ChasingMercer commited on
Commit
031ddf6
1 Parent(s): 7e9bfa9
Files changed (3) hide show
  1. all_results.json +8 -0
  2. train_results.json +8 -0
  3. trainer_state.json +1102 -0
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 4.3513294045912105e+18,
4
+ "train_loss": 0.08035437971183718,
5
+ "train_runtime": 5091.7575,
6
+ "train_samples_per_second": 11.034,
7
+ "train_steps_per_second": 0.345
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 4.3513294045912105e+18,
4
+ "train_loss": 0.08035437971183718,
5
+ "train_runtime": 5091.7575,
6
+ "train_samples_per_second": 11.034,
7
+ "train_steps_per_second": 0.345
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9976505766766339,
3
+ "best_model_checkpoint": "./beit-base/checkpoint-1755",
4
+ "epoch": 2.9995728321230244,
5
+ "global_step": 1755,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 2.840909090909091e-06,
13
+ "loss": 1.0185,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 5.681818181818182e-06,
19
+ "loss": 0.8387,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 8.522727272727273e-06,
25
+ "loss": 0.6429,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.07,
30
+ "learning_rate": 1.1363636363636365e-05,
31
+ "loss": 0.3847,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.09,
36
+ "learning_rate": 1.4204545454545456e-05,
37
+ "loss": 0.206,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.1,
42
+ "learning_rate": 1.7045454545454546e-05,
43
+ "loss": 0.1805,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "learning_rate": 1.9886363636363638e-05,
49
+ "loss": 0.1222,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.14,
54
+ "learning_rate": 2.272727272727273e-05,
55
+ "loss": 0.0944,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.15,
60
+ "learning_rate": 2.5568181818181817e-05,
61
+ "loss": 0.076,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.17,
66
+ "learning_rate": 2.8409090909090912e-05,
67
+ "loss": 0.1085,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.19,
72
+ "learning_rate": 3.125e-05,
73
+ "loss": 0.1311,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.21,
78
+ "learning_rate": 3.409090909090909e-05,
79
+ "loss": 0.0749,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.22,
84
+ "learning_rate": 3.6931818181818184e-05,
85
+ "loss": 0.0745,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "learning_rate": 3.9772727272727275e-05,
91
+ "loss": 0.0682,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.26,
96
+ "learning_rate": 4.261363636363637e-05,
97
+ "loss": 0.0558,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.27,
102
+ "learning_rate": 4.545454545454546e-05,
103
+ "loss": 0.0471,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.29,
108
+ "learning_rate": 4.829545454545455e-05,
109
+ "loss": 0.0662,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.31,
114
+ "learning_rate": 4.987333755541482e-05,
115
+ "loss": 0.1774,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.32,
120
+ "learning_rate": 4.955668144395187e-05,
121
+ "loss": 0.1695,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.34,
126
+ "learning_rate": 4.9240025332488924e-05,
127
+ "loss": 0.0893,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "learning_rate": 4.8923369221025965e-05,
133
+ "loss": 0.096,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.38,
138
+ "learning_rate": 4.860671310956302e-05,
139
+ "loss": 0.0852,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.39,
144
+ "learning_rate": 4.829005699810007e-05,
145
+ "loss": 0.0878,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.41,
150
+ "learning_rate": 4.7973400886637115e-05,
151
+ "loss": 0.1237,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.43,
156
+ "learning_rate": 4.765674477517416e-05,
157
+ "loss": 0.0915,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.44,
162
+ "learning_rate": 4.734008866371121e-05,
163
+ "loss": 0.0439,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 0.46,
168
+ "learning_rate": 4.702343255224826e-05,
169
+ "loss": 0.1507,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "learning_rate": 4.670677644078531e-05,
175
+ "loss": 0.1061,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 0.5,
180
+ "learning_rate": 4.6390120329322354e-05,
181
+ "loss": 0.0967,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 0.51,
186
+ "learning_rate": 4.607346421785941e-05,
187
+ "loss": 0.0994,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 0.53,
192
+ "learning_rate": 4.5756808106396456e-05,
193
+ "loss": 0.0976,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 0.55,
198
+ "learning_rate": 4.5440151994933504e-05,
199
+ "loss": 0.1442,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 0.56,
204
+ "learning_rate": 4.512349588347055e-05,
205
+ "loss": 0.0735,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 0.58,
210
+ "learning_rate": 4.48068397720076e-05,
211
+ "loss": 0.0693,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "learning_rate": 4.449018366054465e-05,
217
+ "loss": 0.0748,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 0.62,
222
+ "learning_rate": 4.41735275490817e-05,
223
+ "loss": 0.0585,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 0.63,
228
+ "learning_rate": 4.385687143761874e-05,
229
+ "loss": 0.0963,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 0.65,
234
+ "learning_rate": 4.35402153261558e-05,
235
+ "loss": 0.0526,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 0.67,
240
+ "learning_rate": 4.3223559214692845e-05,
241
+ "loss": 0.0852,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 0.68,
246
+ "learning_rate": 4.290690310322989e-05,
247
+ "loss": 0.0973,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 0.7,
252
+ "learning_rate": 4.259024699176694e-05,
253
+ "loss": 0.0818,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "learning_rate": 4.2273590880303995e-05,
259
+ "loss": 0.1427,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 0.73,
264
+ "learning_rate": 4.1956934768841036e-05,
265
+ "loss": 0.1151,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 0.75,
270
+ "learning_rate": 4.164027865737809e-05,
271
+ "loss": 0.0536,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 0.77,
276
+ "learning_rate": 4.132362254591514e-05,
277
+ "loss": 0.0959,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 0.79,
282
+ "learning_rate": 4.1006966434452187e-05,
283
+ "loss": 0.0641,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 0.8,
288
+ "learning_rate": 4.0690310322989234e-05,
289
+ "loss": 0.0931,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 0.82,
294
+ "learning_rate": 4.037365421152628e-05,
295
+ "loss": 0.0418,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "learning_rate": 4.005699810006333e-05,
301
+ "loss": 0.0581,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 0.85,
306
+ "learning_rate": 3.9740341988600384e-05,
307
+ "loss": 0.0573,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 0.87,
312
+ "learning_rate": 3.9423685877137425e-05,
313
+ "loss": 0.0911,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 0.89,
318
+ "learning_rate": 3.910702976567448e-05,
319
+ "loss": 0.1501,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 0.91,
324
+ "learning_rate": 3.879037365421153e-05,
325
+ "loss": 0.0979,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 0.92,
330
+ "learning_rate": 3.8473717542748576e-05,
331
+ "loss": 0.0737,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 0.94,
336
+ "learning_rate": 3.8157061431285623e-05,
337
+ "loss": 0.0443,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "learning_rate": 3.784040531982268e-05,
343
+ "loss": 0.0838,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 0.97,
348
+ "learning_rate": 3.752374920835972e-05,
349
+ "loss": 0.0664,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 0.99,
354
+ "learning_rate": 3.7207093096896774e-05,
355
+ "loss": 0.0303,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 1.0,
360
+ "eval_accuracy": 0.9942332336608287,
361
+ "eval_loss": 0.01858363300561905,
362
+ "eval_runtime": 148.0695,
363
+ "eval_samples_per_second": 31.62,
364
+ "eval_steps_per_second": 3.958,
365
+ "step": 585
366
+ },
367
+ {
368
+ "epoch": 1.01,
369
+ "learning_rate": 3.689043698543382e-05,
370
+ "loss": 0.0769,
371
+ "step": 590
372
+ },
373
+ {
374
+ "epoch": 1.03,
375
+ "learning_rate": 3.657378087397087e-05,
376
+ "loss": 0.0621,
377
+ "step": 600
378
+ },
379
+ {
380
+ "epoch": 1.04,
381
+ "learning_rate": 3.625712476250792e-05,
382
+ "loss": 0.0723,
383
+ "step": 610
384
+ },
385
+ {
386
+ "epoch": 1.06,
387
+ "learning_rate": 3.5940468651044965e-05,
388
+ "loss": 0.0851,
389
+ "step": 620
390
+ },
391
+ {
392
+ "epoch": 1.08,
393
+ "learning_rate": 3.562381253958201e-05,
394
+ "loss": 0.0889,
395
+ "step": 630
396
+ },
397
+ {
398
+ "epoch": 1.09,
399
+ "learning_rate": 3.530715642811907e-05,
400
+ "loss": 0.0799,
401
+ "step": 640
402
+ },
403
+ {
404
+ "epoch": 1.11,
405
+ "learning_rate": 3.499050031665611e-05,
406
+ "loss": 0.0558,
407
+ "step": 650
408
+ },
409
+ {
410
+ "epoch": 1.13,
411
+ "learning_rate": 3.467384420519316e-05,
412
+ "loss": 0.0548,
413
+ "step": 660
414
+ },
415
+ {
416
+ "epoch": 1.15,
417
+ "learning_rate": 3.435718809373021e-05,
418
+ "loss": 0.0844,
419
+ "step": 670
420
+ },
421
+ {
422
+ "epoch": 1.16,
423
+ "learning_rate": 3.404053198226726e-05,
424
+ "loss": 0.0374,
425
+ "step": 680
426
+ },
427
+ {
428
+ "epoch": 1.18,
429
+ "learning_rate": 3.3723875870804306e-05,
430
+ "loss": 0.0829,
431
+ "step": 690
432
+ },
433
+ {
434
+ "epoch": 1.2,
435
+ "learning_rate": 3.340721975934136e-05,
436
+ "loss": 0.0581,
437
+ "step": 700
438
+ },
439
+ {
440
+ "epoch": 1.21,
441
+ "learning_rate": 3.30905636478784e-05,
442
+ "loss": 0.073,
443
+ "step": 710
444
+ },
445
+ {
446
+ "epoch": 1.23,
447
+ "learning_rate": 3.2773907536415456e-05,
448
+ "loss": 0.0713,
449
+ "step": 720
450
+ },
451
+ {
452
+ "epoch": 1.25,
453
+ "learning_rate": 3.2457251424952504e-05,
454
+ "loss": 0.0627,
455
+ "step": 730
456
+ },
457
+ {
458
+ "epoch": 1.26,
459
+ "learning_rate": 3.214059531348955e-05,
460
+ "loss": 0.1106,
461
+ "step": 740
462
+ },
463
+ {
464
+ "epoch": 1.28,
465
+ "learning_rate": 3.18239392020266e-05,
466
+ "loss": 0.0838,
467
+ "step": 750
468
+ },
469
+ {
470
+ "epoch": 1.3,
471
+ "learning_rate": 3.150728309056365e-05,
472
+ "loss": 0.0921,
473
+ "step": 760
474
+ },
475
+ {
476
+ "epoch": 1.32,
477
+ "learning_rate": 3.1190626979100695e-05,
478
+ "loss": 0.0262,
479
+ "step": 770
480
+ },
481
+ {
482
+ "epoch": 1.33,
483
+ "learning_rate": 3.087397086763775e-05,
484
+ "loss": 0.0545,
485
+ "step": 780
486
+ },
487
+ {
488
+ "epoch": 1.35,
489
+ "learning_rate": 3.055731475617479e-05,
490
+ "loss": 0.0867,
491
+ "step": 790
492
+ },
493
+ {
494
+ "epoch": 1.37,
495
+ "learning_rate": 3.0240658644711845e-05,
496
+ "loss": 0.091,
497
+ "step": 800
498
+ },
499
+ {
500
+ "epoch": 1.38,
501
+ "learning_rate": 2.992400253324889e-05,
502
+ "loss": 0.0702,
503
+ "step": 810
504
+ },
505
+ {
506
+ "epoch": 1.4,
507
+ "learning_rate": 2.960734642178594e-05,
508
+ "loss": 0.0888,
509
+ "step": 820
510
+ },
511
+ {
512
+ "epoch": 1.42,
513
+ "learning_rate": 2.929069031032299e-05,
514
+ "loss": 0.0514,
515
+ "step": 830
516
+ },
517
+ {
518
+ "epoch": 1.44,
519
+ "learning_rate": 2.897403419886004e-05,
520
+ "loss": 0.0762,
521
+ "step": 840
522
+ },
523
+ {
524
+ "epoch": 1.45,
525
+ "learning_rate": 2.8657378087397084e-05,
526
+ "loss": 0.0192,
527
+ "step": 850
528
+ },
529
+ {
530
+ "epoch": 1.47,
531
+ "learning_rate": 2.834072197593414e-05,
532
+ "loss": 0.0337,
533
+ "step": 860
534
+ },
535
+ {
536
+ "epoch": 1.49,
537
+ "learning_rate": 2.8024065864471183e-05,
538
+ "loss": 0.0634,
539
+ "step": 870
540
+ },
541
+ {
542
+ "epoch": 1.5,
543
+ "learning_rate": 2.7707409753008234e-05,
544
+ "loss": 0.0435,
545
+ "step": 880
546
+ },
547
+ {
548
+ "epoch": 1.52,
549
+ "learning_rate": 2.7390753641545285e-05,
550
+ "loss": 0.0685,
551
+ "step": 890
552
+ },
553
+ {
554
+ "epoch": 1.54,
555
+ "learning_rate": 2.7074097530082333e-05,
556
+ "loss": 0.1016,
557
+ "step": 900
558
+ },
559
+ {
560
+ "epoch": 1.56,
561
+ "learning_rate": 2.6757441418619384e-05,
562
+ "loss": 0.0797,
563
+ "step": 910
564
+ },
565
+ {
566
+ "epoch": 1.57,
567
+ "learning_rate": 2.644078530715643e-05,
568
+ "loss": 0.0946,
569
+ "step": 920
570
+ },
571
+ {
572
+ "epoch": 1.59,
573
+ "learning_rate": 2.612412919569348e-05,
574
+ "loss": 0.0317,
575
+ "step": 930
576
+ },
577
+ {
578
+ "epoch": 1.61,
579
+ "learning_rate": 2.5807473084230528e-05,
580
+ "loss": 0.0673,
581
+ "step": 940
582
+ },
583
+ {
584
+ "epoch": 1.62,
585
+ "learning_rate": 2.549081697276758e-05,
586
+ "loss": 0.0287,
587
+ "step": 950
588
+ },
589
+ {
590
+ "epoch": 1.64,
591
+ "learning_rate": 2.5174160861304623e-05,
592
+ "loss": 0.0273,
593
+ "step": 960
594
+ },
595
+ {
596
+ "epoch": 1.66,
597
+ "learning_rate": 2.4857504749841674e-05,
598
+ "loss": 0.0404,
599
+ "step": 970
600
+ },
601
+ {
602
+ "epoch": 1.67,
603
+ "learning_rate": 2.4540848638378722e-05,
604
+ "loss": 0.0524,
605
+ "step": 980
606
+ },
607
+ {
608
+ "epoch": 1.69,
609
+ "learning_rate": 2.422419252691577e-05,
610
+ "loss": 0.0372,
611
+ "step": 990
612
+ },
613
+ {
614
+ "epoch": 1.71,
615
+ "learning_rate": 2.3907536415452818e-05,
616
+ "loss": 0.0706,
617
+ "step": 1000
618
+ },
619
+ {
620
+ "epoch": 1.73,
621
+ "learning_rate": 2.359088030398987e-05,
622
+ "loss": 0.0451,
623
+ "step": 1010
624
+ },
625
+ {
626
+ "epoch": 1.74,
627
+ "learning_rate": 2.3274224192526917e-05,
628
+ "loss": 0.0583,
629
+ "step": 1020
630
+ },
631
+ {
632
+ "epoch": 1.76,
633
+ "learning_rate": 2.2957568081063964e-05,
634
+ "loss": 0.0233,
635
+ "step": 1030
636
+ },
637
+ {
638
+ "epoch": 1.78,
639
+ "learning_rate": 2.2640911969601016e-05,
640
+ "loss": 0.0509,
641
+ "step": 1040
642
+ },
643
+ {
644
+ "epoch": 1.79,
645
+ "learning_rate": 2.2324255858138063e-05,
646
+ "loss": 0.0619,
647
+ "step": 1050
648
+ },
649
+ {
650
+ "epoch": 1.81,
651
+ "learning_rate": 2.200759974667511e-05,
652
+ "loss": 0.0945,
653
+ "step": 1060
654
+ },
655
+ {
656
+ "epoch": 1.83,
657
+ "learning_rate": 2.169094363521216e-05,
658
+ "loss": 0.0151,
659
+ "step": 1070
660
+ },
661
+ {
662
+ "epoch": 1.85,
663
+ "learning_rate": 2.137428752374921e-05,
664
+ "loss": 0.0675,
665
+ "step": 1080
666
+ },
667
+ {
668
+ "epoch": 1.86,
669
+ "learning_rate": 2.1057631412286258e-05,
670
+ "loss": 0.0548,
671
+ "step": 1090
672
+ },
673
+ {
674
+ "epoch": 1.88,
675
+ "learning_rate": 2.0740975300823306e-05,
676
+ "loss": 0.0327,
677
+ "step": 1100
678
+ },
679
+ {
680
+ "epoch": 1.9,
681
+ "learning_rate": 2.0424319189360357e-05,
682
+ "loss": 0.1223,
683
+ "step": 1110
684
+ },
685
+ {
686
+ "epoch": 1.91,
687
+ "learning_rate": 2.0107663077897405e-05,
688
+ "loss": 0.1124,
689
+ "step": 1120
690
+ },
691
+ {
692
+ "epoch": 1.93,
693
+ "learning_rate": 1.9791006966434452e-05,
694
+ "loss": 0.0497,
695
+ "step": 1130
696
+ },
697
+ {
698
+ "epoch": 1.95,
699
+ "learning_rate": 1.94743508549715e-05,
700
+ "loss": 0.0703,
701
+ "step": 1140
702
+ },
703
+ {
704
+ "epoch": 1.97,
705
+ "learning_rate": 1.915769474350855e-05,
706
+ "loss": 0.0212,
707
+ "step": 1150
708
+ },
709
+ {
710
+ "epoch": 1.98,
711
+ "learning_rate": 1.88410386320456e-05,
712
+ "loss": 0.1229,
713
+ "step": 1160
714
+ },
715
+ {
716
+ "epoch": 2.0,
717
+ "learning_rate": 1.8524382520582647e-05,
718
+ "loss": 0.0374,
719
+ "step": 1170
720
+ },
721
+ {
722
+ "epoch": 2.0,
723
+ "eval_accuracy": 0.9955147372917557,
724
+ "eval_loss": 0.015015287324786186,
725
+ "eval_runtime": 154.7173,
726
+ "eval_samples_per_second": 30.262,
727
+ "eval_steps_per_second": 3.788,
728
+ "step": 1170
729
+ },
730
+ {
731
+ "epoch": 2.02,
732
+ "learning_rate": 1.8207726409119695e-05,
733
+ "loss": 0.0663,
734
+ "step": 1180
735
+ },
736
+ {
737
+ "epoch": 2.03,
738
+ "learning_rate": 1.7891070297656746e-05,
739
+ "loss": 0.0235,
740
+ "step": 1190
741
+ },
742
+ {
743
+ "epoch": 2.05,
744
+ "learning_rate": 1.7574414186193794e-05,
745
+ "loss": 0.0491,
746
+ "step": 1200
747
+ },
748
+ {
749
+ "epoch": 2.07,
750
+ "learning_rate": 1.725775807473084e-05,
751
+ "loss": 0.0361,
752
+ "step": 1210
753
+ },
754
+ {
755
+ "epoch": 2.09,
756
+ "learning_rate": 1.6941101963267893e-05,
757
+ "loss": 0.0354,
758
+ "step": 1220
759
+ },
760
+ {
761
+ "epoch": 2.1,
762
+ "learning_rate": 1.662444585180494e-05,
763
+ "loss": 0.0222,
764
+ "step": 1230
765
+ },
766
+ {
767
+ "epoch": 2.12,
768
+ "learning_rate": 1.6307789740341988e-05,
769
+ "loss": 0.0769,
770
+ "step": 1240
771
+ },
772
+ {
773
+ "epoch": 2.14,
774
+ "learning_rate": 1.5991133628879036e-05,
775
+ "loss": 0.0423,
776
+ "step": 1250
777
+ },
778
+ {
779
+ "epoch": 2.15,
780
+ "learning_rate": 1.5674477517416087e-05,
781
+ "loss": 0.0301,
782
+ "step": 1260
783
+ },
784
+ {
785
+ "epoch": 2.17,
786
+ "learning_rate": 1.5357821405953135e-05,
787
+ "loss": 0.0092,
788
+ "step": 1270
789
+ },
790
+ {
791
+ "epoch": 2.19,
792
+ "learning_rate": 1.5041165294490185e-05,
793
+ "loss": 0.0443,
794
+ "step": 1280
795
+ },
796
+ {
797
+ "epoch": 2.21,
798
+ "learning_rate": 1.4724509183027232e-05,
799
+ "loss": 0.0504,
800
+ "step": 1290
801
+ },
802
+ {
803
+ "epoch": 2.22,
804
+ "learning_rate": 1.4407853071564282e-05,
805
+ "loss": 0.0229,
806
+ "step": 1300
807
+ },
808
+ {
809
+ "epoch": 2.24,
810
+ "learning_rate": 1.409119696010133e-05,
811
+ "loss": 0.077,
812
+ "step": 1310
813
+ },
814
+ {
815
+ "epoch": 2.26,
816
+ "learning_rate": 1.3774540848638379e-05,
817
+ "loss": 0.0168,
818
+ "step": 1320
819
+ },
820
+ {
821
+ "epoch": 2.27,
822
+ "learning_rate": 1.3457884737175427e-05,
823
+ "loss": 0.0509,
824
+ "step": 1330
825
+ },
826
+ {
827
+ "epoch": 2.29,
828
+ "learning_rate": 1.3141228625712476e-05,
829
+ "loss": 0.1056,
830
+ "step": 1340
831
+ },
832
+ {
833
+ "epoch": 2.31,
834
+ "learning_rate": 1.2824572514249524e-05,
835
+ "loss": 0.0267,
836
+ "step": 1350
837
+ },
838
+ {
839
+ "epoch": 2.32,
840
+ "learning_rate": 1.2507916402786574e-05,
841
+ "loss": 0.0302,
842
+ "step": 1360
843
+ },
844
+ {
845
+ "epoch": 2.34,
846
+ "learning_rate": 1.2191260291323623e-05,
847
+ "loss": 0.067,
848
+ "step": 1370
849
+ },
850
+ {
851
+ "epoch": 2.36,
852
+ "learning_rate": 1.1874604179860673e-05,
853
+ "loss": 0.0344,
854
+ "step": 1380
855
+ },
856
+ {
857
+ "epoch": 2.38,
858
+ "learning_rate": 1.1557948068397722e-05,
859
+ "loss": 0.0408,
860
+ "step": 1390
861
+ },
862
+ {
863
+ "epoch": 2.39,
864
+ "learning_rate": 1.124129195693477e-05,
865
+ "loss": 0.0313,
866
+ "step": 1400
867
+ },
868
+ {
869
+ "epoch": 2.41,
870
+ "learning_rate": 1.092463584547182e-05,
871
+ "loss": 0.0065,
872
+ "step": 1410
873
+ },
874
+ {
875
+ "epoch": 2.43,
876
+ "learning_rate": 1.0607979734008867e-05,
877
+ "loss": 0.0507,
878
+ "step": 1420
879
+ },
880
+ {
881
+ "epoch": 2.44,
882
+ "learning_rate": 1.0291323622545917e-05,
883
+ "loss": 0.0182,
884
+ "step": 1430
885
+ },
886
+ {
887
+ "epoch": 2.46,
888
+ "learning_rate": 9.974667511082964e-06,
889
+ "loss": 0.0413,
890
+ "step": 1440
891
+ },
892
+ {
893
+ "epoch": 2.48,
894
+ "learning_rate": 9.658011399620014e-06,
895
+ "loss": 0.0626,
896
+ "step": 1450
897
+ },
898
+ {
899
+ "epoch": 2.5,
900
+ "learning_rate": 9.341355288157062e-06,
901
+ "loss": 0.0222,
902
+ "step": 1460
903
+ },
904
+ {
905
+ "epoch": 2.51,
906
+ "learning_rate": 9.024699176694111e-06,
907
+ "loss": 0.0529,
908
+ "step": 1470
909
+ },
910
+ {
911
+ "epoch": 2.53,
912
+ "learning_rate": 8.70804306523116e-06,
913
+ "loss": 0.0042,
914
+ "step": 1480
915
+ },
916
+ {
917
+ "epoch": 2.55,
918
+ "learning_rate": 8.391386953768208e-06,
919
+ "loss": 0.051,
920
+ "step": 1490
921
+ },
922
+ {
923
+ "epoch": 2.56,
924
+ "learning_rate": 8.074730842305258e-06,
925
+ "loss": 0.0377,
926
+ "step": 1500
927
+ },
928
+ {
929
+ "epoch": 2.58,
930
+ "learning_rate": 7.758074730842306e-06,
931
+ "loss": 0.0639,
932
+ "step": 1510
933
+ },
934
+ {
935
+ "epoch": 2.6,
936
+ "learning_rate": 7.441418619379355e-06,
937
+ "loss": 0.0532,
938
+ "step": 1520
939
+ },
940
+ {
941
+ "epoch": 2.62,
942
+ "learning_rate": 7.124762507916404e-06,
943
+ "loss": 0.0601,
944
+ "step": 1530
945
+ },
946
+ {
947
+ "epoch": 2.63,
948
+ "learning_rate": 6.808106396453452e-06,
949
+ "loss": 0.0118,
950
+ "step": 1540
951
+ },
952
+ {
953
+ "epoch": 2.65,
954
+ "learning_rate": 6.491450284990501e-06,
955
+ "loss": 0.069,
956
+ "step": 1550
957
+ },
958
+ {
959
+ "epoch": 2.67,
960
+ "learning_rate": 6.1747941735275496e-06,
961
+ "loss": 0.1187,
962
+ "step": 1560
963
+ },
964
+ {
965
+ "epoch": 2.68,
966
+ "learning_rate": 5.858138062064598e-06,
967
+ "loss": 0.0612,
968
+ "step": 1570
969
+ },
970
+ {
971
+ "epoch": 2.7,
972
+ "learning_rate": 5.541481950601647e-06,
973
+ "loss": 0.0387,
974
+ "step": 1580
975
+ },
976
+ {
977
+ "epoch": 2.72,
978
+ "learning_rate": 5.2248258391386955e-06,
979
+ "loss": 0.0172,
980
+ "step": 1590
981
+ },
982
+ {
983
+ "epoch": 2.73,
984
+ "learning_rate": 4.908169727675745e-06,
985
+ "loss": 0.0591,
986
+ "step": 1600
987
+ },
988
+ {
989
+ "epoch": 2.75,
990
+ "learning_rate": 4.5915136162127936e-06,
991
+ "loss": 0.0594,
992
+ "step": 1610
993
+ },
994
+ {
995
+ "epoch": 2.77,
996
+ "learning_rate": 4.274857504749842e-06,
997
+ "loss": 0.0249,
998
+ "step": 1620
999
+ },
1000
+ {
1001
+ "epoch": 2.79,
1002
+ "learning_rate": 3.958201393286891e-06,
1003
+ "loss": 0.0707,
1004
+ "step": 1630
1005
+ },
1006
+ {
1007
+ "epoch": 2.8,
1008
+ "learning_rate": 3.6415452818239395e-06,
1009
+ "loss": 0.0472,
1010
+ "step": 1640
1011
+ },
1012
+ {
1013
+ "epoch": 2.82,
1014
+ "learning_rate": 3.324889170360988e-06,
1015
+ "loss": 0.0371,
1016
+ "step": 1650
1017
+ },
1018
+ {
1019
+ "epoch": 2.84,
1020
+ "learning_rate": 3.008233058898037e-06,
1021
+ "loss": 0.0076,
1022
+ "step": 1660
1023
+ },
1024
+ {
1025
+ "epoch": 2.85,
1026
+ "learning_rate": 2.6915769474350858e-06,
1027
+ "loss": 0.0263,
1028
+ "step": 1670
1029
+ },
1030
+ {
1031
+ "epoch": 2.87,
1032
+ "learning_rate": 2.3749208359721344e-06,
1033
+ "loss": 0.0256,
1034
+ "step": 1680
1035
+ },
1036
+ {
1037
+ "epoch": 2.89,
1038
+ "learning_rate": 2.058264724509183e-06,
1039
+ "loss": 0.0349,
1040
+ "step": 1690
1041
+ },
1042
+ {
1043
+ "epoch": 2.91,
1044
+ "learning_rate": 1.7416086130462319e-06,
1045
+ "loss": 0.0315,
1046
+ "step": 1700
1047
+ },
1048
+ {
1049
+ "epoch": 2.92,
1050
+ "learning_rate": 1.4249525015832807e-06,
1051
+ "loss": 0.0217,
1052
+ "step": 1710
1053
+ },
1054
+ {
1055
+ "epoch": 2.94,
1056
+ "learning_rate": 1.1082963901203294e-06,
1057
+ "loss": 0.0409,
1058
+ "step": 1720
1059
+ },
1060
+ {
1061
+ "epoch": 2.96,
1062
+ "learning_rate": 7.916402786573782e-07,
1063
+ "loss": 0.0194,
1064
+ "step": 1730
1065
+ },
1066
+ {
1067
+ "epoch": 2.97,
1068
+ "learning_rate": 4.749841671944269e-07,
1069
+ "loss": 0.0258,
1070
+ "step": 1740
1071
+ },
1072
+ {
1073
+ "epoch": 2.99,
1074
+ "learning_rate": 1.5832805573147563e-07,
1075
+ "loss": 0.0559,
1076
+ "step": 1750
1077
+ },
1078
+ {
1079
+ "epoch": 3.0,
1080
+ "eval_accuracy": 0.9976505766766339,
1081
+ "eval_loss": 0.011637404561042786,
1082
+ "eval_runtime": 150.0146,
1083
+ "eval_samples_per_second": 31.21,
1084
+ "eval_steps_per_second": 3.906,
1085
+ "step": 1755
1086
+ },
1087
+ {
1088
+ "epoch": 3.0,
1089
+ "step": 1755,
1090
+ "total_flos": 4.3513294045912105e+18,
1091
+ "train_loss": 0.08035437971183718,
1092
+ "train_runtime": 5091.7575,
1093
+ "train_samples_per_second": 11.034,
1094
+ "train_steps_per_second": 0.345
1095
+ }
1096
+ ],
1097
+ "max_steps": 1755,
1098
+ "num_train_epochs": 3,
1099
+ "total_flos": 4.3513294045912105e+18,
1100
+ "trial_name": null,
1101
+ "trial_params": null
1102
+ }