siddharth963 commited on
Commit
df686ff
1 Parent(s): 147ae51

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "eval_accuracy": 0.8855140186915887,
4
+ "eval_loss": 0.3418597877025604,
5
+ "eval_runtime": 84.9732,
6
+ "eval_samples_per_second": 50.369,
7
+ "eval_steps_per_second": 1.577,
8
+ "total_flos": 1.3257453564799912e+19,
9
+ "train_loss": 0.4158173976984239,
10
+ "train_runtime": 6524.231,
11
+ "train_samples_per_second": 26.236,
12
+ "train_steps_per_second": 0.204
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "eval_accuracy": 0.8855140186915887,
4
+ "eval_loss": 0.3418597877025604,
5
+ "eval_runtime": 84.9732,
6
+ "eval_samples_per_second": 50.369,
7
+ "eval_steps_per_second": 1.577
8
+ }
runs/Oct12_17-37-40_bf41a5613949/events.out.tfevents.1665603337.bf41a5613949.17.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2c16782de112e9efc9ab3779cddaf855cbcc035567589035dabb2a707beff01
3
+ size 363
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.99,
3
+ "total_flos": 1.3257453564799912e+19,
4
+ "train_loss": 0.4158173976984239,
5
+ "train_runtime": 6524.231,
6
+ "train_samples_per_second": 26.236,
7
+ "train_steps_per_second": 0.204
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,913 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8855140186915887,
3
+ "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-cassava3/checkpoint-1064",
4
+ "epoch": 9.994392523364485,
5
+ "global_step": 1330,
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": 3.7593984962406014e-06,
13
+ "loss": 1.5873,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.15,
18
+ "learning_rate": 7.518796992481203e-06,
19
+ "loss": 1.5448,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.22,
24
+ "learning_rate": 1.1278195488721805e-05,
25
+ "loss": 1.4433,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.3,
30
+ "learning_rate": 1.5037593984962406e-05,
31
+ "loss": 1.2664,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.37,
36
+ "learning_rate": 1.8796992481203007e-05,
37
+ "loss": 1.1417,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.45,
42
+ "learning_rate": 2.255639097744361e-05,
43
+ "loss": 1.0343,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.52,
48
+ "learning_rate": 2.6315789473684212e-05,
49
+ "loss": 0.9283,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.6,
54
+ "learning_rate": 3.007518796992481e-05,
55
+ "loss": 0.8431,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.67,
60
+ "learning_rate": 3.3834586466165414e-05,
61
+ "loss": 0.764,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.75,
66
+ "learning_rate": 3.759398496240601e-05,
67
+ "loss": 0.7234,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.82,
72
+ "learning_rate": 4.135338345864662e-05,
73
+ "loss": 0.6497,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.9,
78
+ "learning_rate": 4.511278195488722e-05,
79
+ "loss": 0.6193,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.97,
84
+ "learning_rate": 4.887218045112782e-05,
85
+ "loss": 0.5624,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.99,
90
+ "eval_accuracy": 0.8165887850467289,
91
+ "eval_loss": 0.5866310596466064,
92
+ "eval_runtime": 93.7262,
93
+ "eval_samples_per_second": 45.665,
94
+ "eval_steps_per_second": 1.43,
95
+ "step": 133
96
+ },
97
+ {
98
+ "epoch": 1.05,
99
+ "learning_rate": 4.970760233918128e-05,
100
+ "loss": 0.6142,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 1.13,
105
+ "learning_rate": 4.928989139515455e-05,
106
+ "loss": 0.5625,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.2,
111
+ "learning_rate": 4.887218045112782e-05,
112
+ "loss": 0.5064,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.28,
117
+ "learning_rate": 4.8454469507101085e-05,
118
+ "loss": 0.5359,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.35,
123
+ "learning_rate": 4.803675856307435e-05,
124
+ "loss": 0.4852,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.43,
129
+ "learning_rate": 4.761904761904762e-05,
130
+ "loss": 0.4538,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.5,
135
+ "learning_rate": 4.720133667502089e-05,
136
+ "loss": 0.4947,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.58,
141
+ "learning_rate": 4.678362573099415e-05,
142
+ "loss": 0.4568,
143
+ "step": 210
144
+ },
145
+ {
146
+ "epoch": 1.65,
147
+ "learning_rate": 4.6365914786967416e-05,
148
+ "loss": 0.4609,
149
+ "step": 220
150
+ },
151
+ {
152
+ "epoch": 1.73,
153
+ "learning_rate": 4.5948203842940684e-05,
154
+ "loss": 0.4919,
155
+ "step": 230
156
+ },
157
+ {
158
+ "epoch": 1.8,
159
+ "learning_rate": 4.553049289891395e-05,
160
+ "loss": 0.5054,
161
+ "step": 240
162
+ },
163
+ {
164
+ "epoch": 1.87,
165
+ "learning_rate": 4.511278195488722e-05,
166
+ "loss": 0.4133,
167
+ "step": 250
168
+ },
169
+ {
170
+ "epoch": 1.95,
171
+ "learning_rate": 4.4695071010860486e-05,
172
+ "loss": 0.4717,
173
+ "step": 260
174
+ },
175
+ {
176
+ "epoch": 1.99,
177
+ "eval_accuracy": 0.8691588785046729,
178
+ "eval_loss": 0.4245365262031555,
179
+ "eval_runtime": 92.5089,
180
+ "eval_samples_per_second": 46.266,
181
+ "eval_steps_per_second": 1.449,
182
+ "step": 266
183
+ },
184
+ {
185
+ "epoch": 2.03,
186
+ "learning_rate": 4.4277360066833754e-05,
187
+ "loss": 0.4125,
188
+ "step": 270
189
+ },
190
+ {
191
+ "epoch": 2.1,
192
+ "learning_rate": 4.3859649122807014e-05,
193
+ "loss": 0.4693,
194
+ "step": 280
195
+ },
196
+ {
197
+ "epoch": 2.18,
198
+ "learning_rate": 4.344193817878028e-05,
199
+ "loss": 0.444,
200
+ "step": 290
201
+ },
202
+ {
203
+ "epoch": 2.25,
204
+ "learning_rate": 4.302422723475355e-05,
205
+ "loss": 0.376,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.33,
210
+ "learning_rate": 4.260651629072682e-05,
211
+ "loss": 0.3967,
212
+ "step": 310
213
+ },
214
+ {
215
+ "epoch": 2.4,
216
+ "learning_rate": 4.2188805346700084e-05,
217
+ "loss": 0.4269,
218
+ "step": 320
219
+ },
220
+ {
221
+ "epoch": 2.48,
222
+ "learning_rate": 4.177109440267335e-05,
223
+ "loss": 0.4132,
224
+ "step": 330
225
+ },
226
+ {
227
+ "epoch": 2.55,
228
+ "learning_rate": 4.135338345864662e-05,
229
+ "loss": 0.4072,
230
+ "step": 340
231
+ },
232
+ {
233
+ "epoch": 2.63,
234
+ "learning_rate": 4.093567251461988e-05,
235
+ "loss": 0.4125,
236
+ "step": 350
237
+ },
238
+ {
239
+ "epoch": 2.7,
240
+ "learning_rate": 4.051796157059315e-05,
241
+ "loss": 0.4108,
242
+ "step": 360
243
+ },
244
+ {
245
+ "epoch": 2.78,
246
+ "learning_rate": 4.0100250626566415e-05,
247
+ "loss": 0.4008,
248
+ "step": 370
249
+ },
250
+ {
251
+ "epoch": 2.85,
252
+ "learning_rate": 3.968253968253968e-05,
253
+ "loss": 0.3809,
254
+ "step": 380
255
+ },
256
+ {
257
+ "epoch": 2.93,
258
+ "learning_rate": 3.926482873851295e-05,
259
+ "loss": 0.4105,
260
+ "step": 390
261
+ },
262
+ {
263
+ "epoch": 2.99,
264
+ "eval_accuracy": 0.8810747663551401,
265
+ "eval_loss": 0.37081509828567505,
266
+ "eval_runtime": 92.7737,
267
+ "eval_samples_per_second": 46.134,
268
+ "eval_steps_per_second": 1.444,
269
+ "step": 399
270
+ },
271
+ {
272
+ "epoch": 3.01,
273
+ "learning_rate": 3.884711779448622e-05,
274
+ "loss": 0.4151,
275
+ "step": 400
276
+ },
277
+ {
278
+ "epoch": 3.08,
279
+ "learning_rate": 3.8429406850459485e-05,
280
+ "loss": 0.3712,
281
+ "step": 410
282
+ },
283
+ {
284
+ "epoch": 3.16,
285
+ "learning_rate": 3.8011695906432746e-05,
286
+ "loss": 0.395,
287
+ "step": 420
288
+ },
289
+ {
290
+ "epoch": 3.23,
291
+ "learning_rate": 3.759398496240601e-05,
292
+ "loss": 0.3889,
293
+ "step": 430
294
+ },
295
+ {
296
+ "epoch": 3.31,
297
+ "learning_rate": 3.717627401837928e-05,
298
+ "loss": 0.343,
299
+ "step": 440
300
+ },
301
+ {
302
+ "epoch": 3.38,
303
+ "learning_rate": 3.675856307435255e-05,
304
+ "loss": 0.3729,
305
+ "step": 450
306
+ },
307
+ {
308
+ "epoch": 3.46,
309
+ "learning_rate": 3.6340852130325816e-05,
310
+ "loss": 0.4083,
311
+ "step": 460
312
+ },
313
+ {
314
+ "epoch": 3.53,
315
+ "learning_rate": 3.592314118629908e-05,
316
+ "loss": 0.4003,
317
+ "step": 470
318
+ },
319
+ {
320
+ "epoch": 3.61,
321
+ "learning_rate": 3.5505430242272344e-05,
322
+ "loss": 0.379,
323
+ "step": 480
324
+ },
325
+ {
326
+ "epoch": 3.68,
327
+ "learning_rate": 3.508771929824561e-05,
328
+ "loss": 0.3461,
329
+ "step": 490
330
+ },
331
+ {
332
+ "epoch": 3.76,
333
+ "learning_rate": 3.467000835421888e-05,
334
+ "loss": 0.3892,
335
+ "step": 500
336
+ },
337
+ {
338
+ "epoch": 3.83,
339
+ "learning_rate": 3.4252297410192146e-05,
340
+ "loss": 0.4018,
341
+ "step": 510
342
+ },
343
+ {
344
+ "epoch": 3.9,
345
+ "learning_rate": 3.3834586466165414e-05,
346
+ "loss": 0.4032,
347
+ "step": 520
348
+ },
349
+ {
350
+ "epoch": 3.98,
351
+ "learning_rate": 3.341687552213868e-05,
352
+ "loss": 0.3753,
353
+ "step": 530
354
+ },
355
+ {
356
+ "epoch": 3.99,
357
+ "eval_accuracy": 0.8787383177570094,
358
+ "eval_loss": 0.36458322405815125,
359
+ "eval_runtime": 86.4304,
360
+ "eval_samples_per_second": 49.52,
361
+ "eval_steps_per_second": 1.55,
362
+ "step": 532
363
+ },
364
+ {
365
+ "epoch": 4.06,
366
+ "learning_rate": 3.299916457811195e-05,
367
+ "loss": 0.3589,
368
+ "step": 540
369
+ },
370
+ {
371
+ "epoch": 4.13,
372
+ "learning_rate": 3.258145363408521e-05,
373
+ "loss": 0.344,
374
+ "step": 550
375
+ },
376
+ {
377
+ "epoch": 4.21,
378
+ "learning_rate": 3.216374269005848e-05,
379
+ "loss": 0.3571,
380
+ "step": 560
381
+ },
382
+ {
383
+ "epoch": 4.28,
384
+ "learning_rate": 3.1746031746031745e-05,
385
+ "loss": 0.3762,
386
+ "step": 570
387
+ },
388
+ {
389
+ "epoch": 4.36,
390
+ "learning_rate": 3.132832080200501e-05,
391
+ "loss": 0.3885,
392
+ "step": 580
393
+ },
394
+ {
395
+ "epoch": 4.43,
396
+ "learning_rate": 3.091060985797828e-05,
397
+ "loss": 0.3582,
398
+ "step": 590
399
+ },
400
+ {
401
+ "epoch": 4.51,
402
+ "learning_rate": 3.0492898913951544e-05,
403
+ "loss": 0.333,
404
+ "step": 600
405
+ },
406
+ {
407
+ "epoch": 4.58,
408
+ "learning_rate": 3.007518796992481e-05,
409
+ "loss": 0.3758,
410
+ "step": 610
411
+ },
412
+ {
413
+ "epoch": 4.66,
414
+ "learning_rate": 2.965747702589808e-05,
415
+ "loss": 0.3531,
416
+ "step": 620
417
+ },
418
+ {
419
+ "epoch": 4.73,
420
+ "learning_rate": 2.9239766081871346e-05,
421
+ "loss": 0.3627,
422
+ "step": 630
423
+ },
424
+ {
425
+ "epoch": 4.81,
426
+ "learning_rate": 2.882205513784461e-05,
427
+ "loss": 0.3575,
428
+ "step": 640
429
+ },
430
+ {
431
+ "epoch": 4.88,
432
+ "learning_rate": 2.8404344193817878e-05,
433
+ "loss": 0.3805,
434
+ "step": 650
435
+ },
436
+ {
437
+ "epoch": 4.96,
438
+ "learning_rate": 2.7986633249791145e-05,
439
+ "loss": 0.2997,
440
+ "step": 660
441
+ },
442
+ {
443
+ "epoch": 4.99,
444
+ "eval_accuracy": 0.8780373831775701,
445
+ "eval_loss": 0.36552494764328003,
446
+ "eval_runtime": 85.8197,
447
+ "eval_samples_per_second": 49.872,
448
+ "eval_steps_per_second": 1.561,
449
+ "step": 665
450
+ },
451
+ {
452
+ "epoch": 5.04,
453
+ "learning_rate": 2.756892230576441e-05,
454
+ "loss": 0.3479,
455
+ "step": 670
456
+ },
457
+ {
458
+ "epoch": 5.11,
459
+ "learning_rate": 2.7151211361737677e-05,
460
+ "loss": 0.3436,
461
+ "step": 680
462
+ },
463
+ {
464
+ "epoch": 5.19,
465
+ "learning_rate": 2.6733500417710944e-05,
466
+ "loss": 0.3158,
467
+ "step": 690
468
+ },
469
+ {
470
+ "epoch": 5.26,
471
+ "learning_rate": 2.6315789473684212e-05,
472
+ "loss": 0.312,
473
+ "step": 700
474
+ },
475
+ {
476
+ "epoch": 5.34,
477
+ "learning_rate": 2.5898078529657476e-05,
478
+ "loss": 0.3412,
479
+ "step": 710
480
+ },
481
+ {
482
+ "epoch": 5.41,
483
+ "learning_rate": 2.5480367585630744e-05,
484
+ "loss": 0.3333,
485
+ "step": 720
486
+ },
487
+ {
488
+ "epoch": 5.49,
489
+ "learning_rate": 2.506265664160401e-05,
490
+ "loss": 0.3272,
491
+ "step": 730
492
+ },
493
+ {
494
+ "epoch": 5.56,
495
+ "learning_rate": 2.4644945697577275e-05,
496
+ "loss": 0.3344,
497
+ "step": 740
498
+ },
499
+ {
500
+ "epoch": 5.64,
501
+ "learning_rate": 2.4227234753550543e-05,
502
+ "loss": 0.3864,
503
+ "step": 750
504
+ },
505
+ {
506
+ "epoch": 5.71,
507
+ "learning_rate": 2.380952380952381e-05,
508
+ "loss": 0.3322,
509
+ "step": 760
510
+ },
511
+ {
512
+ "epoch": 5.79,
513
+ "learning_rate": 2.3391812865497074e-05,
514
+ "loss": 0.3333,
515
+ "step": 770
516
+ },
517
+ {
518
+ "epoch": 5.86,
519
+ "learning_rate": 2.2974101921470342e-05,
520
+ "loss": 0.3277,
521
+ "step": 780
522
+ },
523
+ {
524
+ "epoch": 5.93,
525
+ "learning_rate": 2.255639097744361e-05,
526
+ "loss": 0.3176,
527
+ "step": 790
528
+ },
529
+ {
530
+ "epoch": 5.99,
531
+ "eval_accuracy": 0.8822429906542056,
532
+ "eval_loss": 0.35449549555778503,
533
+ "eval_runtime": 90.8772,
534
+ "eval_samples_per_second": 47.097,
535
+ "eval_steps_per_second": 1.475,
536
+ "step": 798
537
+ },
538
+ {
539
+ "epoch": 6.01,
540
+ "learning_rate": 2.2138680033416877e-05,
541
+ "loss": 0.3596,
542
+ "step": 800
543
+ },
544
+ {
545
+ "epoch": 6.09,
546
+ "learning_rate": 2.172096908939014e-05,
547
+ "loss": 0.2984,
548
+ "step": 810
549
+ },
550
+ {
551
+ "epoch": 6.16,
552
+ "learning_rate": 2.130325814536341e-05,
553
+ "loss": 0.3512,
554
+ "step": 820
555
+ },
556
+ {
557
+ "epoch": 6.24,
558
+ "learning_rate": 2.0885547201336676e-05,
559
+ "loss": 0.3356,
560
+ "step": 830
561
+ },
562
+ {
563
+ "epoch": 6.31,
564
+ "learning_rate": 2.046783625730994e-05,
565
+ "loss": 0.3269,
566
+ "step": 840
567
+ },
568
+ {
569
+ "epoch": 6.39,
570
+ "learning_rate": 2.0050125313283208e-05,
571
+ "loss": 0.2984,
572
+ "step": 850
573
+ },
574
+ {
575
+ "epoch": 6.46,
576
+ "learning_rate": 1.9632414369256475e-05,
577
+ "loss": 0.3279,
578
+ "step": 860
579
+ },
580
+ {
581
+ "epoch": 6.54,
582
+ "learning_rate": 1.9214703425229743e-05,
583
+ "loss": 0.2778,
584
+ "step": 870
585
+ },
586
+ {
587
+ "epoch": 6.61,
588
+ "learning_rate": 1.8796992481203007e-05,
589
+ "loss": 0.3247,
590
+ "step": 880
591
+ },
592
+ {
593
+ "epoch": 6.69,
594
+ "learning_rate": 1.8379281537176274e-05,
595
+ "loss": 0.3076,
596
+ "step": 890
597
+ },
598
+ {
599
+ "epoch": 6.76,
600
+ "learning_rate": 1.796157059314954e-05,
601
+ "loss": 0.3283,
602
+ "step": 900
603
+ },
604
+ {
605
+ "epoch": 6.84,
606
+ "learning_rate": 1.7543859649122806e-05,
607
+ "loss": 0.2951,
608
+ "step": 910
609
+ },
610
+ {
611
+ "epoch": 6.91,
612
+ "learning_rate": 1.7126148705096073e-05,
613
+ "loss": 0.3126,
614
+ "step": 920
615
+ },
616
+ {
617
+ "epoch": 6.99,
618
+ "learning_rate": 1.670843776106934e-05,
619
+ "loss": 0.2849,
620
+ "step": 930
621
+ },
622
+ {
623
+ "epoch": 6.99,
624
+ "eval_accuracy": 0.8850467289719626,
625
+ "eval_loss": 0.3441033363342285,
626
+ "eval_runtime": 85.0261,
627
+ "eval_samples_per_second": 50.337,
628
+ "eval_steps_per_second": 1.576,
629
+ "step": 931
630
+ },
631
+ {
632
+ "epoch": 7.07,
633
+ "learning_rate": 1.6290726817042605e-05,
634
+ "loss": 0.3229,
635
+ "step": 940
636
+ },
637
+ {
638
+ "epoch": 7.14,
639
+ "learning_rate": 1.5873015873015872e-05,
640
+ "loss": 0.2997,
641
+ "step": 950
642
+ },
643
+ {
644
+ "epoch": 7.22,
645
+ "learning_rate": 1.545530492898914e-05,
646
+ "loss": 0.2961,
647
+ "step": 960
648
+ },
649
+ {
650
+ "epoch": 7.29,
651
+ "learning_rate": 1.5037593984962406e-05,
652
+ "loss": 0.3217,
653
+ "step": 970
654
+ },
655
+ {
656
+ "epoch": 7.37,
657
+ "learning_rate": 1.4619883040935673e-05,
658
+ "loss": 0.282,
659
+ "step": 980
660
+ },
661
+ {
662
+ "epoch": 7.44,
663
+ "learning_rate": 1.4202172096908939e-05,
664
+ "loss": 0.2629,
665
+ "step": 990
666
+ },
667
+ {
668
+ "epoch": 7.52,
669
+ "learning_rate": 1.3784461152882205e-05,
670
+ "loss": 0.3394,
671
+ "step": 1000
672
+ },
673
+ {
674
+ "epoch": 7.59,
675
+ "learning_rate": 1.3366750208855472e-05,
676
+ "loss": 0.3002,
677
+ "step": 1010
678
+ },
679
+ {
680
+ "epoch": 7.67,
681
+ "learning_rate": 1.2949039264828738e-05,
682
+ "loss": 0.3042,
683
+ "step": 1020
684
+ },
685
+ {
686
+ "epoch": 7.74,
687
+ "learning_rate": 1.2531328320802006e-05,
688
+ "loss": 0.2942,
689
+ "step": 1030
690
+ },
691
+ {
692
+ "epoch": 7.81,
693
+ "learning_rate": 1.2113617376775271e-05,
694
+ "loss": 0.3077,
695
+ "step": 1040
696
+ },
697
+ {
698
+ "epoch": 7.89,
699
+ "learning_rate": 1.1695906432748537e-05,
700
+ "loss": 0.298,
701
+ "step": 1050
702
+ },
703
+ {
704
+ "epoch": 7.96,
705
+ "learning_rate": 1.1278195488721805e-05,
706
+ "loss": 0.2931,
707
+ "step": 1060
708
+ },
709
+ {
710
+ "epoch": 7.99,
711
+ "eval_accuracy": 0.8855140186915887,
712
+ "eval_loss": 0.3418597877025604,
713
+ "eval_runtime": 84.489,
714
+ "eval_samples_per_second": 50.657,
715
+ "eval_steps_per_second": 1.586,
716
+ "step": 1064
717
+ },
718
+ {
719
+ "epoch": 8.04,
720
+ "learning_rate": 1.086048454469507e-05,
721
+ "loss": 0.3204,
722
+ "step": 1070
723
+ },
724
+ {
725
+ "epoch": 8.12,
726
+ "learning_rate": 1.0442773600668338e-05,
727
+ "loss": 0.2924,
728
+ "step": 1080
729
+ },
730
+ {
731
+ "epoch": 8.19,
732
+ "learning_rate": 1.0025062656641604e-05,
733
+ "loss": 0.2871,
734
+ "step": 1090
735
+ },
736
+ {
737
+ "epoch": 8.27,
738
+ "learning_rate": 9.607351712614871e-06,
739
+ "loss": 0.2625,
740
+ "step": 1100
741
+ },
742
+ {
743
+ "epoch": 8.34,
744
+ "learning_rate": 9.189640768588137e-06,
745
+ "loss": 0.2914,
746
+ "step": 1110
747
+ },
748
+ {
749
+ "epoch": 8.42,
750
+ "learning_rate": 8.771929824561403e-06,
751
+ "loss": 0.2909,
752
+ "step": 1120
753
+ },
754
+ {
755
+ "epoch": 8.49,
756
+ "learning_rate": 8.35421888053467e-06,
757
+ "loss": 0.28,
758
+ "step": 1130
759
+ },
760
+ {
761
+ "epoch": 8.57,
762
+ "learning_rate": 7.936507936507936e-06,
763
+ "loss": 0.2703,
764
+ "step": 1140
765
+ },
766
+ {
767
+ "epoch": 8.64,
768
+ "learning_rate": 7.518796992481203e-06,
769
+ "loss": 0.3228,
770
+ "step": 1150
771
+ },
772
+ {
773
+ "epoch": 8.72,
774
+ "learning_rate": 7.1010860484544695e-06,
775
+ "loss": 0.3069,
776
+ "step": 1160
777
+ },
778
+ {
779
+ "epoch": 8.79,
780
+ "learning_rate": 6.683375104427736e-06,
781
+ "loss": 0.2699,
782
+ "step": 1170
783
+ },
784
+ {
785
+ "epoch": 8.87,
786
+ "learning_rate": 6.265664160401003e-06,
787
+ "loss": 0.281,
788
+ "step": 1180
789
+ },
790
+ {
791
+ "epoch": 8.94,
792
+ "learning_rate": 5.8479532163742686e-06,
793
+ "loss": 0.27,
794
+ "step": 1190
795
+ },
796
+ {
797
+ "epoch": 8.99,
798
+ "eval_accuracy": 0.8848130841121495,
799
+ "eval_loss": 0.341878741979599,
800
+ "eval_runtime": 84.7924,
801
+ "eval_samples_per_second": 50.476,
802
+ "eval_steps_per_second": 1.58,
803
+ "step": 1197
804
+ },
805
+ {
806
+ "epoch": 9.02,
807
+ "learning_rate": 5.430242272347535e-06,
808
+ "loss": 0.2934,
809
+ "step": 1200
810
+ },
811
+ {
812
+ "epoch": 9.1,
813
+ "learning_rate": 5.012531328320802e-06,
814
+ "loss": 0.2769,
815
+ "step": 1210
816
+ },
817
+ {
818
+ "epoch": 9.17,
819
+ "learning_rate": 4.5948203842940685e-06,
820
+ "loss": 0.2903,
821
+ "step": 1220
822
+ },
823
+ {
824
+ "epoch": 9.25,
825
+ "learning_rate": 4.177109440267335e-06,
826
+ "loss": 0.3092,
827
+ "step": 1230
828
+ },
829
+ {
830
+ "epoch": 9.32,
831
+ "learning_rate": 3.7593984962406014e-06,
832
+ "loss": 0.2552,
833
+ "step": 1240
834
+ },
835
+ {
836
+ "epoch": 9.4,
837
+ "learning_rate": 3.341687552213868e-06,
838
+ "loss": 0.3038,
839
+ "step": 1250
840
+ },
841
+ {
842
+ "epoch": 9.47,
843
+ "learning_rate": 2.9239766081871343e-06,
844
+ "loss": 0.2546,
845
+ "step": 1260
846
+ },
847
+ {
848
+ "epoch": 9.55,
849
+ "learning_rate": 2.506265664160401e-06,
850
+ "loss": 0.2846,
851
+ "step": 1270
852
+ },
853
+ {
854
+ "epoch": 9.62,
855
+ "learning_rate": 2.0885547201336676e-06,
856
+ "loss": 0.2794,
857
+ "step": 1280
858
+ },
859
+ {
860
+ "epoch": 9.7,
861
+ "learning_rate": 1.670843776106934e-06,
862
+ "loss": 0.2541,
863
+ "step": 1290
864
+ },
865
+ {
866
+ "epoch": 9.77,
867
+ "learning_rate": 1.2531328320802005e-06,
868
+ "loss": 0.2613,
869
+ "step": 1300
870
+ },
871
+ {
872
+ "epoch": 9.84,
873
+ "learning_rate": 8.35421888053467e-07,
874
+ "loss": 0.2633,
875
+ "step": 1310
876
+ },
877
+ {
878
+ "epoch": 9.92,
879
+ "learning_rate": 4.177109440267335e-07,
880
+ "loss": 0.2787,
881
+ "step": 1320
882
+ },
883
+ {
884
+ "epoch": 9.99,
885
+ "learning_rate": 0.0,
886
+ "loss": 0.2927,
887
+ "step": 1330
888
+ },
889
+ {
890
+ "epoch": 9.99,
891
+ "eval_accuracy": 0.8852803738317757,
892
+ "eval_loss": 0.34026381373405457,
893
+ "eval_runtime": 84.6819,
894
+ "eval_samples_per_second": 50.542,
895
+ "eval_steps_per_second": 1.582,
896
+ "step": 1330
897
+ },
898
+ {
899
+ "epoch": 9.99,
900
+ "step": 1330,
901
+ "total_flos": 1.3257453564799912e+19,
902
+ "train_loss": 0.4158173976984239,
903
+ "train_runtime": 6524.231,
904
+ "train_samples_per_second": 26.236,
905
+ "train_steps_per_second": 0.204
906
+ }
907
+ ],
908
+ "max_steps": 1330,
909
+ "num_train_epochs": 10,
910
+ "total_flos": 1.3257453564799912e+19,
911
+ "trial_name": null,
912
+ "trial_params": null
913
+ }