yueqis commited on
Commit
5bd9633
·
verified ·
1 Parent(s): 3337938

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +489 -0
trainer_state.json ADDED
@@ -0,0 +1,489 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.3822651045256147,
6
+ "eval_steps": 500,
7
+ "global_step": 650,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.036756260050539856,
14
+ "grad_norm": 1.136057734489441,
15
+ "learning_rate": 1.0975609756097562e-05,
16
+ "loss": 1.1345,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.07351252010107971,
21
+ "grad_norm": 0.5754098296165466,
22
+ "learning_rate": 2.3170731707317075e-05,
23
+ "loss": 0.9743,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.11026878015161957,
28
+ "grad_norm": 0.5011259317398071,
29
+ "learning_rate": 3.5365853658536584e-05,
30
+ "loss": 0.9405,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.14702504020215942,
35
+ "grad_norm": 0.453047513961792,
36
+ "learning_rate": 4.75609756097561e-05,
37
+ "loss": 0.9082,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.1837813002526993,
42
+ "grad_norm": 0.3960385024547577,
43
+ "learning_rate": 4.998685537302135e-05,
44
+ "loss": 0.8931,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.22053756030323915,
49
+ "grad_norm": 0.41752490401268005,
50
+ "learning_rate": 4.9933479014963055e-05,
51
+ "loss": 0.873,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.257293820353779,
56
+ "grad_norm": 0.4332248270511627,
57
+ "learning_rate": 4.9839137016353147e-05,
58
+ "loss": 0.8851,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.29405008040431885,
63
+ "grad_norm": 0.41378816962242126,
64
+ "learning_rate": 4.970398438084758e-05,
65
+ "loss": 0.9094,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.33080634045485874,
70
+ "grad_norm": 0.40073126554489136,
71
+ "learning_rate": 4.952824316387163e-05,
72
+ "loss": 0.8963,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.3675626005053986,
77
+ "grad_norm": 0.49767372012138367,
78
+ "learning_rate": 4.931220210778332e-05,
79
+ "loss": 0.8639,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.4043188605559384,
84
+ "grad_norm": 0.3762946128845215,
85
+ "learning_rate": 4.905621616747054e-05,
86
+ "loss": 0.858,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.4410751206064783,
91
+ "grad_norm": 0.37841302156448364,
92
+ "learning_rate": 4.876070592716105e-05,
93
+ "loss": 0.8514,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.47783138065701813,
98
+ "grad_norm": 0.39247384667396545,
99
+ "learning_rate": 4.842615690940373e-05,
100
+ "loss": 0.848,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.514587640707558,
105
+ "grad_norm": 0.36395716667175293,
106
+ "learning_rate": 4.80531187773563e-05,
107
+ "loss": 0.8735,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.5513439007580979,
112
+ "grad_norm": 0.369722455739975,
113
+ "learning_rate": 4.7642204431690206e-05,
114
+ "loss": 0.8525,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.5881001608086377,
119
+ "grad_norm": 0.3555547297000885,
120
+ "learning_rate": 4.71940890035964e-05,
121
+ "loss": 0.8892,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.6248564208591776,
126
+ "grad_norm": 0.36071062088012695,
127
+ "learning_rate": 4.6709508745546575e-05,
128
+ "loss": 0.835,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.6616126809097175,
133
+ "grad_norm": 0.33451011776924133,
134
+ "learning_rate": 4.618925982163232e-05,
135
+ "loss": 0.8515,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.6983689409602573,
140
+ "grad_norm": 0.3431645631790161,
141
+ "learning_rate": 4.563419699946956e-05,
142
+ "loss": 0.862,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.7351252010107971,
147
+ "grad_norm": 0.3229110538959503,
148
+ "learning_rate": 4.504523224581762e-05,
149
+ "loss": 0.8708,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.771881461061337,
154
+ "grad_norm": 0.3718363642692566,
155
+ "learning_rate": 4.442333322822028e-05,
156
+ "loss": 0.8445,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.8086377211118768,
161
+ "grad_norm": 0.389436811208725,
162
+ "learning_rate": 4.376952172513046e-05,
163
+ "loss": 0.8711,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.8453939811624167,
168
+ "grad_norm": 0.38827431201934814,
169
+ "learning_rate": 4.308487194713097e-05,
170
+ "loss": 0.884,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.8821502412129566,
175
+ "grad_norm": 0.34896695613861084,
176
+ "learning_rate": 4.2370508772009334e-05,
177
+ "loss": 0.8524,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.9189065012634965,
182
+ "grad_norm": 0.35946622490882874,
183
+ "learning_rate": 4.162760589658649e-05,
184
+ "loss": 0.847,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.9556627613140363,
189
+ "grad_norm": 0.4428149461746216,
190
+ "learning_rate": 4.0857383908336076e-05,
191
+ "loss": 0.8585,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.9924190213645762,
196
+ "grad_norm": 0.3346478044986725,
197
+ "learning_rate": 4.0061108279962336e-05,
198
+ "loss": 0.8696,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 1.025729382035378,
203
+ "grad_norm": 0.47293058037757874,
204
+ "learning_rate": 3.924008729023185e-05,
205
+ "loss": 0.7543,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 1.0624856420859177,
210
+ "grad_norm": 0.35213029384613037,
211
+ "learning_rate": 3.8395669874474915e-05,
212
+ "loss": 0.7643,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 1.0992419021364577,
217
+ "grad_norm": 0.3495759069919586,
218
+ "learning_rate": 3.752924340828837e-05,
219
+ "loss": 0.7147,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 1.1359981621869975,
224
+ "grad_norm": 0.33258387446403503,
225
+ "learning_rate": 3.66422314280811e-05,
226
+ "loss": 0.7164,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 1.1727544222375372,
231
+ "grad_norm": 0.3696906566619873,
232
+ "learning_rate": 3.573609129220748e-05,
233
+ "loss": 0.6989,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 1.2095106822880772,
238
+ "grad_norm": 0.3357951045036316,
239
+ "learning_rate": 3.4812311786531426e-05,
240
+ "loss": 0.7079,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 1.246266942338617,
245
+ "grad_norm": 0.3242957592010498,
246
+ "learning_rate": 3.387241067835513e-05,
247
+ "loss": 0.7193,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 1.283023202389157,
252
+ "grad_norm": 0.34572961926460266,
253
+ "learning_rate": 3.2917932222731325e-05,
254
+ "loss": 0.7268,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 1.3197794624396968,
259
+ "grad_norm": 0.33331671357154846,
260
+ "learning_rate": 3.195044462525636e-05,
261
+ "loss": 0.7029,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 1.3565357224902366,
266
+ "grad_norm": 0.3374245762825012,
267
+ "learning_rate": 3.097153746551248e-05,
268
+ "loss": 0.7378,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 1.3932919825407764,
273
+ "grad_norm": 0.32257580757141113,
274
+ "learning_rate": 2.998281908539272e-05,
275
+ "loss": 0.7119,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 1.4300482425913164,
280
+ "grad_norm": 0.31712806224823,
281
+ "learning_rate": 2.898591394659948e-05,
282
+ "loss": 0.7254,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 1.4668045026418561,
287
+ "grad_norm": 0.32865336537361145,
288
+ "learning_rate": 2.798245996165813e-05,
289
+ "loss": 0.7233,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 1.5035607626923961,
294
+ "grad_norm": 0.33550089597702026,
295
+ "learning_rate": 2.697410580283107e-05,
296
+ "loss": 0.7412,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 1.540317022742936,
301
+ "grad_norm": 0.3236851394176483,
302
+ "learning_rate": 2.5962508193353542e-05,
303
+ "loss": 0.7113,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 1.5770732827934757,
308
+ "grad_norm": 0.32259735465049744,
309
+ "learning_rate": 2.4949329185441666e-05,
310
+ "loss": 0.7152,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 1.6138295428440155,
315
+ "grad_norm": 0.3176753520965576,
316
+ "learning_rate": 2.3936233429545054e-05,
317
+ "loss": 0.7301,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 1.6505858028945555,
322
+ "grad_norm": 0.3171616792678833,
323
+ "learning_rate": 2.2924885439330477e-05,
324
+ "loss": 0.71,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 1.6873420629450955,
329
+ "grad_norm": 0.34114307165145874,
330
+ "learning_rate": 2.1916946856890293e-05,
331
+ "loss": 0.7058,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 1.7240983229956353,
336
+ "grad_norm": 0.3279208838939667,
337
+ "learning_rate": 2.0914073722668788e-05,
338
+ "loss": 0.6893,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 1.760854583046175,
343
+ "grad_norm": 0.34568503499031067,
344
+ "learning_rate": 1.9917913754592183e-05,
345
+ "loss": 0.7079,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 1.7976108430967148,
350
+ "grad_norm": 0.3167240023612976,
351
+ "learning_rate": 1.8930103640872313e-05,
352
+ "loss": 0.7249,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 1.8343671031472548,
357
+ "grad_norm": 0.31115061044692993,
358
+ "learning_rate": 1.795226635093226e-05,
359
+ "loss": 0.686,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 1.8711233631977946,
364
+ "grad_norm": 0.31580448150634766,
365
+ "learning_rate": 1.6986008468871783e-05,
366
+ "loss": 0.6973,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 1.9078796232483346,
371
+ "grad_norm": 0.3217413127422333,
372
+ "learning_rate": 1.6032917553853936e-05,
373
+ "loss": 0.6929,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 1.9446358832988744,
378
+ "grad_norm": 0.31961649656295776,
379
+ "learning_rate": 1.509455953174948e-05,
380
+ "loss": 0.6886,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 1.9813921433494142,
385
+ "grad_norm": 0.32721585035324097,
386
+ "learning_rate": 1.4172476122324806e-05,
387
+ "loss": 0.6806,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 2.014702504020216,
392
+ "grad_norm": 0.513198733329773,
393
+ "learning_rate": 1.3268182306200405e-05,
394
+ "loss": 0.6179,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 2.051458764070756,
399
+ "grad_norm": 0.3842564523220062,
400
+ "learning_rate": 1.2383163835741692e-05,
401
+ "loss": 0.5208,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 2.0882150241212956,
406
+ "grad_norm": 0.3991941809654236,
407
+ "learning_rate": 1.1518874793971754e-05,
408
+ "loss": 0.5122,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 2.1249712841718353,
413
+ "grad_norm": 0.3792611062526703,
414
+ "learning_rate": 1.0676735205516788e-05,
415
+ "loss": 0.4966,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 2.1617275442223756,
420
+ "grad_norm": 0.41136300563812256,
421
+ "learning_rate": 9.858128703509348e-06,
422
+ "loss": 0.5039,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 2.1984838042729153,
427
+ "grad_norm": 0.40203046798706055,
428
+ "learning_rate": 9.064400256282757e-06,
429
+ "loss": 0.5089,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 2.235240064323455,
434
+ "grad_norm": 0.4289482533931732,
435
+ "learning_rate": 8.296853957591638e-06,
436
+ "loss": 0.4895,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 2.271996324373995,
441
+ "grad_norm": 0.4751777946949005,
442
+ "learning_rate": 7.556750883989324e-06,
443
+ "loss": 0.4958,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 2.3087525844245347,
448
+ "grad_norm": 0.39896517992019653,
449
+ "learning_rate": 6.845307022882405e-06,
450
+ "loss": 0.4994,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 2.3455088444750745,
455
+ "grad_norm": 0.40157681703567505,
456
+ "learning_rate": 6.1636912746665695e-06,
457
+ "loss": 0.5074,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 2.3822651045256147,
462
+ "grad_norm": 0.3953118622303009,
463
+ "learning_rate": 5.5130235322263634e-06,
464
+ "loss": 0.4776,
465
+ "step": 650
466
+ }
467
+ ],
468
+ "logging_steps": 10,
469
+ "max_steps": 816,
470
+ "num_input_tokens_seen": 0,
471
+ "num_train_epochs": 3,
472
+ "save_steps": 50,
473
+ "stateful_callbacks": {
474
+ "TrainerControl": {
475
+ "args": {
476
+ "should_epoch_stop": false,
477
+ "should_evaluate": false,
478
+ "should_log": false,
479
+ "should_save": true,
480
+ "should_training_stop": false
481
+ },
482
+ "attributes": {}
483
+ }
484
+ },
485
+ "total_flos": 1.616348144730112e+16,
486
+ "train_batch_size": 1,
487
+ "trial_name": null,
488
+ "trial_params": null
489
+ }