Johhny1201 commited on
Commit
e9534f2
·
verified ·
1 Parent(s): 8e560c9

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +603 -0
trainer_state.json ADDED
@@ -0,0 +1,603 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9993746091307066,
5
+ "eval_steps": 500,
6
+ "global_step": 799,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.012507817385866166,
13
+ "grad_norm": 15.50371147983014,
14
+ "learning_rate": 6.25e-08,
15
+ "loss": 2.39,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.025015634771732333,
20
+ "grad_norm": 16.377043356637174,
21
+ "learning_rate": 1.25e-07,
22
+ "loss": 2.4206,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0375234521575985,
27
+ "grad_norm": 16.596531841198026,
28
+ "learning_rate": 1.875e-07,
29
+ "loss": 2.3735,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.050031269543464665,
34
+ "grad_norm": 11.7365845286366,
35
+ "learning_rate": 2.5e-07,
36
+ "loss": 2.3416,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.06253908692933083,
41
+ "grad_norm": 11.18800914938059,
42
+ "learning_rate": 3.1249999999999997e-07,
43
+ "loss": 2.3512,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.075046904315197,
48
+ "grad_norm": 7.018014667910915,
49
+ "learning_rate": 3.75e-07,
50
+ "loss": 2.2364,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.08755472170106317,
55
+ "grad_norm": 6.333030569859783,
56
+ "learning_rate": 4.375e-07,
57
+ "loss": 2.2577,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.10006253908692933,
62
+ "grad_norm": 4.7577630065749155,
63
+ "learning_rate": 5e-07,
64
+ "loss": 2.2443,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.1125703564727955,
69
+ "grad_norm": 4.270646018694652,
70
+ "learning_rate": 4.997613931640598e-07,
71
+ "loss": 2.0968,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.12507817385866166,
76
+ "grad_norm": 3.7719895745693144,
77
+ "learning_rate": 4.990460281220164e-07,
78
+ "loss": 2.0786,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.13758599124452783,
83
+ "grad_norm": 3.822159552325772,
84
+ "learning_rate": 4.978552704017834e-07,
85
+ "loss": 1.9348,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.150093808630394,
90
+ "grad_norm": 3.6639855496417577,
91
+ "learning_rate": 4.961913929868171e-07,
92
+ "loss": 1.9247,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.16260162601626016,
97
+ "grad_norm": 3.901015564156351,
98
+ "learning_rate": 4.940575719773205e-07,
99
+ "loss": 1.9058,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.17510944340212634,
104
+ "grad_norm": 3.6368127187289336,
105
+ "learning_rate": 4.914578805275297e-07,
106
+ "loss": 1.8596,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.18761726078799248,
111
+ "grad_norm": 2.992979752090568,
112
+ "learning_rate": 4.883972810706548e-07,
113
+ "loss": 1.8314,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.20012507817385866,
118
+ "grad_norm": 3.3873843484182466,
119
+ "learning_rate": 4.848816158463157e-07,
120
+ "loss": 1.8075,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.21263289555972484,
125
+ "grad_norm": 3.3435874749901937,
126
+ "learning_rate": 4.809175957485557e-07,
127
+ "loss": 1.772,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.225140712945591,
132
+ "grad_norm": 3.4374510709255492,
133
+ "learning_rate": 4.765127875157198e-07,
134
+ "loss": 1.7533,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.23764853033145716,
139
+ "grad_norm": 3.6689585209658873,
140
+ "learning_rate": 4.7167559928665085e-07,
141
+ "loss": 1.7891,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.2501563477173233,
146
+ "grad_norm": 3.8434986912697187,
147
+ "learning_rate": 4.664152645507743e-07,
148
+ "loss": 1.8084,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.2626641651031895,
153
+ "grad_norm": 2.8530782226054003,
154
+ "learning_rate": 4.6074182452270883e-07,
155
+ "loss": 1.7419,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.27517198248905567,
160
+ "grad_norm": 3.4661505185006183,
161
+ "learning_rate": 4.546661089750462e-07,
162
+ "loss": 1.7693,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.2876797998749218,
167
+ "grad_norm": 3.8663130075517875,
168
+ "learning_rate": 4.4819971556588977e-07,
169
+ "loss": 1.7335,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.300187617260788,
174
+ "grad_norm": 3.105644191017463,
175
+ "learning_rate": 4.4135498770060986e-07,
176
+ "loss": 1.8318,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.31269543464665417,
181
+ "grad_norm": 2.9687504404830465,
182
+ "learning_rate": 4.3414499097007697e-07,
183
+ "loss": 1.7344,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.3252032520325203,
188
+ "grad_norm": 3.475648300031111,
189
+ "learning_rate": 4.265834882103472e-07,
190
+ "loss": 1.7508,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.33771106941838647,
195
+ "grad_norm": 2.9742467574673226,
196
+ "learning_rate": 4.186849132314081e-07,
197
+ "loss": 1.7215,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.35021888680425267,
202
+ "grad_norm": 3.1787099419252667,
203
+ "learning_rate": 4.1046434326513303e-07,
204
+ "loss": 1.752,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.3627267041901188,
209
+ "grad_norm": 3.3558933667673667,
210
+ "learning_rate": 4.0193747018503614e-07,
211
+ "loss": 1.7647,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.37523452157598497,
216
+ "grad_norm": 3.5099375875164407,
217
+ "learning_rate": 3.9312057055276636e-07,
218
+ "loss": 1.7102,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.3877423389618512,
223
+ "grad_norm": 3.591142245670379,
224
+ "learning_rate": 3.8403047454851615e-07,
225
+ "loss": 1.7913,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.4002501563477173,
230
+ "grad_norm": 3.527758971554222,
231
+ "learning_rate": 3.7468453384465326e-07,
232
+ "loss": 1.757,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.41275797373358347,
237
+ "grad_norm": 2.9495017861006976,
238
+ "learning_rate": 3.6510058848389946e-07,
239
+ "loss": 1.7136,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.4252657911194497,
244
+ "grad_norm": 3.586714858280363,
245
+ "learning_rate": 3.5529693282528167e-07,
246
+ "loss": 1.7576,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.4377736085053158,
251
+ "grad_norm": 3.549806024215502,
252
+ "learning_rate": 3.4529228062285864e-07,
253
+ "loss": 1.7853,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.450281425891182,
258
+ "grad_norm": 3.3751243069059726,
259
+ "learning_rate": 3.351057293038841e-07,
260
+ "loss": 1.7072,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.4627892432770482,
265
+ "grad_norm": 3.0260065815967834,
266
+ "learning_rate": 3.247567235145928e-07,
267
+ "loss": 1.6807,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.47529706066291433,
272
+ "grad_norm": 2.640773300956867,
273
+ "learning_rate": 3.1426501800319697e-07,
274
+ "loss": 1.6653,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.4878048780487805,
279
+ "grad_norm": 3.2410844977370252,
280
+ "learning_rate": 3.03650639910942e-07,
281
+ "loss": 1.669,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.5003126954346466,
286
+ "grad_norm": 3.390139626995187,
287
+ "learning_rate": 2.9293385054320446e-07,
288
+ "loss": 1.6675,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.5128205128205128,
293
+ "grad_norm": 3.032547484860861,
294
+ "learning_rate": 2.8213510669360425e-07,
295
+ "loss": 1.6581,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.525328330206379,
300
+ "grad_norm": 3.0579494870999597,
301
+ "learning_rate": 2.712750215949578e-07,
302
+ "loss": 1.6945,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.5378361475922452,
307
+ "grad_norm": 3.3503627954919537,
308
+ "learning_rate": 2.6037432557161284e-07,
309
+ "loss": 1.7825,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.5503439649781113,
314
+ "grad_norm": 3.257201806161219,
315
+ "learning_rate": 2.494538264682706e-07,
316
+ "loss": 1.7377,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.5628517823639775,
321
+ "grad_norm": 3.6489839167255274,
322
+ "learning_rate": 2.385343699308345e-07,
323
+ "loss": 1.7268,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.5753595997498436,
328
+ "grad_norm": 3.105178719456824,
329
+ "learning_rate": 2.2763679961510135e-07,
330
+ "loss": 1.7919,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.5878674171357098,
335
+ "grad_norm": 3.050954751841531,
336
+ "learning_rate": 2.1678191739925102e-07,
337
+ "loss": 1.7826,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.600375234521576,
342
+ "grad_norm": 3.445788669098754,
343
+ "learning_rate": 2.0599044367608349e-07,
344
+ "loss": 1.6794,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.6128830519074422,
349
+ "grad_norm": 3.268302307641069,
350
+ "learning_rate": 1.9528297780080066e-07,
351
+ "loss": 1.7104,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.6253908692933083,
356
+ "grad_norm": 3.170030908287175,
357
+ "learning_rate": 1.8467995876983001e-07,
358
+ "loss": 1.6521,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.6253908692933083,
363
+ "eval_loss": 1.9212596416473389,
364
+ "eval_runtime": 52.2748,
365
+ "eval_samples_per_second": 108.752,
366
+ "eval_steps_per_second": 13.601,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.6378986866791745,
371
+ "grad_norm": 3.0945261612885893,
372
+ "learning_rate": 1.742016262057507e-07,
373
+ "loss": 1.748,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.6504065040650406,
378
+ "grad_norm": 3.5211162697926484,
379
+ "learning_rate": 1.6386798172279508e-07,
380
+ "loss": 1.702,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.6629143214509068,
385
+ "grad_norm": 2.82400028169299,
386
+ "learning_rate": 1.5369875074667365e-07,
387
+ "loss": 1.6578,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.6754221388367729,
392
+ "grad_norm": 3.4126613341000342,
393
+ "learning_rate": 1.4371334486160352e-07,
394
+ "loss": 1.6457,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.6879299562226392,
399
+ "grad_norm": 3.080219874111456,
400
+ "learning_rate": 1.3393082475641544e-07,
401
+ "loss": 1.6502,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.7004377736085053,
406
+ "grad_norm": 3.2091545865497135,
407
+ "learning_rate": 1.2436986384046786e-07,
408
+ "loss": 1.8363,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.7129455909943715,
413
+ "grad_norm": 2.9795468307340998,
414
+ "learning_rate": 1.150487125988224e-07,
415
+ "loss": 1.646,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.7254534083802376,
420
+ "grad_norm": 3.3814030136771636,
421
+ "learning_rate": 1.0598516375471978e-07,
422
+ "loss": 1.7004,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.7379612257661038,
427
+ "grad_norm": 2.9635741473039086,
428
+ "learning_rate": 9.719651830585668e-08,
429
+ "loss": 1.6428,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.7504690431519699,
434
+ "grad_norm": 3.341303101666117,
435
+ "learning_rate": 8.86995524992951e-08,
436
+ "loss": 1.6991,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.7629768605378362,
441
+ "grad_norm": 3.2990151037358117,
442
+ "learning_rate": 8.051048580804456e-08,
443
+ "loss": 1.7138,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.7754846779237023,
448
+ "grad_norm": 3.7604514079322544,
449
+ "learning_rate": 7.264494997044521e-08,
450
+ "loss": 1.6684,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.7879924953095685,
455
+ "grad_norm": 3.1090607489783264,
456
+ "learning_rate": 6.51179591514505e-08,
457
+ "loss": 1.6755,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.8005003126954346,
462
+ "grad_norm": 3.218850148666728,
463
+ "learning_rate": 5.794388128276798e-08,
464
+ "loss": 1.6433,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.8130081300813008,
469
+ "grad_norm": 3.381481766181872,
470
+ "learning_rate": 5.113641063656618e-08,
471
+ "loss": 1.7482,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.8255159474671669,
476
+ "grad_norm": 3.271540252470752,
477
+ "learning_rate": 4.470854168509819e-08,
478
+ "loss": 1.7041,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.8380237648530331,
483
+ "grad_norm": 2.8589321850258265,
484
+ "learning_rate": 3.867254429614281e-08,
485
+ "loss": 1.7051,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.8505315822388994,
490
+ "grad_norm": 2.8732647074299535,
491
+ "learning_rate": 3.3039940311609866e-08,
492
+ "loss": 1.6833,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.8630393996247655,
497
+ "grad_norm": 3.354421287892463,
498
+ "learning_rate": 2.782148155401806e-08,
499
+ "loss": 1.7021,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.8755472170106317,
504
+ "grad_norm": 3.255677211190932,
505
+ "learning_rate": 2.302712930282849e-08,
506
+ "loss": 1.724,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.8880550343964978,
511
+ "grad_norm": 3.6035792434299814,
512
+ "learning_rate": 1.8666035279809533e-08,
513
+ "loss": 1.6296,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.900562851782364,
518
+ "grad_norm": 2.932020685282119,
519
+ "learning_rate": 1.474652417972974e-08,
520
+ "loss": 1.7291,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.9130706691682301,
525
+ "grad_norm": 3.1442941331116767,
526
+ "learning_rate": 1.1276077779725335e-08,
527
+ "loss": 1.7443,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.9255784865540964,
532
+ "grad_norm": 2.856048609833897,
533
+ "learning_rate": 8.261320657674626e-09,
534
+ "loss": 1.6627,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.9380863039399625,
539
+ "grad_norm": 3.1836824042626897,
540
+ "learning_rate": 5.708007546841892e-09,
541
+ "loss": 1.631,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.9505941213258287,
546
+ "grad_norm": 3.246017077240641,
547
+ "learning_rate": 3.621012350927377e-09,
548
+ "loss": 1.7077,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.9631019387116948,
553
+ "grad_norm": 3.3755110020876815,
554
+ "learning_rate": 2.004318840493707e-09,
555
+ "loss": 1.7075,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.975609756097561,
560
+ "grad_norm": 3.215190396177525,
561
+ "learning_rate": 8.610130485265655e-10,
562
+ "loss": 1.6703,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.9881175734834271,
567
+ "grad_norm": 3.0778055684285497,
568
+ "learning_rate": 1.932773796461673e-10,
569
+ "loss": 1.6873,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.9993746091307066,
574
+ "step": 799,
575
+ "total_flos": 10053203263488.0,
576
+ "train_loss": 1.7967723003764624,
577
+ "train_runtime": 569.5544,
578
+ "train_samples_per_second": 89.832,
579
+ "train_steps_per_second": 1.403
580
+ }
581
+ ],
582
+ "logging_steps": 10,
583
+ "max_steps": 799,
584
+ "num_input_tokens_seen": 0,
585
+ "num_train_epochs": 1,
586
+ "save_steps": 5000,
587
+ "stateful_callbacks": {
588
+ "TrainerControl": {
589
+ "args": {
590
+ "should_epoch_stop": false,
591
+ "should_evaluate": false,
592
+ "should_log": false,
593
+ "should_save": true,
594
+ "should_training_stop": true
595
+ },
596
+ "attributes": {}
597
+ }
598
+ },
599
+ "total_flos": 10053203263488.0,
600
+ "train_batch_size": 4,
601
+ "trial_name": null,
602
+ "trial_params": null
603
+ }