Weili commited on
Commit
792a931
1 Parent(s): 4a26b1a

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9876,
4
+ "eval_loss": 0.04271771386265755,
5
+ "eval_runtime": 148.5576,
6
+ "eval_samples_per_second": 67.314,
7
+ "eval_steps_per_second": 2.107,
8
+ "total_flos": 1.161843208194687e+19,
9
+ "train_loss": 0.3334213079550327,
10
+ "train_runtime": 6069.2812,
11
+ "train_samples_per_second": 24.715,
12
+ "train_steps_per_second": 0.193
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9876,
4
+ "eval_loss": 0.04271771386265755,
5
+ "eval_runtime": 148.5576,
6
+ "eval_samples_per_second": 67.314,
7
+ "eval_steps_per_second": 2.107
8
+ }
runs/Dec07_00-51-57_20578b4f8f65/events.out.tfevents.1670380916.20578b4f8f65.74.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad350d3999cc63295063a573d75fcca5aa47c0100aad88109dab2fb16c296d47
3
+ size 363
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 1.161843208194687e+19,
4
+ "train_loss": 0.3334213079550327,
5
+ "train_runtime": 6069.2812,
6
+ "train_samples_per_second": 24.715,
7
+ "train_steps_per_second": 0.193
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,754 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9876,
3
+ "best_model_checkpoint": "vit-base-patch16-224-finetuned-cifar10/checkpoint-1170",
4
+ "epoch": 2.998080614203455,
5
+ "global_step": 1170,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 4.273504273504274e-06,
13
+ "loss": 2.389,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 8.547008547008548e-06,
19
+ "loss": 2.2927,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.08,
24
+ "learning_rate": 1.282051282051282e-05,
25
+ "loss": 2.1326,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.1,
30
+ "learning_rate": 1.7094017094017095e-05,
31
+ "loss": 1.9038,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.13,
36
+ "learning_rate": 2.1367521367521368e-05,
37
+ "loss": 1.5749,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.15,
42
+ "learning_rate": 2.564102564102564e-05,
43
+ "loss": 1.2323,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.18,
48
+ "learning_rate": 2.9914529914529915e-05,
49
+ "loss": 0.9157,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.2,
54
+ "learning_rate": 3.418803418803419e-05,
55
+ "loss": 0.6675,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.23,
60
+ "learning_rate": 3.846153846153846e-05,
61
+ "loss": 0.5038,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.26,
66
+ "learning_rate": 4.2735042735042735e-05,
67
+ "loss": 0.4423,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.28,
72
+ "learning_rate": 4.700854700854701e-05,
73
+ "loss": 0.3788,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.31,
78
+ "learning_rate": 4.985754985754986e-05,
79
+ "loss": 0.3663,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.33,
84
+ "learning_rate": 4.938271604938271e-05,
85
+ "loss": 0.3431,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.36,
90
+ "learning_rate": 4.890788224121557e-05,
91
+ "loss": 0.3941,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.38,
96
+ "learning_rate": 4.8433048433048433e-05,
97
+ "loss": 0.3056,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.41,
102
+ "learning_rate": 4.7958214624881294e-05,
103
+ "loss": 0.2976,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.44,
108
+ "learning_rate": 4.7483380816714154e-05,
109
+ "loss": 0.3306,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.46,
114
+ "learning_rate": 4.700854700854701e-05,
115
+ "loss": 0.2931,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.49,
120
+ "learning_rate": 4.653371320037987e-05,
121
+ "loss": 0.3375,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.51,
126
+ "learning_rate": 4.605887939221273e-05,
127
+ "loss": 0.3069,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.54,
132
+ "learning_rate": 4.558404558404559e-05,
133
+ "loss": 0.3015,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.56,
138
+ "learning_rate": 4.510921177587845e-05,
139
+ "loss": 0.2746,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.59,
144
+ "learning_rate": 4.463437796771131e-05,
145
+ "loss": 0.3211,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.61,
150
+ "learning_rate": 4.415954415954416e-05,
151
+ "loss": 0.2963,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.64,
156
+ "learning_rate": 4.368471035137702e-05,
157
+ "loss": 0.3042,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.67,
162
+ "learning_rate": 4.3209876543209875e-05,
163
+ "loss": 0.2472,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.69,
168
+ "learning_rate": 4.2735042735042735e-05,
169
+ "loss": 0.2699,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.72,
174
+ "learning_rate": 4.2260208926875595e-05,
175
+ "loss": 0.2684,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.74,
180
+ "learning_rate": 4.1785375118708455e-05,
181
+ "loss": 0.3115,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.77,
186
+ "learning_rate": 4.131054131054131e-05,
187
+ "loss": 0.2885,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.79,
192
+ "learning_rate": 4.083570750237417e-05,
193
+ "loss": 0.2865,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.82,
198
+ "learning_rate": 4.036087369420703e-05,
199
+ "loss": 0.3073,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.84,
204
+ "learning_rate": 3.988603988603989e-05,
205
+ "loss": 0.2608,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.87,
210
+ "learning_rate": 3.941120607787275e-05,
211
+ "loss": 0.2675,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.9,
216
+ "learning_rate": 3.893637226970561e-05,
217
+ "loss": 0.2603,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.92,
222
+ "learning_rate": 3.846153846153846e-05,
223
+ "loss": 0.2564,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.95,
228
+ "learning_rate": 3.798670465337132e-05,
229
+ "loss": 0.2582,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.97,
234
+ "learning_rate": 3.7511870845204176e-05,
235
+ "loss": 0.2916,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 1.0,
240
+ "learning_rate": 3.7037037037037037e-05,
241
+ "loss": 0.2518,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 1.0,
246
+ "eval_accuracy": 0.9821,
247
+ "eval_loss": 0.060857731848955154,
248
+ "eval_runtime": 146.1804,
249
+ "eval_samples_per_second": 68.409,
250
+ "eval_steps_per_second": 2.141,
251
+ "step": 390
252
+ },
253
+ {
254
+ "epoch": 1.03,
255
+ "learning_rate": 3.65622032288699e-05,
256
+ "loss": 0.2976,
257
+ "step": 400
258
+ },
259
+ {
260
+ "epoch": 1.05,
261
+ "learning_rate": 3.608736942070276e-05,
262
+ "loss": 0.2397,
263
+ "step": 410
264
+ },
265
+ {
266
+ "epoch": 1.08,
267
+ "learning_rate": 3.561253561253561e-05,
268
+ "loss": 0.2125,
269
+ "step": 420
270
+ },
271
+ {
272
+ "epoch": 1.1,
273
+ "learning_rate": 3.513770180436847e-05,
274
+ "loss": 0.2754,
275
+ "step": 430
276
+ },
277
+ {
278
+ "epoch": 1.13,
279
+ "learning_rate": 3.466286799620133e-05,
280
+ "loss": 0.2553,
281
+ "step": 440
282
+ },
283
+ {
284
+ "epoch": 1.15,
285
+ "learning_rate": 3.418803418803419e-05,
286
+ "loss": 0.2723,
287
+ "step": 450
288
+ },
289
+ {
290
+ "epoch": 1.18,
291
+ "learning_rate": 3.371320037986705e-05,
292
+ "loss": 0.2522,
293
+ "step": 460
294
+ },
295
+ {
296
+ "epoch": 1.2,
297
+ "learning_rate": 3.323836657169991e-05,
298
+ "loss": 0.2485,
299
+ "step": 470
300
+ },
301
+ {
302
+ "epoch": 1.23,
303
+ "learning_rate": 3.2763532763532764e-05,
304
+ "loss": 0.2449,
305
+ "step": 480
306
+ },
307
+ {
308
+ "epoch": 1.26,
309
+ "learning_rate": 3.2288698955365625e-05,
310
+ "loss": 0.2093,
311
+ "step": 490
312
+ },
313
+ {
314
+ "epoch": 1.28,
315
+ "learning_rate": 3.181386514719848e-05,
316
+ "loss": 0.2502,
317
+ "step": 500
318
+ },
319
+ {
320
+ "epoch": 1.31,
321
+ "learning_rate": 3.133903133903134e-05,
322
+ "loss": 0.2166,
323
+ "step": 510
324
+ },
325
+ {
326
+ "epoch": 1.33,
327
+ "learning_rate": 3.08641975308642e-05,
328
+ "loss": 0.219,
329
+ "step": 520
330
+ },
331
+ {
332
+ "epoch": 1.36,
333
+ "learning_rate": 3.0389363722697055e-05,
334
+ "loss": 0.2045,
335
+ "step": 530
336
+ },
337
+ {
338
+ "epoch": 1.38,
339
+ "learning_rate": 2.9914529914529915e-05,
340
+ "loss": 0.2775,
341
+ "step": 540
342
+ },
343
+ {
344
+ "epoch": 1.41,
345
+ "learning_rate": 2.9439696106362775e-05,
346
+ "loss": 0.2127,
347
+ "step": 550
348
+ },
349
+ {
350
+ "epoch": 1.44,
351
+ "learning_rate": 2.8964862298195632e-05,
352
+ "loss": 0.2148,
353
+ "step": 560
354
+ },
355
+ {
356
+ "epoch": 1.46,
357
+ "learning_rate": 2.8490028490028492e-05,
358
+ "loss": 0.2258,
359
+ "step": 570
360
+ },
361
+ {
362
+ "epoch": 1.49,
363
+ "learning_rate": 2.8015194681861352e-05,
364
+ "loss": 0.2414,
365
+ "step": 580
366
+ },
367
+ {
368
+ "epoch": 1.51,
369
+ "learning_rate": 2.754036087369421e-05,
370
+ "loss": 0.1817,
371
+ "step": 590
372
+ },
373
+ {
374
+ "epoch": 1.54,
375
+ "learning_rate": 2.706552706552707e-05,
376
+ "loss": 0.2315,
377
+ "step": 600
378
+ },
379
+ {
380
+ "epoch": 1.56,
381
+ "learning_rate": 2.6590693257359926e-05,
382
+ "loss": 0.2415,
383
+ "step": 610
384
+ },
385
+ {
386
+ "epoch": 1.59,
387
+ "learning_rate": 2.611585944919278e-05,
388
+ "loss": 0.2264,
389
+ "step": 620
390
+ },
391
+ {
392
+ "epoch": 1.61,
393
+ "learning_rate": 2.564102564102564e-05,
394
+ "loss": 0.2178,
395
+ "step": 630
396
+ },
397
+ {
398
+ "epoch": 1.64,
399
+ "learning_rate": 2.51661918328585e-05,
400
+ "loss": 0.2287,
401
+ "step": 640
402
+ },
403
+ {
404
+ "epoch": 1.67,
405
+ "learning_rate": 2.4691358024691357e-05,
406
+ "loss": 0.2046,
407
+ "step": 650
408
+ },
409
+ {
410
+ "epoch": 1.69,
411
+ "learning_rate": 2.4216524216524217e-05,
412
+ "loss": 0.2366,
413
+ "step": 660
414
+ },
415
+ {
416
+ "epoch": 1.72,
417
+ "learning_rate": 2.3741690408357077e-05,
418
+ "loss": 0.2455,
419
+ "step": 670
420
+ },
421
+ {
422
+ "epoch": 1.74,
423
+ "learning_rate": 2.3266856600189934e-05,
424
+ "loss": 0.2038,
425
+ "step": 680
426
+ },
427
+ {
428
+ "epoch": 1.77,
429
+ "learning_rate": 2.2792022792022794e-05,
430
+ "loss": 0.2258,
431
+ "step": 690
432
+ },
433
+ {
434
+ "epoch": 1.79,
435
+ "learning_rate": 2.2317188983855654e-05,
436
+ "loss": 0.2259,
437
+ "step": 700
438
+ },
439
+ {
440
+ "epoch": 1.82,
441
+ "learning_rate": 2.184235517568851e-05,
442
+ "loss": 0.2106,
443
+ "step": 710
444
+ },
445
+ {
446
+ "epoch": 1.84,
447
+ "learning_rate": 2.1367521367521368e-05,
448
+ "loss": 0.1748,
449
+ "step": 720
450
+ },
451
+ {
452
+ "epoch": 1.87,
453
+ "learning_rate": 2.0892687559354228e-05,
454
+ "loss": 0.222,
455
+ "step": 730
456
+ },
457
+ {
458
+ "epoch": 1.9,
459
+ "learning_rate": 2.0417853751187084e-05,
460
+ "loss": 0.2288,
461
+ "step": 740
462
+ },
463
+ {
464
+ "epoch": 1.92,
465
+ "learning_rate": 1.9943019943019945e-05,
466
+ "loss": 0.2094,
467
+ "step": 750
468
+ },
469
+ {
470
+ "epoch": 1.95,
471
+ "learning_rate": 1.9468186134852805e-05,
472
+ "loss": 0.2189,
473
+ "step": 760
474
+ },
475
+ {
476
+ "epoch": 1.97,
477
+ "learning_rate": 1.899335232668566e-05,
478
+ "loss": 0.198,
479
+ "step": 770
480
+ },
481
+ {
482
+ "epoch": 2.0,
483
+ "learning_rate": 1.8518518518518518e-05,
484
+ "loss": 0.1985,
485
+ "step": 780
486
+ },
487
+ {
488
+ "epoch": 2.0,
489
+ "eval_accuracy": 0.983,
490
+ "eval_loss": 0.053233180195093155,
491
+ "eval_runtime": 146.402,
492
+ "eval_samples_per_second": 68.305,
493
+ "eval_steps_per_second": 2.138,
494
+ "step": 780
495
+ },
496
+ {
497
+ "epoch": 2.03,
498
+ "learning_rate": 1.804368471035138e-05,
499
+ "loss": 0.2092,
500
+ "step": 790
501
+ },
502
+ {
503
+ "epoch": 2.05,
504
+ "learning_rate": 1.7568850902184235e-05,
505
+ "loss": 0.2125,
506
+ "step": 800
507
+ },
508
+ {
509
+ "epoch": 2.08,
510
+ "learning_rate": 1.7094017094017095e-05,
511
+ "loss": 0.1673,
512
+ "step": 810
513
+ },
514
+ {
515
+ "epoch": 2.1,
516
+ "learning_rate": 1.6619183285849956e-05,
517
+ "loss": 0.2104,
518
+ "step": 820
519
+ },
520
+ {
521
+ "epoch": 2.13,
522
+ "learning_rate": 1.6144349477682812e-05,
523
+ "loss": 0.2577,
524
+ "step": 830
525
+ },
526
+ {
527
+ "epoch": 2.15,
528
+ "learning_rate": 1.566951566951567e-05,
529
+ "loss": 0.1826,
530
+ "step": 840
531
+ },
532
+ {
533
+ "epoch": 2.18,
534
+ "learning_rate": 1.5194681861348528e-05,
535
+ "loss": 0.1934,
536
+ "step": 850
537
+ },
538
+ {
539
+ "epoch": 2.2,
540
+ "learning_rate": 1.4719848053181388e-05,
541
+ "loss": 0.2158,
542
+ "step": 860
543
+ },
544
+ {
545
+ "epoch": 2.23,
546
+ "learning_rate": 1.4245014245014246e-05,
547
+ "loss": 0.1742,
548
+ "step": 870
549
+ },
550
+ {
551
+ "epoch": 2.26,
552
+ "learning_rate": 1.3770180436847105e-05,
553
+ "loss": 0.1709,
554
+ "step": 880
555
+ },
556
+ {
557
+ "epoch": 2.28,
558
+ "learning_rate": 1.3295346628679963e-05,
559
+ "loss": 0.2186,
560
+ "step": 890
561
+ },
562
+ {
563
+ "epoch": 2.31,
564
+ "learning_rate": 1.282051282051282e-05,
565
+ "loss": 0.2159,
566
+ "step": 900
567
+ },
568
+ {
569
+ "epoch": 2.33,
570
+ "learning_rate": 1.2345679012345678e-05,
571
+ "loss": 0.1856,
572
+ "step": 910
573
+ },
574
+ {
575
+ "epoch": 2.36,
576
+ "learning_rate": 1.1870845204178538e-05,
577
+ "loss": 0.1712,
578
+ "step": 920
579
+ },
580
+ {
581
+ "epoch": 2.38,
582
+ "learning_rate": 1.1396011396011397e-05,
583
+ "loss": 0.1614,
584
+ "step": 930
585
+ },
586
+ {
587
+ "epoch": 2.41,
588
+ "learning_rate": 1.0921177587844255e-05,
589
+ "loss": 0.2024,
590
+ "step": 940
591
+ },
592
+ {
593
+ "epoch": 2.44,
594
+ "learning_rate": 1.0446343779677114e-05,
595
+ "loss": 0.1926,
596
+ "step": 950
597
+ },
598
+ {
599
+ "epoch": 2.46,
600
+ "learning_rate": 9.971509971509972e-06,
601
+ "loss": 0.2085,
602
+ "step": 960
603
+ },
604
+ {
605
+ "epoch": 2.49,
606
+ "learning_rate": 9.49667616334283e-06,
607
+ "loss": 0.18,
608
+ "step": 970
609
+ },
610
+ {
611
+ "epoch": 2.51,
612
+ "learning_rate": 9.02184235517569e-06,
613
+ "loss": 0.1745,
614
+ "step": 980
615
+ },
616
+ {
617
+ "epoch": 2.54,
618
+ "learning_rate": 8.547008547008548e-06,
619
+ "loss": 0.1818,
620
+ "step": 990
621
+ },
622
+ {
623
+ "epoch": 2.56,
624
+ "learning_rate": 8.072174738841406e-06,
625
+ "loss": 0.1867,
626
+ "step": 1000
627
+ },
628
+ {
629
+ "epoch": 2.59,
630
+ "learning_rate": 7.597340930674264e-06,
631
+ "loss": 0.1979,
632
+ "step": 1010
633
+ },
634
+ {
635
+ "epoch": 2.61,
636
+ "learning_rate": 7.122507122507123e-06,
637
+ "loss": 0.2179,
638
+ "step": 1020
639
+ },
640
+ {
641
+ "epoch": 2.64,
642
+ "learning_rate": 6.6476733143399815e-06,
643
+ "loss": 0.1913,
644
+ "step": 1030
645
+ },
646
+ {
647
+ "epoch": 2.67,
648
+ "learning_rate": 6.172839506172839e-06,
649
+ "loss": 0.1558,
650
+ "step": 1040
651
+ },
652
+ {
653
+ "epoch": 2.69,
654
+ "learning_rate": 5.6980056980056985e-06,
655
+ "loss": 0.1476,
656
+ "step": 1050
657
+ },
658
+ {
659
+ "epoch": 2.72,
660
+ "learning_rate": 5.223171889838557e-06,
661
+ "loss": 0.1661,
662
+ "step": 1060
663
+ },
664
+ {
665
+ "epoch": 2.74,
666
+ "learning_rate": 4.748338081671415e-06,
667
+ "loss": 0.1894,
668
+ "step": 1070
669
+ },
670
+ {
671
+ "epoch": 2.77,
672
+ "learning_rate": 4.273504273504274e-06,
673
+ "loss": 0.1751,
674
+ "step": 1080
675
+ },
676
+ {
677
+ "epoch": 2.79,
678
+ "learning_rate": 3.798670465337132e-06,
679
+ "loss": 0.1744,
680
+ "step": 1090
681
+ },
682
+ {
683
+ "epoch": 2.82,
684
+ "learning_rate": 3.3238366571699908e-06,
685
+ "loss": 0.1865,
686
+ "step": 1100
687
+ },
688
+ {
689
+ "epoch": 2.84,
690
+ "learning_rate": 2.8490028490028492e-06,
691
+ "loss": 0.1676,
692
+ "step": 1110
693
+ },
694
+ {
695
+ "epoch": 2.87,
696
+ "learning_rate": 2.3741690408357077e-06,
697
+ "loss": 0.2155,
698
+ "step": 1120
699
+ },
700
+ {
701
+ "epoch": 2.9,
702
+ "learning_rate": 1.899335232668566e-06,
703
+ "loss": 0.1763,
704
+ "step": 1130
705
+ },
706
+ {
707
+ "epoch": 2.92,
708
+ "learning_rate": 1.4245014245014246e-06,
709
+ "loss": 0.1727,
710
+ "step": 1140
711
+ },
712
+ {
713
+ "epoch": 2.95,
714
+ "learning_rate": 9.49667616334283e-07,
715
+ "loss": 0.1891,
716
+ "step": 1150
717
+ },
718
+ {
719
+ "epoch": 2.97,
720
+ "learning_rate": 4.748338081671415e-07,
721
+ "loss": 0.1833,
722
+ "step": 1160
723
+ },
724
+ {
725
+ "epoch": 3.0,
726
+ "learning_rate": 0.0,
727
+ "loss": 0.197,
728
+ "step": 1170
729
+ },
730
+ {
731
+ "epoch": 3.0,
732
+ "eval_accuracy": 0.9876,
733
+ "eval_loss": 0.04271771386265755,
734
+ "eval_runtime": 146.374,
735
+ "eval_samples_per_second": 68.318,
736
+ "eval_steps_per_second": 2.138,
737
+ "step": 1170
738
+ },
739
+ {
740
+ "epoch": 3.0,
741
+ "step": 1170,
742
+ "total_flos": 1.161843208194687e+19,
743
+ "train_loss": 0.3334213079550327,
744
+ "train_runtime": 6069.2812,
745
+ "train_samples_per_second": 24.715,
746
+ "train_steps_per_second": 0.193
747
+ }
748
+ ],
749
+ "max_steps": 1170,
750
+ "num_train_epochs": 3,
751
+ "total_flos": 1.161843208194687e+19,
752
+ "trial_name": null,
753
+ "trial_params": null
754
+ }