Soulaimen commited on
Commit
11f7b81
1 Parent(s): 66cb62f

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9781420765027322,
4
+ "eval_loss": 0.1102537140250206,
5
+ "eval_runtime": 27.0948,
6
+ "eval_samples_per_second": 33.77,
7
+ "eval_steps_per_second": 4.244,
8
+ "total_flos": 1.7471888881490534e+18,
9
+ "train_loss": 0.49037140356440123,
10
+ "train_runtime": 8760.81,
11
+ "train_samples_per_second": 9.391,
12
+ "train_steps_per_second": 0.168
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.9781420765027322,
4
+ "eval_loss": 0.1102537140250206,
5
+ "eval_runtime": 27.0948,
6
+ "eval_samples_per_second": 33.77,
7
+ "eval_steps_per_second": 4.244
8
+ }
runs/May04_21-32-08_8b61ba85a7e1/events.out.tfevents.1683244744.8b61ba85a7e1.301.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36b2724a0822fa65209254be945387e29cb34d5ed2f9c9d3fbde86de3b0e20c9
3
+ size 411
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.7471888881490534e+18,
4
+ "train_loss": 0.49037140356440123,
5
+ "train_runtime": 8760.81,
6
+ "train_samples_per_second": 9.391,
7
+ "train_steps_per_second": 0.168
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,997 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9781420765027322,
3
+ "best_model_checkpoint": "resnet-50-shortSleeveCleanedData/checkpoint-1470",
4
+ "epoch": 10.0,
5
+ "global_step": 1470,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.07,
12
+ "learning_rate": 3.3333333333333335e-05,
13
+ "loss": 1.0923,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.14,
18
+ "learning_rate": 4.982817869415808e-05,
19
+ "loss": 1.0936,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.2,
24
+ "learning_rate": 4.948453608247423e-05,
25
+ "loss": 1.0859,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.27,
30
+ "learning_rate": 4.9140893470790375e-05,
31
+ "loss": 1.0726,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.34,
36
+ "learning_rate": 4.879725085910653e-05,
37
+ "loss": 1.0655,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.41,
42
+ "learning_rate": 4.845360824742268e-05,
43
+ "loss": 1.0659,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.48,
48
+ "learning_rate": 4.810996563573883e-05,
49
+ "loss": 1.0552,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.54,
54
+ "learning_rate": 4.776632302405499e-05,
55
+ "loss": 1.0441,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.61,
60
+ "learning_rate": 4.7422680412371134e-05,
61
+ "loss": 1.0318,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.68,
66
+ "learning_rate": 4.7079037800687284e-05,
67
+ "loss": 1.0237,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.75,
72
+ "learning_rate": 4.673539518900344e-05,
73
+ "loss": 1.0043,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.82,
78
+ "learning_rate": 4.639175257731959e-05,
79
+ "loss": 1.0046,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.88,
84
+ "learning_rate": 4.6048109965635736e-05,
85
+ "loss": 0.9857,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.95,
90
+ "learning_rate": 4.570446735395189e-05,
91
+ "loss": 0.973,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 1.0,
96
+ "eval_accuracy": 0.726775956284153,
97
+ "eval_loss": 0.9370501637458801,
98
+ "eval_runtime": 577.0712,
99
+ "eval_samples_per_second": 1.586,
100
+ "eval_steps_per_second": 0.199,
101
+ "step": 147
102
+ },
103
+ {
104
+ "epoch": 1.02,
105
+ "learning_rate": 4.536082474226804e-05,
106
+ "loss": 0.9592,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.09,
111
+ "learning_rate": 4.5017182130584194e-05,
112
+ "loss": 0.9381,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.16,
117
+ "learning_rate": 4.467353951890035e-05,
118
+ "loss": 0.9292,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.22,
123
+ "learning_rate": 4.4329896907216494e-05,
124
+ "loss": 0.8919,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.29,
129
+ "learning_rate": 4.3986254295532645e-05,
130
+ "loss": 0.8828,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.36,
135
+ "learning_rate": 4.36426116838488e-05,
136
+ "loss": 0.8452,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.43,
141
+ "learning_rate": 4.329896907216495e-05,
142
+ "loss": 0.8414,
143
+ "step": 210
144
+ },
145
+ {
146
+ "epoch": 1.5,
147
+ "learning_rate": 4.2955326460481096e-05,
148
+ "loss": 0.8229,
149
+ "step": 220
150
+ },
151
+ {
152
+ "epoch": 1.56,
153
+ "learning_rate": 4.261168384879725e-05,
154
+ "loss": 0.8034,
155
+ "step": 230
156
+ },
157
+ {
158
+ "epoch": 1.63,
159
+ "learning_rate": 4.2268041237113404e-05,
160
+ "loss": 0.7824,
161
+ "step": 240
162
+ },
163
+ {
164
+ "epoch": 1.7,
165
+ "learning_rate": 4.1924398625429554e-05,
166
+ "loss": 0.7358,
167
+ "step": 250
168
+ },
169
+ {
170
+ "epoch": 1.77,
171
+ "learning_rate": 4.158075601374571e-05,
172
+ "loss": 0.7344,
173
+ "step": 260
174
+ },
175
+ {
176
+ "epoch": 1.84,
177
+ "learning_rate": 4.1237113402061855e-05,
178
+ "loss": 0.6981,
179
+ "step": 270
180
+ },
181
+ {
182
+ "epoch": 1.9,
183
+ "learning_rate": 4.0893470790378005e-05,
184
+ "loss": 0.6869,
185
+ "step": 280
186
+ },
187
+ {
188
+ "epoch": 1.97,
189
+ "learning_rate": 4.054982817869416e-05,
190
+ "loss": 0.6565,
191
+ "step": 290
192
+ },
193
+ {
194
+ "epoch": 2.0,
195
+ "eval_accuracy": 0.8710382513661202,
196
+ "eval_loss": 0.5519522428512573,
197
+ "eval_runtime": 27.0894,
198
+ "eval_samples_per_second": 33.777,
199
+ "eval_steps_per_second": 4.245,
200
+ "step": 294
201
+ },
202
+ {
203
+ "epoch": 2.04,
204
+ "learning_rate": 4.020618556701031e-05,
205
+ "loss": 0.663,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.11,
210
+ "learning_rate": 3.9862542955326463e-05,
211
+ "loss": 0.648,
212
+ "step": 310
213
+ },
214
+ {
215
+ "epoch": 2.18,
216
+ "learning_rate": 3.9518900343642614e-05,
217
+ "loss": 0.6162,
218
+ "step": 320
219
+ },
220
+ {
221
+ "epoch": 2.24,
222
+ "learning_rate": 3.9175257731958764e-05,
223
+ "loss": 0.5981,
224
+ "step": 330
225
+ },
226
+ {
227
+ "epoch": 2.31,
228
+ "learning_rate": 3.8831615120274915e-05,
229
+ "loss": 0.5583,
230
+ "step": 340
231
+ },
232
+ {
233
+ "epoch": 2.38,
234
+ "learning_rate": 3.848797250859107e-05,
235
+ "loss": 0.5563,
236
+ "step": 350
237
+ },
238
+ {
239
+ "epoch": 2.45,
240
+ "learning_rate": 3.8144329896907216e-05,
241
+ "loss": 0.563,
242
+ "step": 360
243
+ },
244
+ {
245
+ "epoch": 2.52,
246
+ "learning_rate": 3.7800687285223366e-05,
247
+ "loss": 0.534,
248
+ "step": 370
249
+ },
250
+ {
251
+ "epoch": 2.59,
252
+ "learning_rate": 3.745704467353952e-05,
253
+ "loss": 0.5228,
254
+ "step": 380
255
+ },
256
+ {
257
+ "epoch": 2.65,
258
+ "learning_rate": 3.7113402061855674e-05,
259
+ "loss": 0.5041,
260
+ "step": 390
261
+ },
262
+ {
263
+ "epoch": 2.72,
264
+ "learning_rate": 3.6769759450171824e-05,
265
+ "loss": 0.4817,
266
+ "step": 400
267
+ },
268
+ {
269
+ "epoch": 2.79,
270
+ "learning_rate": 3.6426116838487974e-05,
271
+ "loss": 0.5135,
272
+ "step": 410
273
+ },
274
+ {
275
+ "epoch": 2.86,
276
+ "learning_rate": 3.6082474226804125e-05,
277
+ "loss": 0.4676,
278
+ "step": 420
279
+ },
280
+ {
281
+ "epoch": 2.93,
282
+ "learning_rate": 3.5738831615120275e-05,
283
+ "loss": 0.4882,
284
+ "step": 430
285
+ },
286
+ {
287
+ "epoch": 2.99,
288
+ "learning_rate": 3.539518900343643e-05,
289
+ "loss": 0.4609,
290
+ "step": 440
291
+ },
292
+ {
293
+ "epoch": 3.0,
294
+ "eval_accuracy": 0.9278688524590164,
295
+ "eval_loss": 0.29834744334220886,
296
+ "eval_runtime": 25.7025,
297
+ "eval_samples_per_second": 35.6,
298
+ "eval_steps_per_second": 4.474,
299
+ "step": 441
300
+ },
301
+ {
302
+ "epoch": 3.06,
303
+ "learning_rate": 3.5051546391752576e-05,
304
+ "loss": 0.4741,
305
+ "step": 450
306
+ },
307
+ {
308
+ "epoch": 3.13,
309
+ "learning_rate": 3.4707903780068726e-05,
310
+ "loss": 0.4635,
311
+ "step": 460
312
+ },
313
+ {
314
+ "epoch": 3.2,
315
+ "learning_rate": 3.4364261168384884e-05,
316
+ "loss": 0.4645,
317
+ "step": 470
318
+ },
319
+ {
320
+ "epoch": 3.27,
321
+ "learning_rate": 3.4020618556701034e-05,
322
+ "loss": 0.4558,
323
+ "step": 480
324
+ },
325
+ {
326
+ "epoch": 3.33,
327
+ "learning_rate": 3.3676975945017185e-05,
328
+ "loss": 0.4318,
329
+ "step": 490
330
+ },
331
+ {
332
+ "epoch": 3.4,
333
+ "learning_rate": 3.3333333333333335e-05,
334
+ "loss": 0.4389,
335
+ "step": 500
336
+ },
337
+ {
338
+ "epoch": 3.47,
339
+ "learning_rate": 3.2989690721649485e-05,
340
+ "loss": 0.4614,
341
+ "step": 510
342
+ },
343
+ {
344
+ "epoch": 3.54,
345
+ "learning_rate": 3.2646048109965636e-05,
346
+ "loss": 0.3683,
347
+ "step": 520
348
+ },
349
+ {
350
+ "epoch": 3.61,
351
+ "learning_rate": 3.230240549828179e-05,
352
+ "loss": 0.4364,
353
+ "step": 530
354
+ },
355
+ {
356
+ "epoch": 3.67,
357
+ "learning_rate": 3.1958762886597937e-05,
358
+ "loss": 0.4254,
359
+ "step": 540
360
+ },
361
+ {
362
+ "epoch": 3.74,
363
+ "learning_rate": 3.161512027491409e-05,
364
+ "loss": 0.4524,
365
+ "step": 550
366
+ },
367
+ {
368
+ "epoch": 3.81,
369
+ "learning_rate": 3.1271477663230244e-05,
370
+ "loss": 0.4369,
371
+ "step": 560
372
+ },
373
+ {
374
+ "epoch": 3.88,
375
+ "learning_rate": 3.0927835051546395e-05,
376
+ "loss": 0.3894,
377
+ "step": 570
378
+ },
379
+ {
380
+ "epoch": 3.95,
381
+ "learning_rate": 3.0584192439862545e-05,
382
+ "loss": 0.3937,
383
+ "step": 580
384
+ },
385
+ {
386
+ "epoch": 4.0,
387
+ "eval_accuracy": 0.9486338797814208,
388
+ "eval_loss": 0.20506763458251953,
389
+ "eval_runtime": 26.8515,
390
+ "eval_samples_per_second": 34.076,
391
+ "eval_steps_per_second": 4.283,
392
+ "step": 588
393
+ },
394
+ {
395
+ "epoch": 4.01,
396
+ "learning_rate": 3.0240549828178692e-05,
397
+ "loss": 0.4192,
398
+ "step": 590
399
+ },
400
+ {
401
+ "epoch": 4.08,
402
+ "learning_rate": 2.9896907216494846e-05,
403
+ "loss": 0.4003,
404
+ "step": 600
405
+ },
406
+ {
407
+ "epoch": 4.15,
408
+ "learning_rate": 2.9553264604811e-05,
409
+ "loss": 0.4544,
410
+ "step": 610
411
+ },
412
+ {
413
+ "epoch": 4.22,
414
+ "learning_rate": 2.920962199312715e-05,
415
+ "loss": 0.3996,
416
+ "step": 620
417
+ },
418
+ {
419
+ "epoch": 4.29,
420
+ "learning_rate": 2.8865979381443297e-05,
421
+ "loss": 0.4168,
422
+ "step": 630
423
+ },
424
+ {
425
+ "epoch": 4.35,
426
+ "learning_rate": 2.852233676975945e-05,
427
+ "loss": 0.3964,
428
+ "step": 640
429
+ },
430
+ {
431
+ "epoch": 4.42,
432
+ "learning_rate": 2.81786941580756e-05,
433
+ "loss": 0.4113,
434
+ "step": 650
435
+ },
436
+ {
437
+ "epoch": 4.49,
438
+ "learning_rate": 2.7835051546391755e-05,
439
+ "loss": 0.3912,
440
+ "step": 660
441
+ },
442
+ {
443
+ "epoch": 4.56,
444
+ "learning_rate": 2.749140893470791e-05,
445
+ "loss": 0.3793,
446
+ "step": 670
447
+ },
448
+ {
449
+ "epoch": 4.63,
450
+ "learning_rate": 2.7147766323024053e-05,
451
+ "loss": 0.4175,
452
+ "step": 680
453
+ },
454
+ {
455
+ "epoch": 4.69,
456
+ "learning_rate": 2.6804123711340206e-05,
457
+ "loss": 0.3831,
458
+ "step": 690
459
+ },
460
+ {
461
+ "epoch": 4.76,
462
+ "learning_rate": 2.646048109965636e-05,
463
+ "loss": 0.3818,
464
+ "step": 700
465
+ },
466
+ {
467
+ "epoch": 4.83,
468
+ "learning_rate": 2.611683848797251e-05,
469
+ "loss": 0.3901,
470
+ "step": 710
471
+ },
472
+ {
473
+ "epoch": 4.9,
474
+ "learning_rate": 2.5773195876288658e-05,
475
+ "loss": 0.3541,
476
+ "step": 720
477
+ },
478
+ {
479
+ "epoch": 4.97,
480
+ "learning_rate": 2.542955326460481e-05,
481
+ "loss": 0.3723,
482
+ "step": 730
483
+ },
484
+ {
485
+ "epoch": 5.0,
486
+ "eval_accuracy": 0.9726775956284153,
487
+ "eval_loss": 0.15214987099170685,
488
+ "eval_runtime": 26.7904,
489
+ "eval_samples_per_second": 34.154,
490
+ "eval_steps_per_second": 4.293,
491
+ "step": 735
492
+ },
493
+ {
494
+ "epoch": 5.03,
495
+ "learning_rate": 2.5085910652920962e-05,
496
+ "loss": 0.3624,
497
+ "step": 740
498
+ },
499
+ {
500
+ "epoch": 5.1,
501
+ "learning_rate": 2.4742268041237116e-05,
502
+ "loss": 0.3777,
503
+ "step": 750
504
+ },
505
+ {
506
+ "epoch": 5.17,
507
+ "learning_rate": 2.4398625429553266e-05,
508
+ "loss": 0.3625,
509
+ "step": 760
510
+ },
511
+ {
512
+ "epoch": 5.24,
513
+ "learning_rate": 2.4054982817869417e-05,
514
+ "loss": 0.3478,
515
+ "step": 770
516
+ },
517
+ {
518
+ "epoch": 5.31,
519
+ "learning_rate": 2.3711340206185567e-05,
520
+ "loss": 0.3732,
521
+ "step": 780
522
+ },
523
+ {
524
+ "epoch": 5.37,
525
+ "learning_rate": 2.336769759450172e-05,
526
+ "loss": 0.3508,
527
+ "step": 790
528
+ },
529
+ {
530
+ "epoch": 5.44,
531
+ "learning_rate": 2.3024054982817868e-05,
532
+ "loss": 0.373,
533
+ "step": 800
534
+ },
535
+ {
536
+ "epoch": 5.51,
537
+ "learning_rate": 2.268041237113402e-05,
538
+ "loss": 0.406,
539
+ "step": 810
540
+ },
541
+ {
542
+ "epoch": 5.58,
543
+ "learning_rate": 2.2336769759450175e-05,
544
+ "loss": 0.3409,
545
+ "step": 820
546
+ },
547
+ {
548
+ "epoch": 5.65,
549
+ "learning_rate": 2.1993127147766322e-05,
550
+ "loss": 0.3647,
551
+ "step": 830
552
+ },
553
+ {
554
+ "epoch": 5.71,
555
+ "learning_rate": 2.1649484536082476e-05,
556
+ "loss": 0.3719,
557
+ "step": 840
558
+ },
559
+ {
560
+ "epoch": 5.78,
561
+ "learning_rate": 2.1305841924398627e-05,
562
+ "loss": 0.3631,
563
+ "step": 850
564
+ },
565
+ {
566
+ "epoch": 5.85,
567
+ "learning_rate": 2.0962199312714777e-05,
568
+ "loss": 0.3574,
569
+ "step": 860
570
+ },
571
+ {
572
+ "epoch": 5.92,
573
+ "learning_rate": 2.0618556701030927e-05,
574
+ "loss": 0.3704,
575
+ "step": 870
576
+ },
577
+ {
578
+ "epoch": 5.99,
579
+ "learning_rate": 2.027491408934708e-05,
580
+ "loss": 0.3926,
581
+ "step": 880
582
+ },
583
+ {
584
+ "epoch": 6.0,
585
+ "eval_accuracy": 0.9672131147540983,
586
+ "eval_loss": 0.14902302622795105,
587
+ "eval_runtime": 25.8181,
588
+ "eval_samples_per_second": 35.44,
589
+ "eval_steps_per_second": 4.454,
590
+ "step": 882
591
+ },
592
+ {
593
+ "epoch": 6.05,
594
+ "learning_rate": 1.9931271477663232e-05,
595
+ "loss": 0.3687,
596
+ "step": 890
597
+ },
598
+ {
599
+ "epoch": 6.12,
600
+ "learning_rate": 1.9587628865979382e-05,
601
+ "loss": 0.3601,
602
+ "step": 900
603
+ },
604
+ {
605
+ "epoch": 6.19,
606
+ "learning_rate": 1.9243986254295536e-05,
607
+ "loss": 0.3455,
608
+ "step": 910
609
+ },
610
+ {
611
+ "epoch": 6.26,
612
+ "learning_rate": 1.8900343642611683e-05,
613
+ "loss": 0.3983,
614
+ "step": 920
615
+ },
616
+ {
617
+ "epoch": 6.33,
618
+ "learning_rate": 1.8556701030927837e-05,
619
+ "loss": 0.3627,
620
+ "step": 930
621
+ },
622
+ {
623
+ "epoch": 6.39,
624
+ "learning_rate": 1.8213058419243987e-05,
625
+ "loss": 0.3731,
626
+ "step": 940
627
+ },
628
+ {
629
+ "epoch": 6.46,
630
+ "learning_rate": 1.7869415807560138e-05,
631
+ "loss": 0.3517,
632
+ "step": 950
633
+ },
634
+ {
635
+ "epoch": 6.53,
636
+ "learning_rate": 1.7525773195876288e-05,
637
+ "loss": 0.3834,
638
+ "step": 960
639
+ },
640
+ {
641
+ "epoch": 6.6,
642
+ "learning_rate": 1.7182130584192442e-05,
643
+ "loss": 0.3294,
644
+ "step": 970
645
+ },
646
+ {
647
+ "epoch": 6.67,
648
+ "learning_rate": 1.6838487972508592e-05,
649
+ "loss": 0.301,
650
+ "step": 980
651
+ },
652
+ {
653
+ "epoch": 6.73,
654
+ "learning_rate": 1.6494845360824743e-05,
655
+ "loss": 0.3338,
656
+ "step": 990
657
+ },
658
+ {
659
+ "epoch": 6.8,
660
+ "learning_rate": 1.6151202749140896e-05,
661
+ "loss": 0.3375,
662
+ "step": 1000
663
+ },
664
+ {
665
+ "epoch": 6.87,
666
+ "learning_rate": 1.5807560137457044e-05,
667
+ "loss": 0.359,
668
+ "step": 1010
669
+ },
670
+ {
671
+ "epoch": 6.94,
672
+ "learning_rate": 1.5463917525773197e-05,
673
+ "loss": 0.3326,
674
+ "step": 1020
675
+ },
676
+ {
677
+ "epoch": 7.0,
678
+ "eval_accuracy": 0.9650273224043716,
679
+ "eval_loss": 0.1366710066795349,
680
+ "eval_runtime": 26.9372,
681
+ "eval_samples_per_second": 33.968,
682
+ "eval_steps_per_second": 4.269,
683
+ "step": 1029
684
+ },
685
+ {
686
+ "epoch": 7.01,
687
+ "learning_rate": 1.5120274914089346e-05,
688
+ "loss": 0.3346,
689
+ "step": 1030
690
+ },
691
+ {
692
+ "epoch": 7.07,
693
+ "learning_rate": 1.47766323024055e-05,
694
+ "loss": 0.3585,
695
+ "step": 1040
696
+ },
697
+ {
698
+ "epoch": 7.14,
699
+ "learning_rate": 1.4432989690721649e-05,
700
+ "loss": 0.3262,
701
+ "step": 1050
702
+ },
703
+ {
704
+ "epoch": 7.21,
705
+ "learning_rate": 1.40893470790378e-05,
706
+ "loss": 0.329,
707
+ "step": 1060
708
+ },
709
+ {
710
+ "epoch": 7.28,
711
+ "learning_rate": 1.3745704467353954e-05,
712
+ "loss": 0.3365,
713
+ "step": 1070
714
+ },
715
+ {
716
+ "epoch": 7.35,
717
+ "learning_rate": 1.3402061855670103e-05,
718
+ "loss": 0.3348,
719
+ "step": 1080
720
+ },
721
+ {
722
+ "epoch": 7.41,
723
+ "learning_rate": 1.3058419243986255e-05,
724
+ "loss": 0.3666,
725
+ "step": 1090
726
+ },
727
+ {
728
+ "epoch": 7.48,
729
+ "learning_rate": 1.2714776632302406e-05,
730
+ "loss": 0.3044,
731
+ "step": 1100
732
+ },
733
+ {
734
+ "epoch": 7.55,
735
+ "learning_rate": 1.2371134020618558e-05,
736
+ "loss": 0.3048,
737
+ "step": 1110
738
+ },
739
+ {
740
+ "epoch": 7.62,
741
+ "learning_rate": 1.2027491408934708e-05,
742
+ "loss": 0.2838,
743
+ "step": 1120
744
+ },
745
+ {
746
+ "epoch": 7.69,
747
+ "learning_rate": 1.168384879725086e-05,
748
+ "loss": 0.3713,
749
+ "step": 1130
750
+ },
751
+ {
752
+ "epoch": 7.76,
753
+ "learning_rate": 1.134020618556701e-05,
754
+ "loss": 0.3235,
755
+ "step": 1140
756
+ },
757
+ {
758
+ "epoch": 7.82,
759
+ "learning_rate": 1.0996563573883161e-05,
760
+ "loss": 0.3703,
761
+ "step": 1150
762
+ },
763
+ {
764
+ "epoch": 7.89,
765
+ "learning_rate": 1.0652920962199313e-05,
766
+ "loss": 0.3923,
767
+ "step": 1160
768
+ },
769
+ {
770
+ "epoch": 7.96,
771
+ "learning_rate": 1.0309278350515464e-05,
772
+ "loss": 0.3166,
773
+ "step": 1170
774
+ },
775
+ {
776
+ "epoch": 8.0,
777
+ "eval_accuracy": 0.9737704918032787,
778
+ "eval_loss": 0.11094537377357483,
779
+ "eval_runtime": 26.7688,
780
+ "eval_samples_per_second": 34.182,
781
+ "eval_steps_per_second": 4.296,
782
+ "step": 1176
783
+ },
784
+ {
785
+ "epoch": 8.03,
786
+ "learning_rate": 9.965635738831616e-06,
787
+ "loss": 0.3415,
788
+ "step": 1180
789
+ },
790
+ {
791
+ "epoch": 8.1,
792
+ "learning_rate": 9.621993127147768e-06,
793
+ "loss": 0.3692,
794
+ "step": 1190
795
+ },
796
+ {
797
+ "epoch": 8.16,
798
+ "learning_rate": 9.278350515463918e-06,
799
+ "loss": 0.3247,
800
+ "step": 1200
801
+ },
802
+ {
803
+ "epoch": 8.23,
804
+ "learning_rate": 8.934707903780069e-06,
805
+ "loss": 0.2912,
806
+ "step": 1210
807
+ },
808
+ {
809
+ "epoch": 8.3,
810
+ "learning_rate": 8.591065292096221e-06,
811
+ "loss": 0.3685,
812
+ "step": 1220
813
+ },
814
+ {
815
+ "epoch": 8.37,
816
+ "learning_rate": 8.247422680412371e-06,
817
+ "loss": 0.3102,
818
+ "step": 1230
819
+ },
820
+ {
821
+ "epoch": 8.44,
822
+ "learning_rate": 7.903780068728522e-06,
823
+ "loss": 0.3595,
824
+ "step": 1240
825
+ },
826
+ {
827
+ "epoch": 8.5,
828
+ "learning_rate": 7.560137457044673e-06,
829
+ "loss": 0.3045,
830
+ "step": 1250
831
+ },
832
+ {
833
+ "epoch": 8.57,
834
+ "learning_rate": 7.216494845360824e-06,
835
+ "loss": 0.3092,
836
+ "step": 1260
837
+ },
838
+ {
839
+ "epoch": 8.64,
840
+ "learning_rate": 6.872852233676977e-06,
841
+ "loss": 0.3247,
842
+ "step": 1270
843
+ },
844
+ {
845
+ "epoch": 8.71,
846
+ "learning_rate": 6.529209621993128e-06,
847
+ "loss": 0.3052,
848
+ "step": 1280
849
+ },
850
+ {
851
+ "epoch": 8.78,
852
+ "learning_rate": 6.185567010309279e-06,
853
+ "loss": 0.2978,
854
+ "step": 1290
855
+ },
856
+ {
857
+ "epoch": 8.84,
858
+ "learning_rate": 5.84192439862543e-06,
859
+ "loss": 0.3043,
860
+ "step": 1300
861
+ },
862
+ {
863
+ "epoch": 8.91,
864
+ "learning_rate": 5.498281786941581e-06,
865
+ "loss": 0.3189,
866
+ "step": 1310
867
+ },
868
+ {
869
+ "epoch": 8.98,
870
+ "learning_rate": 5.154639175257732e-06,
871
+ "loss": 0.3492,
872
+ "step": 1320
873
+ },
874
+ {
875
+ "epoch": 9.0,
876
+ "eval_accuracy": 0.9759562841530055,
877
+ "eval_loss": 0.1108308881521225,
878
+ "eval_runtime": 26.5259,
879
+ "eval_samples_per_second": 34.495,
880
+ "eval_steps_per_second": 4.335,
881
+ "step": 1323
882
+ },
883
+ {
884
+ "epoch": 9.05,
885
+ "learning_rate": 4.810996563573884e-06,
886
+ "loss": 0.3084,
887
+ "step": 1330
888
+ },
889
+ {
890
+ "epoch": 9.12,
891
+ "learning_rate": 4.467353951890034e-06,
892
+ "loss": 0.2987,
893
+ "step": 1340
894
+ },
895
+ {
896
+ "epoch": 9.18,
897
+ "learning_rate": 4.123711340206186e-06,
898
+ "loss": 0.3322,
899
+ "step": 1350
900
+ },
901
+ {
902
+ "epoch": 9.25,
903
+ "learning_rate": 3.7800687285223365e-06,
904
+ "loss": 0.293,
905
+ "step": 1360
906
+ },
907
+ {
908
+ "epoch": 9.32,
909
+ "learning_rate": 3.4364261168384886e-06,
910
+ "loss": 0.338,
911
+ "step": 1370
912
+ },
913
+ {
914
+ "epoch": 9.39,
915
+ "learning_rate": 3.0927835051546395e-06,
916
+ "loss": 0.304,
917
+ "step": 1380
918
+ },
919
+ {
920
+ "epoch": 9.46,
921
+ "learning_rate": 2.7491408934707903e-06,
922
+ "loss": 0.3074,
923
+ "step": 1390
924
+ },
925
+ {
926
+ "epoch": 9.52,
927
+ "learning_rate": 2.405498281786942e-06,
928
+ "loss": 0.3073,
929
+ "step": 1400
930
+ },
931
+ {
932
+ "epoch": 9.59,
933
+ "learning_rate": 2.061855670103093e-06,
934
+ "loss": 0.2956,
935
+ "step": 1410
936
+ },
937
+ {
938
+ "epoch": 9.66,
939
+ "learning_rate": 1.7182130584192443e-06,
940
+ "loss": 0.3137,
941
+ "step": 1420
942
+ },
943
+ {
944
+ "epoch": 9.73,
945
+ "learning_rate": 1.3745704467353952e-06,
946
+ "loss": 0.3035,
947
+ "step": 1430
948
+ },
949
+ {
950
+ "epoch": 9.8,
951
+ "learning_rate": 1.0309278350515464e-06,
952
+ "loss": 0.3166,
953
+ "step": 1440
954
+ },
955
+ {
956
+ "epoch": 9.86,
957
+ "learning_rate": 6.872852233676976e-07,
958
+ "loss": 0.2696,
959
+ "step": 1450
960
+ },
961
+ {
962
+ "epoch": 9.93,
963
+ "learning_rate": 3.436426116838488e-07,
964
+ "loss": 0.3482,
965
+ "step": 1460
966
+ },
967
+ {
968
+ "epoch": 10.0,
969
+ "learning_rate": 0.0,
970
+ "loss": 0.3228,
971
+ "step": 1470
972
+ },
973
+ {
974
+ "epoch": 10.0,
975
+ "eval_accuracy": 0.9781420765027322,
976
+ "eval_loss": 0.1102537140250206,
977
+ "eval_runtime": 26.664,
978
+ "eval_samples_per_second": 34.316,
979
+ "eval_steps_per_second": 4.313,
980
+ "step": 1470
981
+ },
982
+ {
983
+ "epoch": 10.0,
984
+ "step": 1470,
985
+ "total_flos": 1.7471888881490534e+18,
986
+ "train_loss": 0.49037140356440123,
987
+ "train_runtime": 8760.81,
988
+ "train_samples_per_second": 9.391,
989
+ "train_steps_per_second": 0.168
990
+ }
991
+ ],
992
+ "max_steps": 1470,
993
+ "num_train_epochs": 10,
994
+ "total_flos": 1.7471888881490534e+18,
995
+ "trial_name": null,
996
+ "trial_params": null
997
+ }