Dnq2025 commited on
Commit
7088e5b
·
verified ·
1 Parent(s): e78d99a

Delete trainer_state.json

Browse files
Files changed (1) hide show
  1. trainer_state.json +0 -578
trainer_state.json DELETED
@@ -1,578 +0,0 @@
1
- {
2
- "best_global_step": null,
3
- "best_metric": null,
4
- "best_model_checkpoint": null,
5
- "epoch": 34.0,
6
- "eval_steps": 132,
7
- "global_step": 4488,
8
- "is_hyper_param_search": false,
9
- "is_local_process_zero": true,
10
- "is_world_process_zero": true,
11
- "log_history": [
12
- {
13
- "epoch": 1.0,
14
- "grad_norm": 49.48432159423828,
15
- "learning_rate": 0.00039166875,
16
- "loss": 57.3675,
17
- "step": 132
18
- },
19
- {
20
- "epoch": 1.0,
21
- "eval_loss": 40.150394439697266,
22
- "eval_mean_iou": 0.3200031710526013,
23
- "eval_runtime": 16.3314,
24
- "eval_samples_per_second": 0.735,
25
- "eval_steps_per_second": 0.184,
26
- "step": 132
27
- },
28
- {
29
- "epoch": 2.0,
30
- "grad_norm": 52.09519577026367,
31
- "learning_rate": 0.00038333750000000004,
32
- "loss": 34.9978,
33
- "step": 264
34
- },
35
- {
36
- "epoch": 2.0,
37
- "eval_loss": 36.060611724853516,
38
- "eval_mean_iou": 0.3679642086623685,
39
- "eval_runtime": 12.7008,
40
- "eval_samples_per_second": 0.945,
41
- "eval_steps_per_second": 0.236,
42
- "step": 264
43
- },
44
- {
45
- "epoch": 3.0,
46
- "grad_norm": 57.55870056152344,
47
- "learning_rate": 0.00037500625,
48
- "loss": 30.9655,
49
- "step": 396
50
- },
51
- {
52
- "epoch": 3.0,
53
- "eval_loss": 35.36885452270508,
54
- "eval_mean_iou": 0.3827357146516952,
55
- "eval_runtime": 12.8032,
56
- "eval_samples_per_second": 0.937,
57
- "eval_steps_per_second": 0.234,
58
- "step": 396
59
- },
60
- {
61
- "epoch": 4.0,
62
- "grad_norm": 47.775787353515625,
63
- "learning_rate": 0.000366675,
64
- "loss": 28.4843,
65
- "step": 528
66
- },
67
- {
68
- "epoch": 4.0,
69
- "eval_loss": 37.1205940246582,
70
- "eval_mean_iou": 0.37594765764226984,
71
- "eval_runtime": 12.7782,
72
- "eval_samples_per_second": 0.939,
73
- "eval_steps_per_second": 0.235,
74
- "step": 528
75
- },
76
- {
77
- "epoch": 5.0,
78
- "grad_norm": 51.64073181152344,
79
- "learning_rate": 0.00035834375000000005,
80
- "loss": 26.3456,
81
- "step": 660
82
- },
83
- {
84
- "epoch": 5.0,
85
- "eval_loss": 37.94709396362305,
86
- "eval_mean_iou": 0.3721762507593381,
87
- "eval_runtime": 12.7428,
88
- "eval_samples_per_second": 0.942,
89
- "eval_steps_per_second": 0.235,
90
- "step": 660
91
- },
92
- {
93
- "epoch": 6.0,
94
- "grad_norm": 58.40945053100586,
95
- "learning_rate": 0.00035001250000000003,
96
- "loss": 24.7825,
97
- "step": 792
98
- },
99
- {
100
- "epoch": 6.0,
101
- "eval_loss": 38.0927619934082,
102
- "eval_mean_iou": 0.37459799925413706,
103
- "eval_runtime": 12.7865,
104
- "eval_samples_per_second": 0.938,
105
- "eval_steps_per_second": 0.235,
106
- "step": 792
107
- },
108
- {
109
- "epoch": 7.0,
110
- "grad_norm": 116.14742279052734,
111
- "learning_rate": 0.00034168125,
112
- "loss": 23.5259,
113
- "step": 924
114
- },
115
- {
116
- "epoch": 7.0,
117
- "eval_loss": 40.069950103759766,
118
- "eval_mean_iou": 0.3704415715125715,
119
- "eval_runtime": 12.8246,
120
- "eval_samples_per_second": 0.936,
121
- "eval_steps_per_second": 0.234,
122
- "step": 924
123
- },
124
- {
125
- "epoch": 8.0,
126
- "grad_norm": 109.31732940673828,
127
- "learning_rate": 0.00033335000000000005,
128
- "loss": 22.6179,
129
- "step": 1056
130
- },
131
- {
132
- "epoch": 8.0,
133
- "eval_loss": 38.507755279541016,
134
- "eval_mean_iou": 0.3908703440162653,
135
- "eval_runtime": 12.8138,
136
- "eval_samples_per_second": 0.936,
137
- "eval_steps_per_second": 0.234,
138
- "step": 1056
139
- },
140
- {
141
- "epoch": 9.0,
142
- "grad_norm": 59.890769958496094,
143
- "learning_rate": 0.00032501875000000004,
144
- "loss": 21.9044,
145
- "step": 1188
146
- },
147
- {
148
- "epoch": 9.0,
149
- "eval_loss": 37.51198959350586,
150
- "eval_mean_iou": 0.4011024752989662,
151
- "eval_runtime": 12.838,
152
- "eval_samples_per_second": 0.935,
153
- "eval_steps_per_second": 0.234,
154
- "step": 1188
155
- },
156
- {
157
- "epoch": 10.0,
158
- "grad_norm": 37.51155471801758,
159
- "learning_rate": 0.0003166875,
160
- "loss": 20.6196,
161
- "step": 1320
162
- },
163
- {
164
- "epoch": 10.0,
165
- "eval_loss": 37.682220458984375,
166
- "eval_mean_iou": 0.3926819827197939,
167
- "eval_runtime": 12.7849,
168
- "eval_samples_per_second": 0.939,
169
- "eval_steps_per_second": 0.235,
170
- "step": 1320
171
- },
172
- {
173
- "epoch": 11.0,
174
- "grad_norm": 43.30808639526367,
175
- "learning_rate": 0.00030835625,
176
- "loss": 19.916,
177
- "step": 1452
178
- },
179
- {
180
- "epoch": 11.0,
181
- "eval_loss": 37.70747756958008,
182
- "eval_mean_iou": 0.3969586400808481,
183
- "eval_runtime": 12.7919,
184
- "eval_samples_per_second": 0.938,
185
- "eval_steps_per_second": 0.235,
186
- "step": 1452
187
- },
188
- {
189
- "epoch": 12.0,
190
- "grad_norm": 36.449153900146484,
191
- "learning_rate": 0.00030002500000000004,
192
- "loss": 19.2906,
193
- "step": 1584
194
- },
195
- {
196
- "epoch": 12.0,
197
- "eval_loss": 37.13621139526367,
198
- "eval_mean_iou": 0.4035316810136001,
199
- "eval_runtime": 12.8017,
200
- "eval_samples_per_second": 0.937,
201
- "eval_steps_per_second": 0.234,
202
- "step": 1584
203
- },
204
- {
205
- "epoch": 13.0,
206
- "grad_norm": 46.17863845825195,
207
- "learning_rate": 0.00029169375,
208
- "loss": 18.8528,
209
- "step": 1716
210
- },
211
- {
212
- "epoch": 13.0,
213
- "eval_loss": 37.92732620239258,
214
- "eval_mean_iou": 0.39573059859578047,
215
- "eval_runtime": 16.1913,
216
- "eval_samples_per_second": 0.741,
217
- "eval_steps_per_second": 0.185,
218
- "step": 1716
219
- },
220
- {
221
- "epoch": 14.0,
222
- "grad_norm": 48.318145751953125,
223
- "learning_rate": 0.0002833625,
224
- "loss": 18.0426,
225
- "step": 1848
226
- },
227
- {
228
- "epoch": 14.0,
229
- "eval_loss": 40.276554107666016,
230
- "eval_mean_iou": 0.40017691692734975,
231
- "eval_runtime": 13.0658,
232
- "eval_samples_per_second": 0.918,
233
- "eval_steps_per_second": 0.23,
234
- "step": 1848
235
- },
236
- {
237
- "epoch": 15.0,
238
- "grad_norm": 36.962890625,
239
- "learning_rate": 0.00027503125,
240
- "loss": 17.5236,
241
- "step": 1980
242
- },
243
- {
244
- "epoch": 15.0,
245
- "eval_loss": 40.92256546020508,
246
- "eval_mean_iou": 0.3867004309900653,
247
- "eval_runtime": 13.0757,
248
- "eval_samples_per_second": 0.918,
249
- "eval_steps_per_second": 0.229,
250
- "step": 1980
251
- },
252
- {
253
- "epoch": 16.0,
254
- "grad_norm": 43.85531234741211,
255
- "learning_rate": 0.00026670000000000003,
256
- "loss": 17.2151,
257
- "step": 2112
258
- },
259
- {
260
- "epoch": 16.0,
261
- "eval_loss": 38.30216979980469,
262
- "eval_mean_iou": 0.41747441141369057,
263
- "eval_runtime": 12.947,
264
- "eval_samples_per_second": 0.927,
265
- "eval_steps_per_second": 0.232,
266
- "step": 2112
267
- },
268
- {
269
- "epoch": 17.0,
270
- "grad_norm": 21.20749855041504,
271
- "learning_rate": 0.00025836874999999996,
272
- "loss": 16.7026,
273
- "step": 2244
274
- },
275
- {
276
- "epoch": 17.0,
277
- "eval_loss": 41.529815673828125,
278
- "eval_mean_iou": 0.3793993317304252,
279
- "eval_runtime": 12.7864,
280
- "eval_samples_per_second": 0.938,
281
- "eval_steps_per_second": 0.235,
282
- "step": 2244
283
- },
284
- {
285
- "epoch": 18.0,
286
- "grad_norm": 24.18674659729004,
287
- "learning_rate": 0.0002500375,
288
- "loss": 16.4939,
289
- "step": 2376
290
- },
291
- {
292
- "epoch": 18.0,
293
- "eval_loss": 38.936214447021484,
294
- "eval_mean_iou": 0.4038423517803273,
295
- "eval_runtime": 12.9296,
296
- "eval_samples_per_second": 0.928,
297
- "eval_steps_per_second": 0.232,
298
- "step": 2376
299
- },
300
- {
301
- "epoch": 19.0,
302
- "grad_norm": 25.2908878326416,
303
- "learning_rate": 0.00024170625,
304
- "loss": 15.9987,
305
- "step": 2508
306
- },
307
- {
308
- "epoch": 19.0,
309
- "eval_loss": 41.31668472290039,
310
- "eval_mean_iou": 0.3934187177194776,
311
- "eval_runtime": 14.4084,
312
- "eval_samples_per_second": 0.833,
313
- "eval_steps_per_second": 0.208,
314
- "step": 2508
315
- },
316
- {
317
- "epoch": 20.0,
318
- "grad_norm": 27.059040069580078,
319
- "learning_rate": 0.000233375,
320
- "loss": 15.6106,
321
- "step": 2640
322
- },
323
- {
324
- "epoch": 20.0,
325
- "eval_loss": 41.9207878112793,
326
- "eval_mean_iou": 0.39292761246836555,
327
- "eval_runtime": 12.8081,
328
- "eval_samples_per_second": 0.937,
329
- "eval_steps_per_second": 0.234,
330
- "step": 2640
331
- },
332
- {
333
- "epoch": 21.0,
334
- "grad_norm": 33.4406852722168,
335
- "learning_rate": 0.00022504375,
336
- "loss": 15.1603,
337
- "step": 2772
338
- },
339
- {
340
- "epoch": 21.0,
341
- "eval_loss": 42.850223541259766,
342
- "eval_mean_iou": 0.39354003303320445,
343
- "eval_runtime": 12.8302,
344
- "eval_samples_per_second": 0.935,
345
- "eval_steps_per_second": 0.234,
346
- "step": 2772
347
- },
348
- {
349
- "epoch": 22.0,
350
- "grad_norm": 35.471824645996094,
351
- "learning_rate": 0.00021671250000000004,
352
- "loss": 14.8704,
353
- "step": 2904
354
- },
355
- {
356
- "epoch": 22.0,
357
- "eval_loss": 43.66987228393555,
358
- "eval_mean_iou": 0.39367395854496623,
359
- "eval_runtime": 12.743,
360
- "eval_samples_per_second": 0.942,
361
- "eval_steps_per_second": 0.235,
362
- "step": 2904
363
- },
364
- {
365
- "epoch": 23.0,
366
- "grad_norm": 32.6788444519043,
367
- "learning_rate": 0.00020838124999999997,
368
- "loss": 14.6412,
369
- "step": 3036
370
- },
371
- {
372
- "epoch": 23.0,
373
- "eval_loss": 41.786964416503906,
374
- "eval_mean_iou": 0.4016933134331057,
375
- "eval_runtime": 12.8539,
376
- "eval_samples_per_second": 0.934,
377
- "eval_steps_per_second": 0.233,
378
- "step": 3036
379
- },
380
- {
381
- "epoch": 24.0,
382
- "grad_norm": 18.433950424194336,
383
- "learning_rate": 0.00020005000000000004,
384
- "loss": 14.3711,
385
- "step": 3168
386
- },
387
- {
388
- "epoch": 24.0,
389
- "eval_loss": 42.654090881347656,
390
- "eval_mean_iou": 0.4053969211518879,
391
- "eval_runtime": 12.8806,
392
- "eval_samples_per_second": 0.932,
393
- "eval_steps_per_second": 0.233,
394
- "step": 3168
395
- },
396
- {
397
- "epoch": 25.0,
398
- "grad_norm": 19.108030319213867,
399
- "learning_rate": 0.00019171875,
400
- "loss": 14.042,
401
- "step": 3300
402
- },
403
- {
404
- "epoch": 25.0,
405
- "eval_loss": 41.77846145629883,
406
- "eval_mean_iou": 0.40953756424959514,
407
- "eval_runtime": 12.8253,
408
- "eval_samples_per_second": 0.936,
409
- "eval_steps_per_second": 0.234,
410
- "step": 3300
411
- },
412
- {
413
- "epoch": 26.0,
414
- "grad_norm": 44.6916618347168,
415
- "learning_rate": 0.00018338750000000003,
416
- "loss": 13.7427,
417
- "step": 3432
418
- },
419
- {
420
- "epoch": 26.0,
421
- "eval_loss": 42.78445053100586,
422
- "eval_mean_iou": 0.39490494739631005,
423
- "eval_runtime": 12.8245,
424
- "eval_samples_per_second": 0.936,
425
- "eval_steps_per_second": 0.234,
426
- "step": 3432
427
- },
428
- {
429
- "epoch": 27.0,
430
- "grad_norm": 11.862845420837402,
431
- "learning_rate": 0.00017505625000000002,
432
- "loss": 13.4682,
433
- "step": 3564
434
- },
435
- {
436
- "epoch": 27.0,
437
- "eval_loss": 39.967960357666016,
438
- "eval_mean_iou": 0.40437776542436404,
439
- "eval_runtime": 12.8842,
440
- "eval_samples_per_second": 0.931,
441
- "eval_steps_per_second": 0.233,
442
- "step": 3564
443
- },
444
- {
445
- "epoch": 28.0,
446
- "grad_norm": 16.17394256591797,
447
- "learning_rate": 0.000166725,
448
- "loss": 13.2038,
449
- "step": 3696
450
- },
451
- {
452
- "epoch": 28.0,
453
- "eval_loss": 44.32680130004883,
454
- "eval_mean_iou": 0.40446819377308363,
455
- "eval_runtime": 12.8407,
456
- "eval_samples_per_second": 0.935,
457
- "eval_steps_per_second": 0.234,
458
- "step": 3696
459
- },
460
- {
461
- "epoch": 29.0,
462
- "grad_norm": 15.816814422607422,
463
- "learning_rate": 0.00015839375,
464
- "loss": 12.9866,
465
- "step": 3828
466
- },
467
- {
468
- "epoch": 29.0,
469
- "eval_loss": 42.994937896728516,
470
- "eval_mean_iou": 0.4172571472732856,
471
- "eval_runtime": 12.7725,
472
- "eval_samples_per_second": 0.94,
473
- "eval_steps_per_second": 0.235,
474
- "step": 3828
475
- },
476
- {
477
- "epoch": 30.0,
478
- "grad_norm": 9.294917106628418,
479
- "learning_rate": 0.00015006250000000002,
480
- "loss": 12.677,
481
- "step": 3960
482
- },
483
- {
484
- "epoch": 30.0,
485
- "eval_loss": 45.5428352355957,
486
- "eval_mean_iou": 0.3972872089864519,
487
- "eval_runtime": 12.8434,
488
- "eval_samples_per_second": 0.934,
489
- "eval_steps_per_second": 0.234,
490
- "step": 3960
491
- },
492
- {
493
- "epoch": 31.0,
494
- "grad_norm": 7.88139009475708,
495
- "learning_rate": 0.00014173125,
496
- "loss": 12.5325,
497
- "step": 4092
498
- },
499
- {
500
- "epoch": 31.0,
501
- "eval_loss": 47.610660552978516,
502
- "eval_mean_iou": 0.38300735666600355,
503
- "eval_runtime": 12.7762,
504
- "eval_samples_per_second": 0.939,
505
- "eval_steps_per_second": 0.235,
506
- "step": 4092
507
- },
508
- {
509
- "epoch": 32.0,
510
- "grad_norm": 9.520607948303223,
511
- "learning_rate": 0.00013340000000000002,
512
- "loss": 12.2756,
513
- "step": 4224
514
- },
515
- {
516
- "epoch": 32.0,
517
- "eval_loss": 46.27323913574219,
518
- "eval_mean_iou": 0.408539339421579,
519
- "eval_runtime": 12.8752,
520
- "eval_samples_per_second": 0.932,
521
- "eval_steps_per_second": 0.233,
522
- "step": 4224
523
- },
524
- {
525
- "epoch": 33.0,
526
- "grad_norm": 15.029850006103516,
527
- "learning_rate": 0.00012506875,
528
- "loss": 12.1823,
529
- "step": 4356
530
- },
531
- {
532
- "epoch": 33.0,
533
- "eval_loss": 46.04867935180664,
534
- "eval_mean_iou": 0.39786737119256266,
535
- "eval_runtime": 12.8881,
536
- "eval_samples_per_second": 0.931,
537
- "eval_steps_per_second": 0.233,
538
- "step": 4356
539
- },
540
- {
541
- "epoch": 34.0,
542
- "grad_norm": 12.38525104522705,
543
- "learning_rate": 0.00011673749999999998,
544
- "loss": 11.9533,
545
- "step": 4488
546
- },
547
- {
548
- "epoch": 34.0,
549
- "eval_loss": 45.46687316894531,
550
- "eval_mean_iou": 0.41451732772514854,
551
- "eval_runtime": 12.8782,
552
- "eval_samples_per_second": 0.932,
553
- "eval_steps_per_second": 0.233,
554
- "step": 4488
555
- }
556
- ],
557
- "logging_steps": 132,
558
- "max_steps": 6336,
559
- "num_input_tokens_seen": 0,
560
- "num_train_epochs": 48,
561
- "save_steps": 132,
562
- "stateful_callbacks": {
563
- "TrainerControl": {
564
- "args": {
565
- "should_epoch_stop": false,
566
- "should_evaluate": false,
567
- "should_log": false,
568
- "should_save": true,
569
- "should_training_stop": false
570
- },
571
- "attributes": {}
572
- }
573
- },
574
- "total_flos": 1.026346797477082e+19,
575
- "train_batch_size": 4,
576
- "trial_name": null,
577
- "trial_params": null
578
- }