Hemgg commited on
Commit
f7c4cb3
·
verified ·
1 Parent(s): 61332a5

Upload 8 files

Browse files
Files changed (4) hide show
  1. optimizer.pt +3 -0
  2. rng_state.pth +3 -0
  3. scheduler.pt +3 -0
  4. trainer_state.json +528 -0
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a13d3988bf45e6c42003fdbc6879518055b7ebe3b8f7f228e9fdec755a6c538
3
+ size 689988619
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a794a6cb9cd4bbd0c53d08db0e20a5536c789bba6f22113385c1c408d58908bd
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9816312e27757c04d02f3a638ca8df7a1931469f11e27684e6ff99d7e823878
3
+ size 1465
trainer_state.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 686,
3
+ "best_metric": 0.893740902474527,
4
+ "best_model_checkpoint": "deepfake_model_Video-MAE-1/checkpoint-686",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 686,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.014577259475218658,
14
+ "grad_norm": 5.3146071434021,
15
+ "learning_rate": 2.608695652173913e-05,
16
+ "loss": 0.6438,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.029154518950437316,
21
+ "grad_norm": 0.1986074000597,
22
+ "learning_rate": 5.507246376811594e-05,
23
+ "loss": 0.4447,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.043731778425655975,
28
+ "grad_norm": 0.06804976612329483,
29
+ "learning_rate": 8.405797101449276e-05,
30
+ "loss": 0.2479,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.05830903790087463,
35
+ "grad_norm": 4.464612007141113,
36
+ "learning_rate": 0.00011304347826086956,
37
+ "loss": 0.7579,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.0728862973760933,
42
+ "grad_norm": 0.01860504038631916,
43
+ "learning_rate": 0.00014202898550724638,
44
+ "loss": 0.2154,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.08746355685131195,
49
+ "grad_norm": 0.0755125880241394,
50
+ "learning_rate": 0.0001710144927536232,
51
+ "loss": 0.3242,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.10204081632653061,
56
+ "grad_norm": 0.4104626476764679,
57
+ "learning_rate": 0.0002,
58
+ "loss": 0.2618,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.11661807580174927,
63
+ "grad_norm": 5.070664882659912,
64
+ "learning_rate": 0.0001967585089141005,
65
+ "loss": 0.4574,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.13119533527696792,
70
+ "grad_norm": 2.585449695587158,
71
+ "learning_rate": 0.00019351701782820098,
72
+ "loss": 0.4542,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.1457725947521866,
77
+ "grad_norm": 0.9481808543205261,
78
+ "learning_rate": 0.00019027552674230147,
79
+ "loss": 0.3373,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.16034985422740525,
84
+ "grad_norm": 0.43055403232574463,
85
+ "learning_rate": 0.00018703403565640197,
86
+ "loss": 0.3928,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.1749271137026239,
91
+ "grad_norm": 0.15950709581375122,
92
+ "learning_rate": 0.00018379254457050244,
93
+ "loss": 0.1172,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.18950437317784258,
98
+ "grad_norm": 8.764716148376465,
99
+ "learning_rate": 0.00018055105348460294,
100
+ "loss": 0.4847,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.20408163265306123,
105
+ "grad_norm": 0.3541601300239563,
106
+ "learning_rate": 0.00017730956239870343,
107
+ "loss": 0.345,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.21865889212827988,
112
+ "grad_norm": 2.651851177215576,
113
+ "learning_rate": 0.0001740680713128039,
114
+ "loss": 0.6826,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.23323615160349853,
119
+ "grad_norm": 0.5931745171546936,
120
+ "learning_rate": 0.0001708265802269044,
121
+ "loss": 0.2021,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.2478134110787172,
126
+ "grad_norm": 1.8013778924942017,
127
+ "learning_rate": 0.00016758508914100487,
128
+ "loss": 0.7516,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.26239067055393583,
133
+ "grad_norm": 1.0193092823028564,
134
+ "learning_rate": 0.00016434359805510534,
135
+ "loss": 0.405,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.27696793002915454,
140
+ "grad_norm": 1.0689607858657837,
141
+ "learning_rate": 0.00016110210696920584,
142
+ "loss": 0.2898,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.2915451895043732,
147
+ "grad_norm": 3.772040605545044,
148
+ "learning_rate": 0.0001578606158833063,
149
+ "loss": 0.5378,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.30612244897959184,
154
+ "grad_norm": 0.8388488292694092,
155
+ "learning_rate": 0.0001546191247974068,
156
+ "loss": 0.4129,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.3206997084548105,
161
+ "grad_norm": 0.44900116324424744,
162
+ "learning_rate": 0.0001513776337115073,
163
+ "loss": 0.2113,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.33527696793002915,
168
+ "grad_norm": 0.29753178358078003,
169
+ "learning_rate": 0.00014813614262560777,
170
+ "loss": 0.4854,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.3498542274052478,
175
+ "grad_norm": 2.6392834186553955,
176
+ "learning_rate": 0.00014489465153970827,
177
+ "loss": 0.5591,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.36443148688046645,
182
+ "grad_norm": 0.7161630988121033,
183
+ "learning_rate": 0.00014165316045380876,
184
+ "loss": 0.3698,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.37900874635568516,
189
+ "grad_norm": 0.4423307180404663,
190
+ "learning_rate": 0.00013841166936790923,
191
+ "loss": 0.4399,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.3935860058309038,
196
+ "grad_norm": 2.786557197570801,
197
+ "learning_rate": 0.00013517017828200973,
198
+ "loss": 0.3868,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.40816326530612246,
203
+ "grad_norm": 2.5099093914031982,
204
+ "learning_rate": 0.00013192868719611023,
205
+ "loss": 0.4658,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.4227405247813411,
210
+ "grad_norm": 1.9165363311767578,
211
+ "learning_rate": 0.0001286871961102107,
212
+ "loss": 0.8003,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.43731778425655976,
217
+ "grad_norm": 1.3797115087509155,
218
+ "learning_rate": 0.0001254457050243112,
219
+ "loss": 0.3938,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.4518950437317784,
224
+ "grad_norm": 1.7510792016983032,
225
+ "learning_rate": 0.0001222042139384117,
226
+ "loss": 0.3926,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.46647230320699706,
231
+ "grad_norm": 1.488336205482483,
232
+ "learning_rate": 0.00011896272285251216,
233
+ "loss": 0.504,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.48104956268221577,
238
+ "grad_norm": 0.9835926294326782,
239
+ "learning_rate": 0.00011572123176661264,
240
+ "loss": 0.2206,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.4956268221574344,
245
+ "grad_norm": 0.7870058417320251,
246
+ "learning_rate": 0.00011247974068071314,
247
+ "loss": 0.3236,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.5102040816326531,
252
+ "grad_norm": 0.633097767829895,
253
+ "learning_rate": 0.00010923824959481361,
254
+ "loss": 0.4667,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.5247813411078717,
259
+ "grad_norm": 2.2080698013305664,
260
+ "learning_rate": 0.0001059967585089141,
261
+ "loss": 0.3491,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.5393586005830904,
266
+ "grad_norm": 0.6938947439193726,
267
+ "learning_rate": 0.0001027552674230146,
268
+ "loss": 0.497,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.5539358600583091,
273
+ "grad_norm": 0.47899454832077026,
274
+ "learning_rate": 9.951377633711507e-05,
275
+ "loss": 0.0584,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.5685131195335277,
280
+ "grad_norm": 0.3724881410598755,
281
+ "learning_rate": 9.627228525121557e-05,
282
+ "loss": 0.3644,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.5830903790087464,
287
+ "grad_norm": 0.5075926184654236,
288
+ "learning_rate": 9.303079416531605e-05,
289
+ "loss": 0.697,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.597667638483965,
294
+ "grad_norm": 0.6416959166526794,
295
+ "learning_rate": 8.978930307941653e-05,
296
+ "loss": 0.2031,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.6122448979591837,
301
+ "grad_norm": 0.49635839462280273,
302
+ "learning_rate": 8.654781199351703e-05,
303
+ "loss": 0.3355,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.6268221574344023,
308
+ "grad_norm": 0.43995216488838196,
309
+ "learning_rate": 8.330632090761751e-05,
310
+ "loss": 0.3582,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.641399416909621,
315
+ "grad_norm": 4.814486980438232,
316
+ "learning_rate": 8.006482982171798e-05,
317
+ "loss": 0.51,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.6559766763848397,
322
+ "grad_norm": 1.9786182641983032,
323
+ "learning_rate": 7.682333873581848e-05,
324
+ "loss": 0.486,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.6705539358600583,
329
+ "grad_norm": 1.8584587574005127,
330
+ "learning_rate": 7.358184764991896e-05,
331
+ "loss": 0.3977,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.685131195335277,
336
+ "grad_norm": 0.7450128793716431,
337
+ "learning_rate": 7.034035656401945e-05,
338
+ "loss": 0.2681,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.6997084548104956,
343
+ "grad_norm": 0.6749535202980042,
344
+ "learning_rate": 6.709886547811994e-05,
345
+ "loss": 0.2679,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.7142857142857143,
350
+ "grad_norm": 0.7746213674545288,
351
+ "learning_rate": 6.385737439222043e-05,
352
+ "loss": 0.6622,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.7288629737609329,
357
+ "grad_norm": 1.8850953578948975,
358
+ "learning_rate": 6.061588330632091e-05,
359
+ "loss": 0.2107,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.7434402332361516,
364
+ "grad_norm": 0.5546265840530396,
365
+ "learning_rate": 5.7374392220421394e-05,
366
+ "loss": 0.2717,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.7580174927113703,
371
+ "grad_norm": 0.39889729022979736,
372
+ "learning_rate": 5.4132901134521884e-05,
373
+ "loss": 0.1229,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.7725947521865889,
378
+ "grad_norm": 0.3102359175682068,
379
+ "learning_rate": 5.089141004862237e-05,
380
+ "loss": 0.2076,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.7871720116618076,
385
+ "grad_norm": 2.644927740097046,
386
+ "learning_rate": 4.764991896272286e-05,
387
+ "loss": 0.3011,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.8017492711370262,
392
+ "grad_norm": 0.3060024678707123,
393
+ "learning_rate": 4.440842787682334e-05,
394
+ "loss": 0.3004,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.8163265306122449,
399
+ "grad_norm": 0.3062104284763336,
400
+ "learning_rate": 4.1166936790923824e-05,
401
+ "loss": 0.1175,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.8309037900874635,
406
+ "grad_norm": 0.3254077434539795,
407
+ "learning_rate": 3.7925445705024314e-05,
408
+ "loss": 0.6553,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.8454810495626822,
413
+ "grad_norm": 5.185217380523682,
414
+ "learning_rate": 3.46839546191248e-05,
415
+ "loss": 0.7195,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.8600583090379009,
420
+ "grad_norm": 2.225529193878174,
421
+ "learning_rate": 3.144246353322529e-05,
422
+ "loss": 0.1991,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.8746355685131195,
427
+ "grad_norm": 0.4137911796569824,
428
+ "learning_rate": 2.820097244732577e-05,
429
+ "loss": 0.2816,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.8892128279883382,
434
+ "grad_norm": 0.3949086368083954,
435
+ "learning_rate": 2.4959481361426257e-05,
436
+ "loss": 0.2029,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.9037900874635568,
441
+ "grad_norm": 2.37062668800354,
442
+ "learning_rate": 2.1717990275526743e-05,
443
+ "loss": 0.609,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.9183673469387755,
448
+ "grad_norm": 2.082977771759033,
449
+ "learning_rate": 1.847649918962723e-05,
450
+ "loss": 1.0544,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.9329446064139941,
455
+ "grad_norm": 0.516295313835144,
456
+ "learning_rate": 1.5235008103727717e-05,
457
+ "loss": 0.0501,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.9475218658892128,
462
+ "grad_norm": 0.4977923035621643,
463
+ "learning_rate": 1.1993517017828202e-05,
464
+ "loss": 0.198,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.9620991253644315,
469
+ "grad_norm": 0.48140740394592285,
470
+ "learning_rate": 8.752025931928688e-06,
471
+ "loss": 0.4284,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.9766763848396501,
476
+ "grad_norm": 0.4845019280910492,
477
+ "learning_rate": 5.510534846029174e-06,
478
+ "loss": 0.2735,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.9912536443148688,
483
+ "grad_norm": 2.2184746265411377,
484
+ "learning_rate": 2.2690437601296597e-06,
485
+ "loss": 0.504,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 1.0,
490
+ "eval_accuracy": 0.893740902474527,
491
+ "eval_loss": 0.36989203095436096,
492
+ "eval_runtime": 401.8752,
493
+ "eval_samples_per_second": 1.709,
494
+ "eval_steps_per_second": 0.428,
495
+ "step": 686
496
+ }
497
+ ],
498
+ "logging_steps": 10,
499
+ "max_steps": 686,
500
+ "num_input_tokens_seen": 0,
501
+ "num_train_epochs": 1,
502
+ "save_steps": 500,
503
+ "stateful_callbacks": {
504
+ "EarlyStoppingCallback": {
505
+ "args": {
506
+ "early_stopping_patience": 1,
507
+ "early_stopping_threshold": 0.0
508
+ },
509
+ "attributes": {
510
+ "early_stopping_patience_counter": 0
511
+ }
512
+ },
513
+ "TrainerControl": {
514
+ "args": {
515
+ "should_epoch_stop": false,
516
+ "should_evaluate": false,
517
+ "should_log": false,
518
+ "should_save": true,
519
+ "should_training_stop": true
520
+ },
521
+ "attributes": {}
522
+ }
523
+ },
524
+ "total_flos": 3.4192012335104655e+18,
525
+ "train_batch_size": 4,
526
+ "trial_name": null,
527
+ "trial_params": null
528
+ }