therealcyberlord commited on
Commit
acf13c9
1 Parent(s): e3e994f

Upload folder using huggingface_hub

Browse files
Files changed (5) hide show
  1. model.safetensors +1 -1
  2. optimizer.pt +1 -1
  3. scheduler.pt +1 -1
  4. trainer_state.json +373 -373
  5. training_args.bin +1 -1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8967e449012f7c944184f3ce6349fd31aa445ea5aa57e3ed9c3ebe1b72e9154e
3
  size 343263964
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c27310485a9b79a035ce96eefa30c3f05ca41b3dbc0a791f8358ec7a4cade6b
3
  size 343263964
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc1e1a549d597eaafe6ed33bbfaef5b45c6b665601d3306f11a2c94af7c827b5
3
  size 686648762
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcd11d58bd2ec412b20b27e86c8f5848d6add1200c349b431a5cd60b5225319b
3
  size 686648762
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:51e639b2ecb2aadfd272d6a9f35fa08de5eb4f8889c32310c3292be755c67c29
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5a7979006addc7da940500294cb37f12b9720d1946ac787792f8aae8227211
3
  size 1064
trainer_state.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "best_metric": 0.16356664896011353,
3
  "best_model_checkpoint": "./vit-indian-food/checkpoint-500",
4
  "epoch": 2.6178010471204187,
5
  "eval_steps": 50,
@@ -10,819 +10,819 @@
10
  "log_history": [
11
  {
12
  "epoch": 0.03,
13
- "grad_norm": 6.967112064361572,
14
- "learning_rate": 0.00019895287958115185,
15
- "loss": 2.3423,
16
  "step": 5
17
  },
18
  {
19
  "epoch": 0.05,
20
- "grad_norm": 6.768218517303467,
21
- "learning_rate": 0.00019790575916230367,
22
- "loss": 1.5082,
23
  "step": 10
24
  },
25
  {
26
  "epoch": 0.08,
27
- "grad_norm": 6.283707141876221,
28
- "learning_rate": 0.0001968586387434555,
29
- "loss": 1.1555,
30
  "step": 15
31
  },
32
  {
33
  "epoch": 0.1,
34
- "grad_norm": 4.167559623718262,
35
- "learning_rate": 0.00019581151832460733,
36
- "loss": 0.7045,
37
  "step": 20
38
  },
39
  {
40
  "epoch": 0.13,
41
- "grad_norm": 6.519095420837402,
42
- "learning_rate": 0.00019476439790575917,
43
- "loss": 0.5942,
44
  "step": 25
45
  },
46
  {
47
  "epoch": 0.16,
48
- "grad_norm": 4.0483479499816895,
49
- "learning_rate": 0.000193717277486911,
50
- "loss": 0.45,
51
  "step": 30
52
  },
53
  {
54
  "epoch": 0.18,
55
- "grad_norm": 4.359745025634766,
56
- "learning_rate": 0.00019267015706806283,
57
- "loss": 0.3833,
58
  "step": 35
59
  },
60
  {
61
  "epoch": 0.21,
62
- "grad_norm": 5.201784610748291,
63
- "learning_rate": 0.00019162303664921465,
64
- "loss": 0.3702,
65
  "step": 40
66
  },
67
  {
68
  "epoch": 0.24,
69
- "grad_norm": 4.259886741638184,
70
- "learning_rate": 0.0001905759162303665,
71
- "loss": 0.3227,
72
  "step": 45
73
  },
74
  {
75
  "epoch": 0.26,
76
- "grad_norm": 4.89500617980957,
77
- "learning_rate": 0.00018952879581151833,
78
- "loss": 0.2368,
79
  "step": 50
80
  },
81
  {
82
  "epoch": 0.26,
83
- "eval_accuracy": 0.9002624671916011,
84
- "eval_loss": 0.33144867420196533,
85
- "eval_precision": 0.9131344269428404,
86
- "eval_recall": 0.9002624671916011,
87
- "eval_runtime": 17.436,
88
- "eval_samples_per_second": 43.703,
89
- "eval_steps_per_second": 2.753,
90
  "step": 50
91
  },
92
  {
93
  "epoch": 0.29,
94
- "grad_norm": 4.650336265563965,
95
- "learning_rate": 0.00018848167539267018,
96
- "loss": 0.4139,
97
  "step": 55
98
  },
99
  {
100
  "epoch": 0.31,
101
- "grad_norm": 4.881960391998291,
102
- "learning_rate": 0.00018743455497382202,
103
- "loss": 0.2109,
104
  "step": 60
105
  },
106
  {
107
  "epoch": 0.34,
108
- "grad_norm": 5.3574066162109375,
109
- "learning_rate": 0.00018638743455497384,
110
- "loss": 0.3076,
111
  "step": 65
112
  },
113
  {
114
  "epoch": 0.37,
115
- "grad_norm": 5.008607864379883,
116
- "learning_rate": 0.00018534031413612568,
117
- "loss": 0.2632,
118
  "step": 70
119
  },
120
  {
121
  "epoch": 0.39,
122
- "grad_norm": 4.561398506164551,
123
- "learning_rate": 0.0001842931937172775,
124
- "loss": 0.2062,
125
  "step": 75
126
  },
127
  {
128
  "epoch": 0.42,
129
- "grad_norm": 5.034663677215576,
130
- "learning_rate": 0.00018324607329842934,
131
- "loss": 0.4752,
132
  "step": 80
133
  },
134
  {
135
  "epoch": 0.45,
136
- "grad_norm": 8.143318176269531,
137
- "learning_rate": 0.00018219895287958115,
138
- "loss": 0.3963,
139
  "step": 85
140
  },
141
  {
142
  "epoch": 0.47,
143
- "grad_norm": 6.556373119354248,
144
- "learning_rate": 0.000181151832460733,
145
- "loss": 0.5088,
146
  "step": 90
147
  },
148
  {
149
  "epoch": 0.5,
150
- "grad_norm": 4.414945125579834,
151
- "learning_rate": 0.0001801047120418848,
152
- "loss": 0.2903,
153
  "step": 95
154
  },
155
  {
156
  "epoch": 0.52,
157
- "grad_norm": 2.464193820953369,
158
- "learning_rate": 0.00017905759162303666,
159
- "loss": 0.2801,
160
  "step": 100
161
  },
162
  {
163
  "epoch": 0.52,
164
- "eval_accuracy": 0.9422572178477691,
165
- "eval_loss": 0.18048594892024994,
166
- "eval_precision": 0.9466471604125616,
167
- "eval_recall": 0.9422572178477691,
168
- "eval_runtime": 17.2463,
169
- "eval_samples_per_second": 44.183,
170
- "eval_steps_per_second": 2.783,
171
  "step": 100
172
  },
173
  {
174
  "epoch": 0.55,
175
- "grad_norm": 5.096113681793213,
176
- "learning_rate": 0.0001780104712041885,
177
- "loss": 0.213,
178
  "step": 105
179
  },
180
  {
181
  "epoch": 0.58,
182
- "grad_norm": 1.4251294136047363,
183
- "learning_rate": 0.00017696335078534032,
184
- "loss": 0.1831,
185
  "step": 110
186
  },
187
  {
188
  "epoch": 0.6,
189
- "grad_norm": 3.766965627670288,
190
- "learning_rate": 0.00017591623036649216,
191
- "loss": 0.1636,
192
  "step": 115
193
  },
194
  {
195
  "epoch": 0.63,
196
- "grad_norm": 3.66446852684021,
197
- "learning_rate": 0.00017486910994764398,
198
- "loss": 0.3075,
199
  "step": 120
200
  },
201
  {
202
  "epoch": 0.65,
203
- "grad_norm": 4.702181339263916,
204
- "learning_rate": 0.00017382198952879582,
205
- "loss": 0.2472,
206
  "step": 125
207
  },
208
  {
209
  "epoch": 0.68,
210
- "grad_norm": 8.463685035705566,
211
- "learning_rate": 0.00017277486910994763,
212
- "loss": 0.4341,
213
  "step": 130
214
  },
215
  {
216
  "epoch": 0.71,
217
- "grad_norm": 5.947971343994141,
218
- "learning_rate": 0.00017172774869109948,
219
- "loss": 0.2291,
220
  "step": 135
221
  },
222
  {
223
  "epoch": 0.73,
224
- "grad_norm": 5.682934761047363,
225
- "learning_rate": 0.00017068062827225132,
226
- "loss": 0.1632,
227
  "step": 140
228
  },
229
  {
230
  "epoch": 0.76,
231
- "grad_norm": 9.100614547729492,
232
- "learning_rate": 0.00016963350785340316,
233
- "loss": 0.307,
234
  "step": 145
235
  },
236
  {
237
  "epoch": 0.79,
238
- "grad_norm": 4.207605361938477,
239
- "learning_rate": 0.00016858638743455498,
240
- "loss": 0.2362,
241
  "step": 150
242
  },
243
  {
244
  "epoch": 0.79,
245
- "eval_accuracy": 0.9186351706036745,
246
- "eval_loss": 0.29145774245262146,
247
- "eval_precision": 0.9273784378254858,
248
- "eval_recall": 0.9186351706036745,
249
- "eval_runtime": 17.4226,
250
- "eval_samples_per_second": 43.736,
251
- "eval_steps_per_second": 2.755,
252
  "step": 150
253
  },
254
  {
255
  "epoch": 0.81,
256
- "grad_norm": 3.4303929805755615,
257
- "learning_rate": 0.00016753926701570682,
258
- "loss": 0.074,
259
  "step": 155
260
  },
261
  {
262
  "epoch": 0.84,
263
- "grad_norm": 4.852244853973389,
264
- "learning_rate": 0.00016649214659685867,
265
- "loss": 0.2731,
266
  "step": 160
267
  },
268
  {
269
  "epoch": 0.86,
270
- "grad_norm": 0.3142894506454468,
271
- "learning_rate": 0.00016544502617801048,
272
- "loss": 0.1295,
273
  "step": 165
274
  },
275
  {
276
  "epoch": 0.89,
277
- "grad_norm": 3.892892360687256,
278
- "learning_rate": 0.00016439790575916233,
279
- "loss": 0.225,
280
  "step": 170
281
  },
282
  {
283
  "epoch": 0.92,
284
- "grad_norm": 4.833043575286865,
285
- "learning_rate": 0.00016335078534031414,
286
- "loss": 0.2674,
287
  "step": 175
288
  },
289
  {
290
  "epoch": 0.94,
291
- "grad_norm": 5.150189399719238,
292
- "learning_rate": 0.00016230366492146599,
293
- "loss": 0.241,
294
  "step": 180
295
  },
296
  {
297
  "epoch": 0.97,
298
- "grad_norm": 5.426727771759033,
299
- "learning_rate": 0.0001612565445026178,
300
- "loss": 0.2049,
301
  "step": 185
302
  },
303
  {
304
  "epoch": 0.99,
305
- "grad_norm": 5.2064738273620605,
306
- "learning_rate": 0.00016020942408376964,
307
- "loss": 0.1756,
308
  "step": 190
309
  },
310
  {
311
  "epoch": 1.02,
312
- "grad_norm": 2.034318685531616,
313
- "learning_rate": 0.00015916230366492146,
314
- "loss": 0.0499,
315
  "step": 195
316
  },
317
  {
318
  "epoch": 1.05,
319
- "grad_norm": 4.4703145027160645,
320
- "learning_rate": 0.0001581151832460733,
321
- "loss": 0.042,
322
  "step": 200
323
  },
324
  {
325
  "epoch": 1.05,
326
- "eval_accuracy": 0.9461942257217848,
327
- "eval_loss": 0.16130226850509644,
328
- "eval_precision": 0.9511635531837818,
329
- "eval_recall": 0.9461942257217848,
330
- "eval_runtime": 17.3623,
331
- "eval_samples_per_second": 43.888,
332
- "eval_steps_per_second": 2.765,
333
  "step": 200
334
  },
335
  {
336
  "epoch": 1.07,
337
- "grad_norm": 0.06716866791248322,
338
- "learning_rate": 0.00015706806282722515,
339
- "loss": 0.0643,
340
  "step": 205
341
  },
342
  {
343
  "epoch": 1.1,
344
- "grad_norm": 7.002842903137207,
345
- "learning_rate": 0.00015602094240837696,
346
- "loss": 0.1028,
347
  "step": 210
348
  },
349
  {
350
  "epoch": 1.13,
351
- "grad_norm": 0.013943655416369438,
352
- "learning_rate": 0.0001549738219895288,
353
- "loss": 0.0941,
354
  "step": 215
355
  },
356
  {
357
  "epoch": 1.15,
358
- "grad_norm": 0.42214348912239075,
359
- "learning_rate": 0.00015392670157068062,
360
- "loss": 0.0273,
361
  "step": 220
362
  },
363
  {
364
  "epoch": 1.18,
365
- "grad_norm": 8.428805351257324,
366
- "learning_rate": 0.00015287958115183247,
367
- "loss": 0.0589,
368
  "step": 225
369
  },
370
  {
371
  "epoch": 1.2,
372
- "grad_norm": 2.468398094177246,
373
- "learning_rate": 0.00015183246073298428,
374
- "loss": 0.0321,
375
  "step": 230
376
  },
377
  {
378
  "epoch": 1.23,
379
- "grad_norm": 0.8661732077598572,
380
- "learning_rate": 0.00015078534031413612,
381
- "loss": 0.0856,
382
  "step": 235
383
  },
384
  {
385
  "epoch": 1.26,
386
- "grad_norm": 0.4541776478290558,
387
- "learning_rate": 0.00014973821989528797,
388
- "loss": 0.0136,
389
  "step": 240
390
  },
391
  {
392
  "epoch": 1.28,
393
- "grad_norm": 4.832011699676514,
394
- "learning_rate": 0.0001486910994764398,
395
- "loss": 0.1087,
396
  "step": 245
397
  },
398
  {
399
  "epoch": 1.31,
400
- "grad_norm": 4.744746685028076,
401
- "learning_rate": 0.00014764397905759163,
402
- "loss": 0.0477,
403
  "step": 250
404
  },
405
  {
406
  "epoch": 1.31,
407
- "eval_accuracy": 0.9514435695538058,
408
- "eval_loss": 0.14496225118637085,
409
- "eval_precision": 0.9534067355318784,
410
- "eval_recall": 0.9514435695538058,
411
- "eval_runtime": 17.4463,
412
- "eval_samples_per_second": 43.677,
413
- "eval_steps_per_second": 2.751,
414
  "step": 250
415
  },
416
  {
417
  "epoch": 1.34,
418
- "grad_norm": 0.053365129977464676,
419
- "learning_rate": 0.00014659685863874347,
420
- "loss": 0.0171,
421
  "step": 255
422
  },
423
  {
424
  "epoch": 1.36,
425
- "grad_norm": 0.41375842690467834,
426
- "learning_rate": 0.00014554973821989531,
427
- "loss": 0.1189,
428
  "step": 260
429
  },
430
  {
431
  "epoch": 1.39,
432
- "grad_norm": 0.05034720525145531,
433
- "learning_rate": 0.00014450261780104713,
434
- "loss": 0.0377,
435
  "step": 265
436
  },
437
  {
438
  "epoch": 1.41,
439
- "grad_norm": 2.498413324356079,
440
- "learning_rate": 0.00014345549738219897,
441
- "loss": 0.0334,
442
  "step": 270
443
  },
444
  {
445
  "epoch": 1.44,
446
- "grad_norm": 0.06190189719200134,
447
- "learning_rate": 0.0001424083769633508,
448
- "loss": 0.0158,
449
  "step": 275
450
  },
451
  {
452
  "epoch": 1.47,
453
- "grad_norm": 1.9862781763076782,
454
- "learning_rate": 0.00014136125654450263,
455
- "loss": 0.0186,
456
  "step": 280
457
  },
458
  {
459
  "epoch": 1.49,
460
- "grad_norm": 0.047805577516555786,
461
- "learning_rate": 0.00014031413612565445,
462
- "loss": 0.0705,
463
  "step": 285
464
  },
465
  {
466
  "epoch": 1.52,
467
- "grad_norm": 0.03765374794602394,
468
- "learning_rate": 0.0001392670157068063,
469
- "loss": 0.1822,
470
  "step": 290
471
  },
472
  {
473
  "epoch": 1.54,
474
- "grad_norm": 5.020593643188477,
475
- "learning_rate": 0.0001382198952879581,
476
- "loss": 0.0375,
477
  "step": 295
478
  },
479
  {
480
  "epoch": 1.57,
481
- "grad_norm": 1.0980048179626465,
482
- "learning_rate": 0.00013717277486910995,
483
- "loss": 0.1297,
484
  "step": 300
485
  },
486
  {
487
  "epoch": 1.57,
488
- "eval_accuracy": 0.937007874015748,
489
- "eval_loss": 0.24536970257759094,
490
- "eval_precision": 0.9442846541005356,
491
- "eval_recall": 0.937007874015748,
492
- "eval_runtime": 17.5121,
493
- "eval_samples_per_second": 43.513,
494
- "eval_steps_per_second": 2.741,
495
  "step": 300
496
  },
497
  {
498
  "epoch": 1.6,
499
- "grad_norm": 3.535846471786499,
500
- "learning_rate": 0.0001361256544502618,
501
- "loss": 0.1071,
502
  "step": 305
503
  },
504
  {
505
  "epoch": 1.62,
506
- "grad_norm": 4.182868957519531,
507
- "learning_rate": 0.0001350785340314136,
508
- "loss": 0.0249,
509
  "step": 310
510
  },
511
  {
512
  "epoch": 1.65,
513
- "grad_norm": 0.36884206533432007,
514
- "learning_rate": 0.00013403141361256545,
515
- "loss": 0.0762,
516
  "step": 315
517
  },
518
  {
519
  "epoch": 1.68,
520
- "grad_norm": 5.14253044128418,
521
- "learning_rate": 0.00013298429319371727,
522
- "loss": 0.0766,
523
  "step": 320
524
  },
525
  {
526
  "epoch": 1.7,
527
- "grad_norm": 1.2552727460861206,
528
- "learning_rate": 0.0001319371727748691,
529
- "loss": 0.0093,
530
  "step": 325
531
  },
532
  {
533
  "epoch": 1.73,
534
- "grad_norm": 3.767280101776123,
535
- "learning_rate": 0.00013089005235602096,
536
- "loss": 0.2232,
537
  "step": 330
538
  },
539
  {
540
  "epoch": 1.75,
541
- "grad_norm": 0.010386645793914795,
542
- "learning_rate": 0.00012984293193717277,
543
- "loss": 0.0026,
544
  "step": 335
545
  },
546
  {
547
  "epoch": 1.78,
548
- "grad_norm": 4.292264461517334,
549
- "learning_rate": 0.00012879581151832462,
550
- "loss": 0.2457,
551
  "step": 340
552
  },
553
  {
554
  "epoch": 1.81,
555
- "grad_norm": 0.024649567902088165,
556
- "learning_rate": 0.00012774869109947646,
557
- "loss": 0.0306,
558
  "step": 345
559
  },
560
  {
561
  "epoch": 1.83,
562
- "grad_norm": 5.098654270172119,
563
- "learning_rate": 0.00012670157068062827,
564
- "loss": 0.1156,
565
  "step": 350
566
  },
567
  {
568
  "epoch": 1.83,
569
- "eval_accuracy": 0.9501312335958005,
570
- "eval_loss": 0.17998863756656647,
571
- "eval_precision": 0.9531660720469141,
572
- "eval_recall": 0.9501312335958005,
573
- "eval_runtime": 17.4136,
574
- "eval_samples_per_second": 43.759,
575
- "eval_steps_per_second": 2.756,
576
  "step": 350
577
  },
578
  {
579
  "epoch": 1.86,
580
- "grad_norm": 0.9651967287063599,
581
- "learning_rate": 0.00012565445026178012,
582
- "loss": 0.0618,
583
  "step": 355
584
  },
585
  {
586
  "epoch": 1.88,
587
- "grad_norm": 2.727735996246338,
588
- "learning_rate": 0.00012460732984293196,
589
- "loss": 0.0705,
590
  "step": 360
591
  },
592
  {
593
  "epoch": 1.91,
594
- "grad_norm": 4.953824043273926,
595
- "learning_rate": 0.00012356020942408378,
596
- "loss": 0.2112,
597
  "step": 365
598
  },
599
  {
600
  "epoch": 1.94,
601
- "grad_norm": 0.08269181102514267,
602
- "learning_rate": 0.00012251308900523562,
603
- "loss": 0.071,
604
  "step": 370
605
  },
606
  {
607
  "epoch": 1.96,
608
- "grad_norm": 0.06396706402301788,
609
- "learning_rate": 0.00012146596858638744,
610
- "loss": 0.1069,
611
  "step": 375
612
  },
613
  {
614
  "epoch": 1.99,
615
- "grad_norm": 0.28295230865478516,
616
- "learning_rate": 0.00012041884816753928,
617
- "loss": 0.0759,
618
  "step": 380
619
  },
620
  {
621
  "epoch": 2.02,
622
- "grad_norm": 1.5805819034576416,
623
- "learning_rate": 0.0001193717277486911,
624
- "loss": 0.0622,
625
  "step": 385
626
  },
627
  {
628
  "epoch": 2.04,
629
- "grad_norm": 2.3917598724365234,
630
- "learning_rate": 0.00011832460732984294,
631
- "loss": 0.0512,
632
  "step": 390
633
  },
634
  {
635
  "epoch": 2.07,
636
- "grad_norm": 0.03971586003899574,
637
- "learning_rate": 0.00011727748691099475,
638
- "loss": 0.0163,
639
  "step": 395
640
  },
641
  {
642
  "epoch": 2.09,
643
- "grad_norm": 0.024471383541822433,
644
- "learning_rate": 0.0001162303664921466,
645
- "loss": 0.0188,
646
  "step": 400
647
  },
648
  {
649
  "epoch": 2.09,
650
- "eval_accuracy": 0.9593175853018373,
651
- "eval_loss": 0.14536795020103455,
652
- "eval_precision": 0.9601421810596481,
653
- "eval_recall": 0.9593175853018373,
654
- "eval_runtime": 17.4,
655
- "eval_samples_per_second": 43.793,
656
- "eval_steps_per_second": 2.759,
657
  "step": 400
658
  },
659
  {
660
  "epoch": 2.12,
661
- "grad_norm": 0.0705149918794632,
662
- "learning_rate": 0.00011518324607329844,
663
- "loss": 0.0062,
664
  "step": 405
665
  },
666
  {
667
  "epoch": 2.15,
668
- "grad_norm": 0.013097544200718403,
669
- "learning_rate": 0.00011413612565445027,
670
- "loss": 0.0191,
671
  "step": 410
672
  },
673
  {
674
  "epoch": 2.17,
675
- "grad_norm": 0.5236541628837585,
676
- "learning_rate": 0.00011308900523560211,
677
- "loss": 0.0403,
678
  "step": 415
679
  },
680
  {
681
  "epoch": 2.2,
682
- "grad_norm": 0.01967461034655571,
683
- "learning_rate": 0.00011204188481675393,
684
- "loss": 0.0032,
685
  "step": 420
686
  },
687
  {
688
  "epoch": 2.23,
689
- "grad_norm": 0.02198900654911995,
690
- "learning_rate": 0.00011099476439790577,
691
- "loss": 0.0019,
692
  "step": 425
693
  },
694
  {
695
  "epoch": 2.25,
696
- "grad_norm": 0.14719614386558533,
697
- "learning_rate": 0.00010994764397905759,
698
- "loss": 0.0052,
699
  "step": 430
700
  },
701
  {
702
  "epoch": 2.28,
703
- "grad_norm": 0.021793629974126816,
704
- "learning_rate": 0.00010890052356020943,
705
- "loss": 0.0105,
706
  "step": 435
707
  },
708
  {
709
  "epoch": 2.3,
710
- "grad_norm": 0.02368028089404106,
711
- "learning_rate": 0.00010785340314136125,
712
- "loss": 0.001,
713
  "step": 440
714
  },
715
  {
716
  "epoch": 2.33,
717
- "grad_norm": 0.010441082529723644,
718
- "learning_rate": 0.00010680628272251309,
719
- "loss": 0.0019,
720
  "step": 445
721
  },
722
  {
723
  "epoch": 2.36,
724
- "grad_norm": 0.015402965247631073,
725
- "learning_rate": 0.00010575916230366492,
726
- "loss": 0.0049,
727
  "step": 450
728
  },
729
  {
730
  "epoch": 2.36,
731
- "eval_accuracy": 0.9711286089238845,
732
- "eval_loss": 0.1082761362195015,
733
- "eval_precision": 0.9714506772867647,
734
- "eval_recall": 0.9711286089238845,
735
- "eval_runtime": 17.6808,
736
- "eval_samples_per_second": 43.098,
737
- "eval_steps_per_second": 2.715,
738
  "step": 450
739
  },
740
  {
741
  "epoch": 2.38,
742
- "grad_norm": 0.07379986345767975,
743
- "learning_rate": 0.00010471204188481676,
744
- "loss": 0.0047,
745
  "step": 455
746
  },
747
  {
748
  "epoch": 2.41,
749
- "grad_norm": 5.459177494049072,
750
- "learning_rate": 0.0001036649214659686,
751
- "loss": 0.0192,
752
  "step": 460
753
  },
754
  {
755
  "epoch": 2.43,
756
- "grad_norm": 0.004949438851326704,
757
- "learning_rate": 0.00010261780104712042,
758
- "loss": 0.0148,
759
  "step": 465
760
  },
761
  {
762
  "epoch": 2.46,
763
- "grad_norm": 0.17624743282794952,
764
- "learning_rate": 0.00010157068062827227,
765
- "loss": 0.0087,
766
  "step": 470
767
  },
768
  {
769
  "epoch": 2.49,
770
- "grad_norm": 0.010408706963062286,
771
- "learning_rate": 0.00010052356020942408,
772
- "loss": 0.0018,
773
  "step": 475
774
  },
775
  {
776
  "epoch": 2.51,
777
- "grad_norm": 0.007278731558471918,
778
- "learning_rate": 9.947643979057593e-05,
779
- "loss": 0.0029,
780
  "step": 480
781
  },
782
  {
783
  "epoch": 2.54,
784
- "grad_norm": 0.01558419968932867,
785
- "learning_rate": 9.842931937172776e-05,
786
- "loss": 0.0549,
787
  "step": 485
788
  },
789
  {
790
  "epoch": 2.57,
791
- "grad_norm": 0.005123753100633621,
792
- "learning_rate": 9.738219895287959e-05,
793
- "loss": 0.0006,
794
  "step": 490
795
  },
796
  {
797
  "epoch": 2.59,
798
- "grad_norm": 0.011126354336738586,
799
- "learning_rate": 9.633507853403142e-05,
800
- "loss": 0.003,
801
  "step": 495
802
  },
803
  {
804
  "epoch": 2.62,
805
- "grad_norm": 0.02012033388018608,
806
- "learning_rate": 9.528795811518324e-05,
807
- "loss": 0.0139,
808
  "step": 500
809
  },
810
  {
811
  "epoch": 2.62,
812
- "eval_accuracy": 0.963254593175853,
813
- "eval_loss": 0.16356664896011353,
814
- "eval_precision": 0.9644146563432678,
815
- "eval_recall": 0.963254593175853,
816
- "eval_runtime": 17.6855,
817
- "eval_samples_per_second": 43.086,
818
- "eval_steps_per_second": 2.714,
819
  "step": 500
820
  }
821
  ],
822
  "logging_steps": 5,
823
- "max_steps": 955,
824
  "num_input_tokens_seen": 0,
825
- "num_train_epochs": 5,
826
  "save_steps": 500,
827
  "total_flos": 6.186131303461724e+17,
828
  "train_batch_size": 16,
 
1
  {
2
+ "best_metric": 0.16896109282970428,
3
  "best_model_checkpoint": "./vit-indian-food/checkpoint-500",
4
  "epoch": 2.6178010471204187,
5
  "eval_steps": 50,
 
10
  "log_history": [
11
  {
12
  "epoch": 0.03,
13
+ "grad_norm": 7.558130264282227,
14
+ "learning_rate": 0.00019947643979057592,
15
+ "loss": 2.4369,
16
  "step": 5
17
  },
18
  {
19
  "epoch": 0.05,
20
+ "grad_norm": 6.5493950843811035,
21
+ "learning_rate": 0.00019895287958115185,
22
+ "loss": 1.6214,
23
  "step": 10
24
  },
25
  {
26
  "epoch": 0.08,
27
+ "grad_norm": 6.205695629119873,
28
+ "learning_rate": 0.00019842931937172776,
29
+ "loss": 1.197,
30
  "step": 15
31
  },
32
  {
33
  "epoch": 0.1,
34
+ "grad_norm": 3.8880319595336914,
35
+ "learning_rate": 0.00019790575916230367,
36
+ "loss": 0.7391,
37
  "step": 20
38
  },
39
  {
40
  "epoch": 0.13,
41
+ "grad_norm": 6.315922260284424,
42
+ "learning_rate": 0.0001973821989528796,
43
+ "loss": 0.6288,
44
  "step": 25
45
  },
46
  {
47
  "epoch": 0.16,
48
+ "grad_norm": 4.008142471313477,
49
+ "learning_rate": 0.0001968586387434555,
50
+ "loss": 0.3582,
51
  "step": 30
52
  },
53
  {
54
  "epoch": 0.18,
55
+ "grad_norm": 4.005300998687744,
56
+ "learning_rate": 0.00019633507853403142,
57
+ "loss": 0.3882,
58
  "step": 35
59
  },
60
  {
61
  "epoch": 0.21,
62
+ "grad_norm": 3.9506969451904297,
63
+ "learning_rate": 0.00019581151832460733,
64
+ "loss": 0.39,
65
  "step": 40
66
  },
67
  {
68
  "epoch": 0.24,
69
+ "grad_norm": 4.074863910675049,
70
+ "learning_rate": 0.00019528795811518326,
71
+ "loss": 0.3413,
72
  "step": 45
73
  },
74
  {
75
  "epoch": 0.26,
76
+ "grad_norm": 4.617203712463379,
77
+ "learning_rate": 0.00019476439790575917,
78
+ "loss": 0.358,
79
  "step": 50
80
  },
81
  {
82
  "epoch": 0.26,
83
+ "eval_accuracy": 0.9081364829396326,
84
+ "eval_loss": 0.32936015725135803,
85
+ "eval_precision": 0.9220515144455581,
86
+ "eval_recall": 0.9081364829396326,
87
+ "eval_runtime": 17.5763,
88
+ "eval_samples_per_second": 43.354,
89
+ "eval_steps_per_second": 2.731,
90
  "step": 50
91
  },
92
  {
93
  "epoch": 0.29,
94
+ "grad_norm": 4.685304641723633,
95
+ "learning_rate": 0.0001942408376963351,
96
+ "loss": 0.3669,
97
  "step": 55
98
  },
99
  {
100
  "epoch": 0.31,
101
+ "grad_norm": 3.119244337081909,
102
+ "learning_rate": 0.000193717277486911,
103
+ "loss": 0.1722,
104
  "step": 60
105
  },
106
  {
107
  "epoch": 0.34,
108
+ "grad_norm": 6.17639684677124,
109
+ "learning_rate": 0.00019319371727748692,
110
+ "loss": 0.405,
111
  "step": 65
112
  },
113
  {
114
  "epoch": 0.37,
115
+ "grad_norm": 4.263852119445801,
116
+ "learning_rate": 0.00019267015706806283,
117
+ "loss": 0.5013,
118
  "step": 70
119
  },
120
  {
121
  "epoch": 0.39,
122
+ "grad_norm": 2.03020977973938,
123
+ "learning_rate": 0.00019214659685863877,
124
+ "loss": 0.2251,
125
  "step": 75
126
  },
127
  {
128
  "epoch": 0.42,
129
+ "grad_norm": 5.0216193199157715,
130
+ "learning_rate": 0.00019162303664921465,
131
+ "loss": 0.3329,
132
  "step": 80
133
  },
134
  {
135
  "epoch": 0.45,
136
+ "grad_norm": 7.692380428314209,
137
+ "learning_rate": 0.00019109947643979058,
138
+ "loss": 0.2191,
139
  "step": 85
140
  },
141
  {
142
  "epoch": 0.47,
143
+ "grad_norm": 5.645913600921631,
144
+ "learning_rate": 0.0001905759162303665,
145
+ "loss": 0.3437,
146
  "step": 90
147
  },
148
  {
149
  "epoch": 0.5,
150
+ "grad_norm": 5.066718578338623,
151
+ "learning_rate": 0.00019005235602094243,
152
+ "loss": 0.4461,
153
  "step": 95
154
  },
155
  {
156
  "epoch": 0.52,
157
+ "grad_norm": 4.659124851226807,
158
+ "learning_rate": 0.00018952879581151833,
159
+ "loss": 0.5051,
160
  "step": 100
161
  },
162
  {
163
  "epoch": 0.52,
164
+ "eval_accuracy": 0.9291338582677166,
165
+ "eval_loss": 0.22470374405384064,
166
+ "eval_precision": 0.9372539303361345,
167
+ "eval_recall": 0.9291338582677166,
168
+ "eval_runtime": 17.4785,
169
+ "eval_samples_per_second": 43.597,
170
+ "eval_steps_per_second": 2.746,
171
  "step": 100
172
  },
173
  {
174
  "epoch": 0.55,
175
+ "grad_norm": 4.5200324058532715,
176
+ "learning_rate": 0.00018900523560209424,
177
+ "loss": 0.2615,
178
  "step": 105
179
  },
180
  {
181
  "epoch": 0.58,
182
+ "grad_norm": 1.2165738344192505,
183
+ "learning_rate": 0.00018848167539267018,
184
+ "loss": 0.2415,
185
  "step": 110
186
  },
187
  {
188
  "epoch": 0.6,
189
+ "grad_norm": 6.801146030426025,
190
+ "learning_rate": 0.00018795811518324608,
191
+ "loss": 0.3234,
192
  "step": 115
193
  },
194
  {
195
  "epoch": 0.63,
196
+ "grad_norm": 0.8039401173591614,
197
+ "learning_rate": 0.00018743455497382202,
198
+ "loss": 0.1962,
199
  "step": 120
200
  },
201
  {
202
  "epoch": 0.65,
203
+ "grad_norm": 4.639632701873779,
204
+ "learning_rate": 0.0001869109947643979,
205
+ "loss": 0.2755,
206
  "step": 125
207
  },
208
  {
209
  "epoch": 0.68,
210
+ "grad_norm": 3.4670636653900146,
211
+ "learning_rate": 0.00018638743455497384,
212
+ "loss": 0.3463,
213
  "step": 130
214
  },
215
  {
216
  "epoch": 0.71,
217
+ "grad_norm": 6.07156229019165,
218
+ "learning_rate": 0.00018586387434554974,
219
+ "loss": 0.3047,
220
  "step": 135
221
  },
222
  {
223
  "epoch": 0.73,
224
+ "grad_norm": 6.347087860107422,
225
+ "learning_rate": 0.00018534031413612568,
226
+ "loss": 0.2364,
227
  "step": 140
228
  },
229
  {
230
  "epoch": 0.76,
231
+ "grad_norm": 5.963706970214844,
232
+ "learning_rate": 0.00018481675392670156,
233
+ "loss": 0.4148,
234
  "step": 145
235
  },
236
  {
237
  "epoch": 0.79,
238
+ "grad_norm": 4.037764072418213,
239
+ "learning_rate": 0.0001842931937172775,
240
+ "loss": 0.1361,
241
  "step": 150
242
  },
243
  {
244
  "epoch": 0.79,
245
+ "eval_accuracy": 0.9409448818897638,
246
+ "eval_loss": 0.21655863523483276,
247
+ "eval_precision": 0.9454930944052194,
248
+ "eval_recall": 0.9409448818897638,
249
+ "eval_runtime": 17.4607,
250
+ "eval_samples_per_second": 43.641,
251
+ "eval_steps_per_second": 2.749,
252
  "step": 150
253
  },
254
  {
255
  "epoch": 0.81,
256
+ "grad_norm": 4.717739582061768,
257
+ "learning_rate": 0.0001837696335078534,
258
+ "loss": 0.1183,
259
  "step": 155
260
  },
261
  {
262
  "epoch": 0.84,
263
+ "grad_norm": 3.9427084922790527,
264
+ "learning_rate": 0.00018324607329842934,
265
+ "loss": 0.1138,
266
  "step": 160
267
  },
268
  {
269
  "epoch": 0.86,
270
+ "grad_norm": 4.915616989135742,
271
+ "learning_rate": 0.00018272251308900525,
272
+ "loss": 0.2988,
273
  "step": 165
274
  },
275
  {
276
  "epoch": 0.89,
277
+ "grad_norm": 6.3770432472229,
278
+ "learning_rate": 0.00018219895287958115,
279
+ "loss": 0.1744,
280
  "step": 170
281
  },
282
  {
283
  "epoch": 0.92,
284
+ "grad_norm": 0.7859281301498413,
285
+ "learning_rate": 0.00018167539267015706,
286
+ "loss": 0.2486,
287
  "step": 175
288
  },
289
  {
290
  "epoch": 0.94,
291
+ "grad_norm": 3.75736141204834,
292
+ "learning_rate": 0.000181151832460733,
293
+ "loss": 0.2637,
294
  "step": 180
295
  },
296
  {
297
  "epoch": 0.97,
298
+ "grad_norm": 6.771958351135254,
299
+ "learning_rate": 0.0001806282722513089,
300
+ "loss": 0.2305,
301
  "step": 185
302
  },
303
  {
304
  "epoch": 0.99,
305
+ "grad_norm": 4.951119899749756,
306
+ "learning_rate": 0.0001801047120418848,
307
+ "loss": 0.1108,
308
  "step": 190
309
  },
310
  {
311
  "epoch": 1.02,
312
+ "grad_norm": 3.362152576446533,
313
+ "learning_rate": 0.00017958115183246075,
314
+ "loss": 0.133,
315
  "step": 195
316
  },
317
  {
318
  "epoch": 1.05,
319
+ "grad_norm": 0.8346299529075623,
320
+ "learning_rate": 0.00017905759162303666,
321
+ "loss": 0.0611,
322
  "step": 200
323
  },
324
  {
325
  "epoch": 1.05,
326
+ "eval_accuracy": 0.9488188976377953,
327
+ "eval_loss": 0.18689152598381042,
328
+ "eval_precision": 0.9533647964651322,
329
+ "eval_recall": 0.9488188976377953,
330
+ "eval_runtime": 17.6111,
331
+ "eval_samples_per_second": 43.268,
332
+ "eval_steps_per_second": 2.726,
333
  "step": 200
334
  },
335
  {
336
  "epoch": 1.07,
337
+ "grad_norm": 0.5746695399284363,
338
+ "learning_rate": 0.0001785340314136126,
339
+ "loss": 0.0241,
340
  "step": 205
341
  },
342
  {
343
  "epoch": 1.1,
344
+ "grad_norm": 2.0743372440338135,
345
+ "learning_rate": 0.0001780104712041885,
346
+ "loss": 0.0662,
347
  "step": 210
348
  },
349
  {
350
  "epoch": 1.13,
351
+ "grad_norm": 1.6814557313919067,
352
+ "learning_rate": 0.0001774869109947644,
353
+ "loss": 0.0635,
354
  "step": 215
355
  },
356
  {
357
  "epoch": 1.15,
358
+ "grad_norm": 2.118739128112793,
359
+ "learning_rate": 0.00017696335078534032,
360
+ "loss": 0.1024,
361
  "step": 220
362
  },
363
  {
364
  "epoch": 1.18,
365
+ "grad_norm": 3.4306182861328125,
366
+ "learning_rate": 0.00017643979057591625,
367
+ "loss": 0.0201,
368
  "step": 225
369
  },
370
  {
371
  "epoch": 1.2,
372
+ "grad_norm": 0.28945377469062805,
373
+ "learning_rate": 0.00017591623036649216,
374
+ "loss": 0.1151,
375
  "step": 230
376
  },
377
  {
378
  "epoch": 1.23,
379
+ "grad_norm": 4.6208648681640625,
380
+ "learning_rate": 0.00017539267015706807,
381
+ "loss": 0.2324,
382
  "step": 235
383
  },
384
  {
385
  "epoch": 1.26,
386
+ "grad_norm": 1.6085395812988281,
387
+ "learning_rate": 0.00017486910994764398,
388
+ "loss": 0.0429,
389
  "step": 240
390
  },
391
  {
392
  "epoch": 1.28,
393
+ "grad_norm": 5.860565185546875,
394
+ "learning_rate": 0.0001743455497382199,
395
+ "loss": 0.0567,
396
  "step": 245
397
  },
398
  {
399
  "epoch": 1.31,
400
+ "grad_norm": 5.421276092529297,
401
+ "learning_rate": 0.00017382198952879582,
402
+ "loss": 0.1037,
403
  "step": 250
404
  },
405
  {
406
  "epoch": 1.31,
407
+ "eval_accuracy": 0.968503937007874,
408
+ "eval_loss": 0.11793948709964752,
409
+ "eval_precision": 0.9693355319509627,
410
+ "eval_recall": 0.968503937007874,
411
+ "eval_runtime": 17.5701,
412
+ "eval_samples_per_second": 43.369,
413
+ "eval_steps_per_second": 2.732,
414
  "step": 250
415
  },
416
  {
417
  "epoch": 1.34,
418
+ "grad_norm": 3.9477956295013428,
419
+ "learning_rate": 0.00017329842931937175,
420
+ "loss": 0.0285,
421
  "step": 255
422
  },
423
  {
424
  "epoch": 1.36,
425
+ "grad_norm": 7.515362739562988,
426
+ "learning_rate": 0.00017277486910994763,
427
+ "loss": 0.2195,
428
  "step": 260
429
  },
430
  {
431
  "epoch": 1.39,
432
+ "grad_norm": 0.1676739752292633,
433
+ "learning_rate": 0.00017225130890052357,
434
+ "loss": 0.0172,
435
  "step": 265
436
  },
437
  {
438
  "epoch": 1.41,
439
+ "grad_norm": 8.235359191894531,
440
+ "learning_rate": 0.00017172774869109948,
441
+ "loss": 0.1756,
442
  "step": 270
443
  },
444
  {
445
  "epoch": 1.44,
446
+ "grad_norm": 7.9198832511901855,
447
+ "learning_rate": 0.0001712041884816754,
448
+ "loss": 0.0663,
449
  "step": 275
450
  },
451
  {
452
  "epoch": 1.47,
453
+ "grad_norm": 6.388627529144287,
454
+ "learning_rate": 0.00017068062827225132,
455
+ "loss": 0.1467,
456
  "step": 280
457
  },
458
  {
459
  "epoch": 1.49,
460
+ "grad_norm": 0.09213338047266006,
461
+ "learning_rate": 0.00017015706806282723,
462
+ "loss": 0.0684,
463
  "step": 285
464
  },
465
  {
466
  "epoch": 1.52,
467
+ "grad_norm": 1.0691012144088745,
468
+ "learning_rate": 0.00016963350785340316,
469
+ "loss": 0.0954,
470
  "step": 290
471
  },
472
  {
473
  "epoch": 1.54,
474
+ "grad_norm": 6.120029449462891,
475
+ "learning_rate": 0.00016910994764397907,
476
+ "loss": 0.0587,
477
  "step": 295
478
  },
479
  {
480
  "epoch": 1.57,
481
+ "grad_norm": 0.0928138718008995,
482
+ "learning_rate": 0.00016858638743455498,
483
+ "loss": 0.0294,
484
  "step": 300
485
  },
486
  {
487
  "epoch": 1.57,
488
+ "eval_accuracy": 0.963254593175853,
489
+ "eval_loss": 0.11820019036531448,
490
+ "eval_precision": 0.9645155573207865,
491
+ "eval_recall": 0.963254593175853,
492
+ "eval_runtime": 17.5014,
493
+ "eval_samples_per_second": 43.539,
494
+ "eval_steps_per_second": 2.743,
495
  "step": 300
496
  },
497
  {
498
  "epoch": 1.6,
499
+ "grad_norm": 0.3131069242954254,
500
+ "learning_rate": 0.0001680628272251309,
501
+ "loss": 0.1184,
502
  "step": 305
503
  },
504
  {
505
  "epoch": 1.62,
506
+ "grad_norm": 0.417850524187088,
507
+ "learning_rate": 0.00016753926701570682,
508
+ "loss": 0.0053,
509
  "step": 310
510
  },
511
  {
512
  "epoch": 1.65,
513
+ "grad_norm": 5.371811389923096,
514
+ "learning_rate": 0.00016701570680628273,
515
+ "loss": 0.1099,
516
  "step": 315
517
  },
518
  {
519
  "epoch": 1.68,
520
+ "grad_norm": 4.09129524230957,
521
+ "learning_rate": 0.00016649214659685867,
522
+ "loss": 0.2501,
523
  "step": 320
524
  },
525
  {
526
  "epoch": 1.7,
527
+ "grad_norm": 0.6515682935714722,
528
+ "learning_rate": 0.00016596858638743455,
529
+ "loss": 0.0528,
530
  "step": 325
531
  },
532
  {
533
  "epoch": 1.73,
534
+ "grad_norm": 0.5656059384346008,
535
+ "learning_rate": 0.00016544502617801048,
536
+ "loss": 0.1378,
537
  "step": 330
538
  },
539
  {
540
  "epoch": 1.75,
541
+ "grad_norm": 0.10378353297710419,
542
+ "learning_rate": 0.0001649214659685864,
543
+ "loss": 0.0329,
544
  "step": 335
545
  },
546
  {
547
  "epoch": 1.78,
548
+ "grad_norm": 2.387305736541748,
549
+ "learning_rate": 0.00016439790575916233,
550
+ "loss": 0.0353,
551
  "step": 340
552
  },
553
  {
554
  "epoch": 1.81,
555
+ "grad_norm": 0.05628238245844841,
556
+ "learning_rate": 0.0001638743455497382,
557
+ "loss": 0.0138,
558
  "step": 345
559
  },
560
  {
561
  "epoch": 1.83,
562
+ "grad_norm": 0.04053734615445137,
563
+ "learning_rate": 0.00016335078534031414,
564
+ "loss": 0.0082,
565
  "step": 350
566
  },
567
  {
568
  "epoch": 1.83,
569
+ "eval_accuracy": 0.963254593175853,
570
+ "eval_loss": 0.11838709563016891,
571
+ "eval_precision": 0.9654537871036863,
572
+ "eval_recall": 0.963254593175853,
573
+ "eval_runtime": 17.4034,
574
+ "eval_samples_per_second": 43.785,
575
+ "eval_steps_per_second": 2.758,
576
  "step": 350
577
  },
578
  {
579
  "epoch": 1.86,
580
+ "grad_norm": 2.287546396255493,
581
+ "learning_rate": 0.00016282722513089005,
582
+ "loss": 0.02,
583
  "step": 355
584
  },
585
  {
586
  "epoch": 1.88,
587
+ "grad_norm": 4.561453342437744,
588
+ "learning_rate": 0.00016230366492146599,
589
+ "loss": 0.203,
590
  "step": 360
591
  },
592
  {
593
  "epoch": 1.91,
594
+ "grad_norm": 4.0477294921875,
595
+ "learning_rate": 0.0001617801047120419,
596
+ "loss": 0.0646,
597
  "step": 365
598
  },
599
  {
600
  "epoch": 1.94,
601
+ "grad_norm": 1.0380969047546387,
602
+ "learning_rate": 0.0001612565445026178,
603
+ "loss": 0.0848,
604
  "step": 370
605
  },
606
  {
607
  "epoch": 1.96,
608
+ "grad_norm": 4.385131359100342,
609
+ "learning_rate": 0.0001607329842931937,
610
+ "loss": 0.166,
611
  "step": 375
612
  },
613
  {
614
  "epoch": 1.99,
615
+ "grad_norm": 0.036621347069740295,
616
+ "learning_rate": 0.00016020942408376964,
617
+ "loss": 0.0871,
618
  "step": 380
619
  },
620
  {
621
  "epoch": 2.02,
622
+ "grad_norm": 0.5929269194602966,
623
+ "learning_rate": 0.00015968586387434555,
624
+ "loss": 0.1271,
625
  "step": 385
626
  },
627
  {
628
  "epoch": 2.04,
629
+ "grad_norm": 0.26594266295433044,
630
+ "learning_rate": 0.00015916230366492146,
631
+ "loss": 0.0114,
632
  "step": 390
633
  },
634
  {
635
  "epoch": 2.07,
636
+ "grad_norm": 0.020196596160531044,
637
+ "learning_rate": 0.0001586387434554974,
638
+ "loss": 0.0081,
639
  "step": 395
640
  },
641
  {
642
  "epoch": 2.09,
643
+ "grad_norm": 0.6509531140327454,
644
+ "learning_rate": 0.0001581151832460733,
645
+ "loss": 0.0206,
646
  "step": 400
647
  },
648
  {
649
  "epoch": 2.09,
650
+ "eval_accuracy": 0.963254593175853,
651
+ "eval_loss": 0.130891352891922,
652
+ "eval_precision": 0.964741369994443,
653
+ "eval_recall": 0.963254593175853,
654
+ "eval_runtime": 17.4584,
655
+ "eval_samples_per_second": 43.647,
656
+ "eval_steps_per_second": 2.749,
657
  "step": 400
658
  },
659
  {
660
  "epoch": 2.12,
661
+ "grad_norm": 0.04883525148034096,
662
+ "learning_rate": 0.00015759162303664924,
663
+ "loss": 0.047,
664
  "step": 405
665
  },
666
  {
667
  "epoch": 2.15,
668
+ "grad_norm": 1.4584039449691772,
669
+ "learning_rate": 0.00015706806282722515,
670
+ "loss": 0.0393,
671
  "step": 410
672
  },
673
  {
674
  "epoch": 2.17,
675
+ "grad_norm": 0.3986053466796875,
676
+ "learning_rate": 0.00015654450261780105,
677
+ "loss": 0.0606,
678
  "step": 415
679
  },
680
  {
681
  "epoch": 2.2,
682
+ "grad_norm": 0.11985136568546295,
683
+ "learning_rate": 0.00015602094240837696,
684
+ "loss": 0.0379,
685
  "step": 420
686
  },
687
  {
688
  "epoch": 2.23,
689
+ "grad_norm": 0.07443628460168839,
690
+ "learning_rate": 0.0001554973821989529,
691
+ "loss": 0.0157,
692
  "step": 425
693
  },
694
  {
695
  "epoch": 2.25,
696
+ "grad_norm": 0.03219376876950264,
697
+ "learning_rate": 0.0001549738219895288,
698
+ "loss": 0.01,
699
  "step": 430
700
  },
701
  {
702
  "epoch": 2.28,
703
+ "grad_norm": 0.17020490765571594,
704
+ "learning_rate": 0.00015445026178010471,
705
+ "loss": 0.0093,
706
  "step": 435
707
  },
708
  {
709
  "epoch": 2.3,
710
+ "grad_norm": 2.6877171993255615,
711
+ "learning_rate": 0.00015392670157068062,
712
+ "loss": 0.021,
713
  "step": 440
714
  },
715
  {
716
  "epoch": 2.33,
717
+ "grad_norm": 0.010094034485518932,
718
+ "learning_rate": 0.00015340314136125656,
719
+ "loss": 0.0083,
720
  "step": 445
721
  },
722
  {
723
  "epoch": 2.36,
724
+ "grad_norm": 0.0077184755355119705,
725
+ "learning_rate": 0.00015287958115183247,
726
+ "loss": 0.0246,
727
  "step": 450
728
  },
729
  {
730
  "epoch": 2.36,
731
+ "eval_accuracy": 0.9553805774278216,
732
+ "eval_loss": 0.17534801363945007,
733
+ "eval_precision": 0.9586171915743325,
734
+ "eval_recall": 0.9553805774278216,
735
+ "eval_runtime": 17.505,
736
+ "eval_samples_per_second": 43.53,
737
+ "eval_steps_per_second": 2.742,
738
  "step": 450
739
  },
740
  {
741
  "epoch": 2.38,
742
+ "grad_norm": 5.202419757843018,
743
+ "learning_rate": 0.0001523560209424084,
744
+ "loss": 0.032,
745
  "step": 455
746
  },
747
  {
748
  "epoch": 2.41,
749
+ "grad_norm": 0.018663793802261353,
750
+ "learning_rate": 0.00015183246073298428,
751
+ "loss": 0.0221,
752
  "step": 460
753
  },
754
  {
755
  "epoch": 2.43,
756
+ "grad_norm": 1.0761280059814453,
757
+ "learning_rate": 0.00015130890052356022,
758
+ "loss": 0.0055,
759
  "step": 465
760
  },
761
  {
762
  "epoch": 2.46,
763
+ "grad_norm": 0.8688761591911316,
764
+ "learning_rate": 0.00015078534031413612,
765
+ "loss": 0.0852,
766
  "step": 470
767
  },
768
  {
769
  "epoch": 2.49,
770
+ "grad_norm": 0.06378360092639923,
771
+ "learning_rate": 0.00015026178010471206,
772
+ "loss": 0.0031,
773
  "step": 475
774
  },
775
  {
776
  "epoch": 2.51,
777
+ "grad_norm": 7.236721992492676,
778
+ "learning_rate": 0.00014973821989528797,
779
+ "loss": 0.0106,
780
  "step": 480
781
  },
782
  {
783
  "epoch": 2.54,
784
+ "grad_norm": 0.013875219039618969,
785
+ "learning_rate": 0.00014921465968586388,
786
+ "loss": 0.0032,
787
  "step": 485
788
  },
789
  {
790
  "epoch": 2.57,
791
+ "grad_norm": 0.031233886256814003,
792
+ "learning_rate": 0.0001486910994764398,
793
+ "loss": 0.0047,
794
  "step": 490
795
  },
796
  {
797
  "epoch": 2.59,
798
+ "grad_norm": 0.2528940439224243,
799
+ "learning_rate": 0.00014816753926701572,
800
+ "loss": 0.0081,
801
  "step": 495
802
  },
803
  {
804
  "epoch": 2.62,
805
+ "grad_norm": 0.02138712629675865,
806
+ "learning_rate": 0.00014764397905759163,
807
+ "loss": 0.0161,
808
  "step": 500
809
  },
810
  {
811
  "epoch": 2.62,
812
+ "eval_accuracy": 0.9514435695538058,
813
+ "eval_loss": 0.16896109282970428,
814
+ "eval_precision": 0.9537490891546931,
815
+ "eval_recall": 0.9514435695538058,
816
+ "eval_runtime": 17.5851,
817
+ "eval_samples_per_second": 43.332,
818
+ "eval_steps_per_second": 2.73,
819
  "step": 500
820
  }
821
  ],
822
  "logging_steps": 5,
823
+ "max_steps": 1910,
824
  "num_input_tokens_seen": 0,
825
+ "num_train_epochs": 10,
826
  "save_steps": 500,
827
  "total_flos": 6.186131303461724e+17,
828
  "train_batch_size": 16,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9351bf119f7d6bd8f47397285059fe8f281c9320b7267ccda968417cdf32d2fa
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20978be5af09aa5e062dc23ab1ae5cd2911d62519dbe8c26b70e5d2de0491f9b
3
  size 4856