1.6,
230
+ "grad_norm": 0.0280978968741894,
231
+ "learning_rate": 3.7500000000000003e-05,
232
+ "loss": 0.0446,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 1.65,
237
+ "grad_norm": 0.028428656261065784,
238
+ "learning_rate": 3.654371533087586e-05,
239
+ "loss": 0.0421,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 1.7,
244
+ "grad_norm": 0.03310118661665954,
245
+ "learning_rate": 3.556545654351749e-05,
246
+ "loss": 0.0436,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 1.75,
251
+ "grad_norm": 0.027621293127998756,
252
+ "learning_rate": 3.456708580912725e-05,
253
+ "loss": 0.0414,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 1.8,
258
+ "grad_norm": 0.031305980977346645,
259
+ "learning_rate": 3.355050358314172e-05,
260
+ "loss": 0.0424,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 1.85,
265
+ "grad_norm": 0.02821237752777818,
266
+ "learning_rate": 3.251764498760683e-05,
267
+ "loss": 0.0413,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 1.9,
272
+ "grad_norm": 0.02595028819757713,
273
+ "learning_rate": 3.147047612756302e-05,
274
+ "loss": 0.0354,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 1.95,
279
+ "grad_norm": 0.028249892893483154,
280
+ "learning_rate": 3.0410990348452573e-05,
281
+ "loss": 0.0425,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 2.0,
286
+ "grad_norm": 0.023890505381380638,
287
+ "learning_rate": 2.9341204441673266e-05,
288
+ "loss": 0.0302,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 2.05,
293
+ "grad_norm": 0.021580387415294815,
294
+ "learning_rate": 2.8263154805501297e-05,
295
+ "loss": 0.0314,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 2.1,
300
+ "grad_norm": 0.020588314809361626,
301
+ "learning_rate": 2.717889356869146e-05,
302
+ "loss": 0.0291,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 2.15,
307
+ "grad_norm": 0.020463281695351575,
308
+ "learning_rate": 2.6090484684133404e-05,
309
+ "loss": 0.0329,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 2.2,
314
+ "grad_norm": 0.04001010673333842,
315
+ "learning_rate": 2.5e-05,
316
+ "loss": 0.0345,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 2.25,
321
+ "grad_norm": 0.020584243302568767,
322
+ "learning_rate": 2.3909515315866605e-05,
323
+ "loss": 0.0272,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 2.3,
328
+ "grad_norm": 0.022447337126769707,
329
+ "learning_rate": 2.2821106431308544e-05,
330
+ "loss": 0.0289,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 2.35,
335
+ "grad_norm": 0.020637635680219337,
336
+ "learning_rate": 2.173684519449872e-05,
337
+ "loss": 0.0274,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 2.4,
342
+ "grad_norm": 0.019683363570100445,
343
+ "learning_rate": 2.0658795558326743e-05,
344
+ "loss": 0.0288,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 2.45,
349
+ "grad_norm": 0.031243945080274442,
350
+ "learning_rate": 1.958900965154743e-05,
351
+ "loss": 0.0387,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 2.5,
356
+ "grad_norm": 0.01997604058204314,
357
+ "learning_rate": 1.852952387243698e-05,
358
+ "loss": 0.0275,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 2.55,
363
+ "grad_norm": 0.022378313541631285,
364
+ "learning_rate": 1.7482355012393177e-05,
365
+ "loss": 0.0297,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 2.6,
370
+ "grad_norm": 0.01971201877928536,
371
+ "learning_rate": 1.6449496416858284e-05,
372
+ "loss": 0.0275,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 2.65,
377
+ "grad_norm": 0.022059176470837674,
378
+ "learning_rate": 1.5432914190872757e-05,
379
+ "loss": 0.0275,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 2.7,
384
+ "grad_norm": 0.01811241798703188,
385
+ "learning_rate": 1.443454345648252e-05,
386
+ "loss": 0.0281,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 2.75,
391
+ "grad_norm": 0.018238279525765698,
392
+ "learning_rate": 1.3456284669124158e-05,
393
+ "loss": 0.0272,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 2.8,
398
+ "grad_norm": 0.02037184481118651,
399
+ "learning_rate": 1.2500000000000006e-05,
400
+ "loss": 0.0283,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 2.85,
405
+ "grad_norm": 0.0192456262694011,
406
+ "learning_rate": 1.1567509791329401e-05,
407
+ "loss": 0.0321,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 2.9,
412
+ "grad_norm": 0.01550821162034626,
413
+ "learning_rate": 1.0660589091223855e-05,
414
+ "loss": 0.0242,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 2.95,
419
+ "grad_norm": 0.01858583076694851,
420
+ "learning_rate": 9.780964274781984e-06,
421
+ "loss": 0.0249,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 3.0,
426
+ "grad_norm": 0.019073828965141038,
427
+ "learning_rate": 8.930309757836517e-06,
428
+ "loss": 0.0269,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 3.05,
433
+ "grad_norm": 0.016131093812404212,
434
+ "learning_rate": 8.110244809608495e-06,
435
+ "loss": 0.0238,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 3.1,
440
+ "grad_norm": 0.017423126710067707,
441
+ "learning_rate": 7.3223304703363135e-06,
442
+ "loss": 0.0249,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 3.15,
447
+ "grad_norm": 0.017662503677228478,
448
+ "learning_rate": 6.568066579746901e-06,
449
+ "loss": 0.0255,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 3.2,
454
+ "grad_norm": 0.01695460814174754,
455
+ "learning_rate": 5.848888922025553e-06,
456
+ "loss": 0.0222,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 3.25,
461
+ "grad_norm": 0.016043806952368577,
462
+ "learning_rate": 5.166166492719124e-06,
463
+ "loss": 0.0236,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 3.3,
468
+ "grad_norm": 0.015613735206659563,
469
+ "learning_rate": 4.521198892775203e-06,
470
+ "loss": 0.0283,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 3.35,
475
+ "grad_norm": 0.016302457782708404,
476
+ "learning_rate": 3.9152138546778625e-06,
477
+ "loss": 0.0222,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 3.4,
482
+ "grad_norm": 0.015644529851688054,
483
+ "learning_rate": 3.3493649053890326e-06,
484
+ "loss": 0.0242,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 3.45,
489
+ "grad_norm": 0.017824515754806754,
490
+ "learning_rate": 2.8247291705444575e-06,
491
+ "loss": 0.0312,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 3.5,
496
+ "grad_norm": 0.01783878167725992,
497
+ "learning_rate": 2.3423053240837515e-06,
498
+ "loss": 0.0298,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 3.55,
503
+ "grad_norm": 0.01623915947132217,
504
+ "learning_rate": 1.9030116872178316e-06,
505
+ "loss": 0.0259,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 3.6,
510
+ "grad_norm": 0.014691878921968545,
511
+ "learning_rate": 1.5076844803522922e-06,
512
+ "loss": 0.0252,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 3.65,
517
+ "grad_norm": 0.0158012594720991,
518
+ "learning_rate": 1.1570762312943295e-06,
519
+ "loss": 0.0248,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 3.7,
524
+ "grad_norm": 0.016417378319219886,
525
+ "learning_rate": 8.51854342773295e-07,
526
+ "loss": 0.0256,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 3.75,
531
+ "grad_norm": 0.014394437548996657,
532
+ "learning_rate": 5.925998220016659e-07,
533
+ "loss": 0.0267,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 3.8,
538
+ "grad_norm": 0.015252447213812402,
539
+ "learning_rate": 3.7980617469479953e-07,
540
+ "loss": 0.0233,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 3.85,
545
+ "grad_norm": 0.022938011099557724,
546
+ "learning_rate": 2.1387846565474045e-07,
547
+ "loss": 0.0331,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 3.9,
552
+ "grad_norm": 0.019316460523955057,
553
+ "learning_rate": 9.513254770636137e-08,
554
+ "loss": 0.0317,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 3.95,
559
+ "grad_norm": 0.017247184889252695,
560
+ "learning_rate": 2.3794460453555047e-08,
561
+ "loss": 0.0251,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 4.0,
566
+ "grad_norm": 0.01626709020519667,
567
+ "learning_rate": 0.0,
568
+ "loss": 0.0234,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 4.0,
573
+ "step": 80,
574
+ "total_flos": 436992876216320.0,
575
+ "train_loss": 0.06461741614621133,
576
+ "train_runtime": 740.6897,
577
+ "train_samples_per_second": 0.848,
578
+ "train_steps_per_second": 0.108
579
+ }
580
+ ],
581
+ "logging_steps": 1,
582
+ "max_steps": 80,
583
+ "num_input_tokens_seen": 0,
584
+ "num_train_epochs": 4,
585
+ "save_steps": 500,
586
+ "stateful_callbacks": {
587
+ "TrainerControl": {
588
+ "args": {
589
+ "should_epoch_stop": false,
590
+ "should_evaluate": false,
591
+ "should_log": false,
592
+ "should_save": true,
593
+ "should_training_stop": true
594
+ },
595
+ "attributes": {}
596
+ }
597
+ },
598
+ "total_flos": 436992876216320.0,
599
+ "train_batch_size": 1,
600
+ "trial_name": null,
601
+ "trial_params": null
602
+ }