motheecreator commited on
Commit
f9b9574
1 Parent(s): 008bc67

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "eval_accuracy": 0.6982446363889663,
4
+ "eval_loss": 0.8523625135421753,
5
+ "eval_runtime": 60.7812,
6
+ "eval_samples_per_second": 118.096,
7
+ "eval_steps_per_second": 3.702,
8
+ "total_flos": 1.1101379964762415e+19,
9
+ "train_loss": 0.7277007077421461,
10
+ "train_runtime": 3200.5261,
11
+ "train_samples_per_second": 44.85,
12
+ "train_steps_per_second": 0.35
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "eval_accuracy": 0.6982446363889663,
4
+ "eval_loss": 0.8523625135421753,
5
+ "eval_runtime": 60.7812,
6
+ "eval_samples_per_second": 118.096,
7
+ "eval_steps_per_second": 3.702
8
+ }
runs/Apr29_23-45-53_786022f41cde/events.out.tfevents.1714437636.786022f41cde.42.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86229f8bf319882e4673c62b1d2cd89660e6269da58e88cdb7ece5ce2fb4a2e4
3
+ size 734
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "total_flos": 1.1101379964762415e+19,
4
+ "train_loss": 0.7277007077421461,
5
+ "train_runtime": 3200.5261,
6
+ "train_samples_per_second": 44.85,
7
+ "train_steps_per_second": 0.35
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,747 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6982446363889663,
3
+ "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned/checkpoint-1120",
4
+ "epoch": 4.988864142538976,
5
+ "eval_steps": 500,
6
+ "global_step": 1120,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "learning_rate": 4.464285714285715e-06,
14
+ "loss": 1.0215,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.09,
19
+ "learning_rate": 8.92857142857143e-06,
20
+ "loss": 1.0059,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.13,
25
+ "learning_rate": 1.3392857142857144e-05,
26
+ "loss": 1.0021,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.18,
31
+ "learning_rate": 1.785714285714286e-05,
32
+ "loss": 1.0043,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.22,
37
+ "learning_rate": 2.2321428571428575e-05,
38
+ "loss": 0.975,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.27,
43
+ "learning_rate": 2.6785714285714288e-05,
44
+ "loss": 0.9459,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.31,
49
+ "learning_rate": 3.125e-05,
50
+ "loss": 0.9092,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.36,
55
+ "learning_rate": 3.571428571428572e-05,
56
+ "loss": 0.911,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.4,
61
+ "learning_rate": 4.017857142857143e-05,
62
+ "loss": 0.961,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.45,
67
+ "learning_rate": 4.464285714285715e-05,
68
+ "loss": 0.9321,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.49,
73
+ "learning_rate": 4.910714285714286e-05,
74
+ "loss": 0.9215,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.53,
79
+ "learning_rate": 4.960317460317461e-05,
80
+ "loss": 0.8922,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.58,
85
+ "learning_rate": 4.910714285714286e-05,
86
+ "loss": 0.9519,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.62,
91
+ "learning_rate": 4.8611111111111115e-05,
92
+ "loss": 0.8926,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.67,
97
+ "learning_rate": 4.811507936507937e-05,
98
+ "loss": 0.8787,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.71,
103
+ "learning_rate": 4.761904761904762e-05,
104
+ "loss": 0.8783,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.76,
109
+ "learning_rate": 4.7123015873015876e-05,
110
+ "loss": 0.8611,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.8,
115
+ "learning_rate": 4.662698412698413e-05,
116
+ "loss": 0.9162,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.85,
121
+ "learning_rate": 4.613095238095239e-05,
122
+ "loss": 0.8853,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.89,
127
+ "learning_rate": 4.563492063492064e-05,
128
+ "loss": 0.8687,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.94,
133
+ "learning_rate": 4.5138888888888894e-05,
134
+ "loss": 0.8379,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.98,
139
+ "learning_rate": 4.464285714285715e-05,
140
+ "loss": 0.8333,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 1.0,
145
+ "eval_accuracy": 0.6473948174979103,
146
+ "eval_loss": 0.9670450687408447,
147
+ "eval_runtime": 60.2016,
148
+ "eval_samples_per_second": 119.233,
149
+ "eval_steps_per_second": 3.737,
150
+ "step": 224
151
+ },
152
+ {
153
+ "epoch": 1.02,
154
+ "learning_rate": 4.41468253968254e-05,
155
+ "loss": 0.8652,
156
+ "step": 230
157
+ },
158
+ {
159
+ "epoch": 1.07,
160
+ "learning_rate": 4.3650793650793655e-05,
161
+ "loss": 0.8686,
162
+ "step": 240
163
+ },
164
+ {
165
+ "epoch": 1.11,
166
+ "learning_rate": 4.315476190476191e-05,
167
+ "loss": 0.8604,
168
+ "step": 250
169
+ },
170
+ {
171
+ "epoch": 1.16,
172
+ "learning_rate": 4.265873015873016e-05,
173
+ "loss": 0.8377,
174
+ "step": 260
175
+ },
176
+ {
177
+ "epoch": 1.2,
178
+ "learning_rate": 4.2162698412698416e-05,
179
+ "loss": 0.8151,
180
+ "step": 270
181
+ },
182
+ {
183
+ "epoch": 1.25,
184
+ "learning_rate": 4.166666666666667e-05,
185
+ "loss": 0.8818,
186
+ "step": 280
187
+ },
188
+ {
189
+ "epoch": 1.29,
190
+ "learning_rate": 4.117063492063492e-05,
191
+ "loss": 0.8108,
192
+ "step": 290
193
+ },
194
+ {
195
+ "epoch": 1.34,
196
+ "learning_rate": 4.067460317460318e-05,
197
+ "loss": 0.8408,
198
+ "step": 300
199
+ },
200
+ {
201
+ "epoch": 1.38,
202
+ "learning_rate": 4.017857142857143e-05,
203
+ "loss": 0.8801,
204
+ "step": 310
205
+ },
206
+ {
207
+ "epoch": 1.43,
208
+ "learning_rate": 3.968253968253968e-05,
209
+ "loss": 0.8194,
210
+ "step": 320
211
+ },
212
+ {
213
+ "epoch": 1.47,
214
+ "learning_rate": 3.918650793650794e-05,
215
+ "loss": 0.862,
216
+ "step": 330
217
+ },
218
+ {
219
+ "epoch": 1.51,
220
+ "learning_rate": 3.8690476190476195e-05,
221
+ "loss": 0.8165,
222
+ "step": 340
223
+ },
224
+ {
225
+ "epoch": 1.56,
226
+ "learning_rate": 3.8194444444444444e-05,
227
+ "loss": 0.8342,
228
+ "step": 350
229
+ },
230
+ {
231
+ "epoch": 1.6,
232
+ "learning_rate": 3.76984126984127e-05,
233
+ "loss": 0.8397,
234
+ "step": 360
235
+ },
236
+ {
237
+ "epoch": 1.65,
238
+ "learning_rate": 3.7202380952380956e-05,
239
+ "loss": 0.8318,
240
+ "step": 370
241
+ },
242
+ {
243
+ "epoch": 1.69,
244
+ "learning_rate": 3.6706349206349205e-05,
245
+ "loss": 0.792,
246
+ "step": 380
247
+ },
248
+ {
249
+ "epoch": 1.74,
250
+ "learning_rate": 3.621031746031746e-05,
251
+ "loss": 0.844,
252
+ "step": 390
253
+ },
254
+ {
255
+ "epoch": 1.78,
256
+ "learning_rate": 3.571428571428572e-05,
257
+ "loss": 0.8179,
258
+ "step": 400
259
+ },
260
+ {
261
+ "epoch": 1.83,
262
+ "learning_rate": 3.521825396825397e-05,
263
+ "loss": 0.7968,
264
+ "step": 410
265
+ },
266
+ {
267
+ "epoch": 1.87,
268
+ "learning_rate": 3.472222222222222e-05,
269
+ "loss": 0.7903,
270
+ "step": 420
271
+ },
272
+ {
273
+ "epoch": 1.92,
274
+ "learning_rate": 3.422619047619048e-05,
275
+ "loss": 0.7977,
276
+ "step": 430
277
+ },
278
+ {
279
+ "epoch": 1.96,
280
+ "learning_rate": 3.3730158730158734e-05,
281
+ "loss": 0.7972,
282
+ "step": 440
283
+ },
284
+ {
285
+ "epoch": 2.0,
286
+ "eval_accuracy": 0.6653663973251602,
287
+ "eval_loss": 0.9123018383979797,
288
+ "eval_runtime": 59.7062,
289
+ "eval_samples_per_second": 120.222,
290
+ "eval_steps_per_second": 3.768,
291
+ "step": 449
292
+ },
293
+ {
294
+ "epoch": 2.0,
295
+ "learning_rate": 3.3234126984126983e-05,
296
+ "loss": 0.8381,
297
+ "step": 450
298
+ },
299
+ {
300
+ "epoch": 2.05,
301
+ "learning_rate": 3.273809523809524e-05,
302
+ "loss": 0.7561,
303
+ "step": 460
304
+ },
305
+ {
306
+ "epoch": 2.09,
307
+ "learning_rate": 3.2242063492063495e-05,
308
+ "loss": 0.7124,
309
+ "step": 470
310
+ },
311
+ {
312
+ "epoch": 2.14,
313
+ "learning_rate": 3.1746031746031745e-05,
314
+ "loss": 0.7322,
315
+ "step": 480
316
+ },
317
+ {
318
+ "epoch": 2.18,
319
+ "learning_rate": 3.125e-05,
320
+ "loss": 0.7348,
321
+ "step": 490
322
+ },
323
+ {
324
+ "epoch": 2.23,
325
+ "learning_rate": 3.075396825396826e-05,
326
+ "loss": 0.7468,
327
+ "step": 500
328
+ },
329
+ {
330
+ "epoch": 2.27,
331
+ "learning_rate": 3.0257936507936506e-05,
332
+ "loss": 0.7106,
333
+ "step": 510
334
+ },
335
+ {
336
+ "epoch": 2.32,
337
+ "learning_rate": 2.9761904761904762e-05,
338
+ "loss": 0.7585,
339
+ "step": 520
340
+ },
341
+ {
342
+ "epoch": 2.36,
343
+ "learning_rate": 2.9265873015873018e-05,
344
+ "loss": 0.7383,
345
+ "step": 530
346
+ },
347
+ {
348
+ "epoch": 2.41,
349
+ "learning_rate": 2.876984126984127e-05,
350
+ "loss": 0.7238,
351
+ "step": 540
352
+ },
353
+ {
354
+ "epoch": 2.45,
355
+ "learning_rate": 2.8273809523809523e-05,
356
+ "loss": 0.7042,
357
+ "step": 550
358
+ },
359
+ {
360
+ "epoch": 2.49,
361
+ "learning_rate": 2.777777777777778e-05,
362
+ "loss": 0.7076,
363
+ "step": 560
364
+ },
365
+ {
366
+ "epoch": 2.54,
367
+ "learning_rate": 2.7281746031746032e-05,
368
+ "loss": 0.7441,
369
+ "step": 570
370
+ },
371
+ {
372
+ "epoch": 2.58,
373
+ "learning_rate": 2.6785714285714288e-05,
374
+ "loss": 0.6887,
375
+ "step": 580
376
+ },
377
+ {
378
+ "epoch": 2.63,
379
+ "learning_rate": 2.628968253968254e-05,
380
+ "loss": 0.7216,
381
+ "step": 590
382
+ },
383
+ {
384
+ "epoch": 2.67,
385
+ "learning_rate": 2.5793650793650796e-05,
386
+ "loss": 0.71,
387
+ "step": 600
388
+ },
389
+ {
390
+ "epoch": 2.72,
391
+ "learning_rate": 2.529761904761905e-05,
392
+ "loss": 0.7265,
393
+ "step": 610
394
+ },
395
+ {
396
+ "epoch": 2.76,
397
+ "learning_rate": 2.4801587301587305e-05,
398
+ "loss": 0.697,
399
+ "step": 620
400
+ },
401
+ {
402
+ "epoch": 2.81,
403
+ "learning_rate": 2.4305555555555558e-05,
404
+ "loss": 0.7578,
405
+ "step": 630
406
+ },
407
+ {
408
+ "epoch": 2.85,
409
+ "learning_rate": 2.380952380952381e-05,
410
+ "loss": 0.6838,
411
+ "step": 640
412
+ },
413
+ {
414
+ "epoch": 2.9,
415
+ "learning_rate": 2.3313492063492066e-05,
416
+ "loss": 0.7082,
417
+ "step": 650
418
+ },
419
+ {
420
+ "epoch": 2.94,
421
+ "learning_rate": 2.281746031746032e-05,
422
+ "loss": 0.6824,
423
+ "step": 660
424
+ },
425
+ {
426
+ "epoch": 2.98,
427
+ "learning_rate": 2.2321428571428575e-05,
428
+ "loss": 0.667,
429
+ "step": 670
430
+ },
431
+ {
432
+ "epoch": 3.0,
433
+ "eval_accuracy": 0.6886319308999721,
434
+ "eval_loss": 0.8676984906196594,
435
+ "eval_runtime": 59.6434,
436
+ "eval_samples_per_second": 120.349,
437
+ "eval_steps_per_second": 3.772,
438
+ "step": 673
439
+ },
440
+ {
441
+ "epoch": 3.03,
442
+ "learning_rate": 2.1825396825396827e-05,
443
+ "loss": 0.6546,
444
+ "step": 680
445
+ },
446
+ {
447
+ "epoch": 3.07,
448
+ "learning_rate": 2.132936507936508e-05,
449
+ "loss": 0.6349,
450
+ "step": 690
451
+ },
452
+ {
453
+ "epoch": 3.12,
454
+ "learning_rate": 2.0833333333333336e-05,
455
+ "loss": 0.6229,
456
+ "step": 700
457
+ },
458
+ {
459
+ "epoch": 3.16,
460
+ "learning_rate": 2.033730158730159e-05,
461
+ "loss": 0.653,
462
+ "step": 710
463
+ },
464
+ {
465
+ "epoch": 3.21,
466
+ "learning_rate": 1.984126984126984e-05,
467
+ "loss": 0.6433,
468
+ "step": 720
469
+ },
470
+ {
471
+ "epoch": 3.25,
472
+ "learning_rate": 1.9345238095238097e-05,
473
+ "loss": 0.6291,
474
+ "step": 730
475
+ },
476
+ {
477
+ "epoch": 3.3,
478
+ "learning_rate": 1.884920634920635e-05,
479
+ "loss": 0.6621,
480
+ "step": 740
481
+ },
482
+ {
483
+ "epoch": 3.34,
484
+ "learning_rate": 1.8353174603174602e-05,
485
+ "loss": 0.6116,
486
+ "step": 750
487
+ },
488
+ {
489
+ "epoch": 3.39,
490
+ "learning_rate": 1.785714285714286e-05,
491
+ "loss": 0.6441,
492
+ "step": 760
493
+ },
494
+ {
495
+ "epoch": 3.43,
496
+ "learning_rate": 1.736111111111111e-05,
497
+ "loss": 0.642,
498
+ "step": 770
499
+ },
500
+ {
501
+ "epoch": 3.47,
502
+ "learning_rate": 1.6865079365079367e-05,
503
+ "loss": 0.6129,
504
+ "step": 780
505
+ },
506
+ {
507
+ "epoch": 3.52,
508
+ "learning_rate": 1.636904761904762e-05,
509
+ "loss": 0.5959,
510
+ "step": 790
511
+ },
512
+ {
513
+ "epoch": 3.56,
514
+ "learning_rate": 1.5873015873015872e-05,
515
+ "loss": 0.6257,
516
+ "step": 800
517
+ },
518
+ {
519
+ "epoch": 3.61,
520
+ "learning_rate": 1.537698412698413e-05,
521
+ "loss": 0.5942,
522
+ "step": 810
523
+ },
524
+ {
525
+ "epoch": 3.65,
526
+ "learning_rate": 1.4880952380952381e-05,
527
+ "loss": 0.5848,
528
+ "step": 820
529
+ },
530
+ {
531
+ "epoch": 3.7,
532
+ "learning_rate": 1.4384920634920635e-05,
533
+ "loss": 0.6071,
534
+ "step": 830
535
+ },
536
+ {
537
+ "epoch": 3.74,
538
+ "learning_rate": 1.388888888888889e-05,
539
+ "loss": 0.6009,
540
+ "step": 840
541
+ },
542
+ {
543
+ "epoch": 3.79,
544
+ "learning_rate": 1.3392857142857144e-05,
545
+ "loss": 0.6541,
546
+ "step": 850
547
+ },
548
+ {
549
+ "epoch": 3.83,
550
+ "learning_rate": 1.2896825396825398e-05,
551
+ "loss": 0.6306,
552
+ "step": 860
553
+ },
554
+ {
555
+ "epoch": 3.88,
556
+ "learning_rate": 1.2400793650793652e-05,
557
+ "loss": 0.6398,
558
+ "step": 870
559
+ },
560
+ {
561
+ "epoch": 3.92,
562
+ "learning_rate": 1.1904761904761905e-05,
563
+ "loss": 0.6285,
564
+ "step": 880
565
+ },
566
+ {
567
+ "epoch": 3.96,
568
+ "learning_rate": 1.140873015873016e-05,
569
+ "loss": 0.5729,
570
+ "step": 890
571
+ },
572
+ {
573
+ "epoch": 4.0,
574
+ "eval_accuracy": 0.6937865700752298,
575
+ "eval_loss": 0.8486846685409546,
576
+ "eval_runtime": 59.646,
577
+ "eval_samples_per_second": 120.343,
578
+ "eval_steps_per_second": 3.772,
579
+ "step": 898
580
+ },
581
+ {
582
+ "epoch": 4.01,
583
+ "learning_rate": 1.0912698412698414e-05,
584
+ "loss": 0.5634,
585
+ "step": 900
586
+ },
587
+ {
588
+ "epoch": 4.05,
589
+ "learning_rate": 1.0416666666666668e-05,
590
+ "loss": 0.5553,
591
+ "step": 910
592
+ },
593
+ {
594
+ "epoch": 4.1,
595
+ "learning_rate": 9.92063492063492e-06,
596
+ "loss": 0.5474,
597
+ "step": 920
598
+ },
599
+ {
600
+ "epoch": 4.14,
601
+ "learning_rate": 9.424603174603175e-06,
602
+ "loss": 0.5299,
603
+ "step": 930
604
+ },
605
+ {
606
+ "epoch": 4.19,
607
+ "learning_rate": 8.92857142857143e-06,
608
+ "loss": 0.5541,
609
+ "step": 940
610
+ },
611
+ {
612
+ "epoch": 4.23,
613
+ "learning_rate": 8.432539682539684e-06,
614
+ "loss": 0.5341,
615
+ "step": 950
616
+ },
617
+ {
618
+ "epoch": 4.28,
619
+ "learning_rate": 7.936507936507936e-06,
620
+ "loss": 0.5315,
621
+ "step": 960
622
+ },
623
+ {
624
+ "epoch": 4.32,
625
+ "learning_rate": 7.4404761904761905e-06,
626
+ "loss": 0.5265,
627
+ "step": 970
628
+ },
629
+ {
630
+ "epoch": 4.37,
631
+ "learning_rate": 6.944444444444445e-06,
632
+ "loss": 0.5063,
633
+ "step": 980
634
+ },
635
+ {
636
+ "epoch": 4.41,
637
+ "learning_rate": 6.448412698412699e-06,
638
+ "loss": 0.53,
639
+ "step": 990
640
+ },
641
+ {
642
+ "epoch": 4.45,
643
+ "learning_rate": 5.9523809523809525e-06,
644
+ "loss": 0.5322,
645
+ "step": 1000
646
+ },
647
+ {
648
+ "epoch": 4.5,
649
+ "learning_rate": 5.456349206349207e-06,
650
+ "loss": 0.5397,
651
+ "step": 1010
652
+ },
653
+ {
654
+ "epoch": 4.54,
655
+ "learning_rate": 4.96031746031746e-06,
656
+ "loss": 0.533,
657
+ "step": 1020
658
+ },
659
+ {
660
+ "epoch": 4.59,
661
+ "learning_rate": 4.464285714285715e-06,
662
+ "loss": 0.6067,
663
+ "step": 1030
664
+ },
665
+ {
666
+ "epoch": 4.63,
667
+ "learning_rate": 3.968253968253968e-06,
668
+ "loss": 0.568,
669
+ "step": 1040
670
+ },
671
+ {
672
+ "epoch": 4.68,
673
+ "learning_rate": 3.4722222222222224e-06,
674
+ "loss": 0.5507,
675
+ "step": 1050
676
+ },
677
+ {
678
+ "epoch": 4.72,
679
+ "learning_rate": 2.9761904761904763e-06,
680
+ "loss": 0.5993,
681
+ "step": 1060
682
+ },
683
+ {
684
+ "epoch": 4.77,
685
+ "learning_rate": 2.48015873015873e-06,
686
+ "loss": 0.5343,
687
+ "step": 1070
688
+ },
689
+ {
690
+ "epoch": 4.81,
691
+ "learning_rate": 1.984126984126984e-06,
692
+ "loss": 0.5301,
693
+ "step": 1080
694
+ },
695
+ {
696
+ "epoch": 4.86,
697
+ "learning_rate": 1.4880952380952381e-06,
698
+ "loss": 0.5446,
699
+ "step": 1090
700
+ },
701
+ {
702
+ "epoch": 4.9,
703
+ "learning_rate": 9.92063492063492e-07,
704
+ "loss": 0.5496,
705
+ "step": 1100
706
+ },
707
+ {
708
+ "epoch": 4.94,
709
+ "learning_rate": 4.96031746031746e-07,
710
+ "loss": 0.5198,
711
+ "step": 1110
712
+ },
713
+ {
714
+ "epoch": 4.99,
715
+ "learning_rate": 0.0,
716
+ "loss": 0.5347,
717
+ "step": 1120
718
+ },
719
+ {
720
+ "epoch": 4.99,
721
+ "eval_accuracy": 0.6982446363889663,
722
+ "eval_loss": 0.8523625135421753,
723
+ "eval_runtime": 60.8207,
724
+ "eval_samples_per_second": 118.019,
725
+ "eval_steps_per_second": 3.699,
726
+ "step": 1120
727
+ },
728
+ {
729
+ "epoch": 4.99,
730
+ "step": 1120,
731
+ "total_flos": 1.1101379964762415e+19,
732
+ "train_loss": 0.7277007077421461,
733
+ "train_runtime": 3200.5261,
734
+ "train_samples_per_second": 44.85,
735
+ "train_steps_per_second": 0.35
736
+ }
737
+ ],
738
+ "logging_steps": 10,
739
+ "max_steps": 1120,
740
+ "num_input_tokens_seen": 0,
741
+ "num_train_epochs": 5,
742
+ "save_steps": 500,
743
+ "total_flos": 1.1101379964762415e+19,
744
+ "train_batch_size": 32,
745
+ "trial_name": null,
746
+ "trial_params": null
747
+ }