Augusto777 commited on
Commit
4334b49
1 Parent(s): 33be873

End of training

Browse files
README.md CHANGED
@@ -21,7 +21,7 @@ model-index:
21
  metrics:
22
  - name: Accuracy
23
  type: accuracy
24
- value: 0.6956521739130435
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on the imagefolder dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 1.0596
35
- - Accuracy: 0.6957
36
 
37
  ## Model description
38
 
 
21
  metrics:
22
  - name: Accuracy
23
  type: accuracy
24
+ value: 0.8478260869565217
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on the imagefolder dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 0.6640
35
+ - Accuracy: 0.8478
36
 
37
  ## Model description
38
 
all_results.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
- "epoch": 73.85,
3
- "eval_accuracy": 0.8695652173913043,
4
- "eval_loss": 0.6163212060928345,
5
- "eval_runtime": 0.6921,
6
- "eval_samples_per_second": 66.464,
7
- "eval_steps_per_second": 2.89,
8
- "train_loss": 0.38355133185784024,
9
- "train_runtime": 282.7663,
10
- "train_samples_per_second": 231.711,
11
- "train_steps_per_second": 1.698
12
  }
 
1
  {
2
+ "epoch": 92.31,
3
+ "eval_accuracy": 0.8478260869565217,
4
+ "eval_loss": 0.6640377044677734,
5
+ "eval_runtime": 0.7832,
6
+ "eval_samples_per_second": 58.731,
7
+ "eval_steps_per_second": 2.554,
8
+ "train_loss": 0.2747100231051445,
9
+ "train_runtime": 347.4081,
10
+ "train_samples_per_second": 235.746,
11
+ "train_steps_per_second": 1.727
12
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 73.85,
3
- "eval_accuracy": 0.8695652173913043,
4
- "eval_loss": 0.6163212060928345,
5
- "eval_runtime": 0.6921,
6
- "eval_samples_per_second": 66.464,
7
- "eval_steps_per_second": 2.89
8
  }
 
1
  {
2
+ "epoch": 92.31,
3
+ "eval_accuracy": 0.8478260869565217,
4
+ "eval_loss": 0.6640377044677734,
5
+ "eval_runtime": 0.7832,
6
+ "eval_samples_per_second": 58.731,
7
+ "eval_steps_per_second": 2.554
8
  }
runs/Jun12_16-35-19_DESKTOP-SKBE9FB/events.out.tfevents.1718232073.DESKTOP-SKBE9FB.14408.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:342a69a88cc986938ee2330d8b69ca17bfb23627bf5f34b2fda1ead76c8f35b6
3
+ size 411
train_results.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "epoch": 73.85,
3
- "train_loss": 0.38355133185784024,
4
- "train_runtime": 282.7663,
5
- "train_samples_per_second": 231.711,
6
- "train_steps_per_second": 1.698
7
  }
 
1
  {
2
+ "epoch": 92.31,
3
+ "train_loss": 0.2747100231051445,
4
+ "train_runtime": 347.4081,
5
+ "train_samples_per_second": 235.746,
6
+ "train_steps_per_second": 1.727
7
  }
trainer_state.json CHANGED
@@ -1,983 +1,1226 @@
1
  {
2
- "best_metric": 0.8695652173913043,
3
  "best_model_checkpoint": "swiftformer-xs-ve-U13-b-80\\checkpoint-175",
4
- "epoch": 73.84615384615384,
5
  "eval_steps": 500,
6
- "global_step": 480,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.92,
13
- "eval_accuracy": 0.10869565217391304,
14
- "eval_loss": 1.4460796117782593,
15
- "eval_runtime": 0.7125,
16
- "eval_samples_per_second": 64.563,
17
- "eval_steps_per_second": 2.807,
18
  "step": 6
19
  },
20
  {
21
  "epoch": 1.54,
22
- "learning_rate": 4.166666666666667e-05,
23
- "loss": 1.3993,
24
  "step": 10
25
  },
26
  {
27
  "epoch": 2.0,
28
- "eval_accuracy": 0.10869565217391304,
29
- "eval_loss": 1.4435073137283325,
30
- "eval_runtime": 0.6148,
31
- "eval_samples_per_second": 74.823,
32
- "eval_steps_per_second": 3.253,
33
  "step": 13
34
  },
35
  {
36
  "epoch": 2.92,
37
- "eval_accuracy": 0.10869565217391304,
38
- "eval_loss": 1.4388779401779175,
39
- "eval_runtime": 0.6122,
40
- "eval_samples_per_second": 75.137,
41
- "eval_steps_per_second": 3.267,
42
  "step": 19
43
  },
44
  {
45
  "epoch": 3.08,
46
- "learning_rate": 8.333333333333334e-05,
47
- "loss": 1.3849,
48
  "step": 20
49
  },
50
  {
51
  "epoch": 4.0,
52
- "eval_accuracy": 0.10869565217391304,
53
- "eval_loss": 1.4283998012542725,
54
- "eval_runtime": 0.6101,
55
- "eval_samples_per_second": 75.4,
56
- "eval_steps_per_second": 3.278,
57
  "step": 26
58
  },
59
  {
60
  "epoch": 4.62,
61
- "learning_rate": 0.000125,
62
- "loss": 1.3287,
63
  "step": 30
64
  },
65
  {
66
  "epoch": 4.92,
67
- "eval_accuracy": 0.13043478260869565,
68
- "eval_loss": 1.4222790002822876,
69
- "eval_runtime": 0.6169,
70
- "eval_samples_per_second": 74.563,
71
- "eval_steps_per_second": 3.242,
72
  "step": 32
73
  },
74
  {
75
  "epoch": 6.0,
76
- "eval_accuracy": 0.10869565217391304,
77
- "eval_loss": 1.4646577835083008,
78
- "eval_runtime": 0.6082,
79
- "eval_samples_per_second": 75.638,
80
- "eval_steps_per_second": 3.289,
81
  "step": 39
82
  },
83
  {
84
  "epoch": 6.15,
85
- "learning_rate": 0.0001666666666666667,
86
- "loss": 1.2128,
87
  "step": 40
88
  },
89
  {
90
  "epoch": 6.92,
91
- "eval_accuracy": 0.17391304347826086,
92
- "eval_loss": 1.4183566570281982,
93
- "eval_runtime": 0.64,
94
- "eval_samples_per_second": 71.878,
95
- "eval_steps_per_second": 3.125,
96
  "step": 45
97
  },
98
  {
99
  "epoch": 7.69,
100
- "learning_rate": 0.0001990740740740741,
101
- "loss": 1.122,
102
  "step": 50
103
  },
104
  {
105
  "epoch": 8.0,
106
- "eval_accuracy": 0.1956521739130435,
107
- "eval_loss": 1.3262405395507812,
108
- "eval_runtime": 0.6384,
109
- "eval_samples_per_second": 72.055,
110
- "eval_steps_per_second": 3.133,
111
  "step": 52
112
  },
113
  {
114
  "epoch": 8.92,
115
- "eval_accuracy": 0.1956521739130435,
116
- "eval_loss": 1.3297909498214722,
117
- "eval_runtime": 0.5935,
118
- "eval_samples_per_second": 77.507,
119
- "eval_steps_per_second": 3.37,
120
  "step": 58
121
  },
122
  {
123
  "epoch": 9.23,
124
- "learning_rate": 0.00019444444444444446,
125
- "loss": 1.0062,
126
  "step": 60
127
  },
128
  {
129
  "epoch": 10.0,
130
- "eval_accuracy": 0.3695652173913043,
131
- "eval_loss": 1.2035472393035889,
132
- "eval_runtime": 0.5924,
133
- "eval_samples_per_second": 77.654,
134
- "eval_steps_per_second": 3.376,
135
  "step": 65
136
  },
137
  {
138
  "epoch": 10.77,
139
- "learning_rate": 0.00018981481481481483,
140
- "loss": 0.872,
141
  "step": 70
142
  },
143
  {
144
  "epoch": 10.92,
145
- "eval_accuracy": 0.32608695652173914,
146
- "eval_loss": 1.3667021989822388,
147
- "eval_runtime": 0.6186,
148
- "eval_samples_per_second": 74.367,
149
- "eval_steps_per_second": 3.233,
150
  "step": 71
151
  },
152
  {
153
  "epoch": 12.0,
154
- "eval_accuracy": 0.43478260869565216,
155
- "eval_loss": 1.201258897781372,
156
- "eval_runtime": 0.6083,
157
- "eval_samples_per_second": 75.623,
158
- "eval_steps_per_second": 3.288,
159
  "step": 78
160
  },
161
  {
162
  "epoch": 12.31,
163
- "learning_rate": 0.0001851851851851852,
164
- "loss": 0.814,
165
  "step": 80
166
  },
167
  {
168
  "epoch": 12.92,
169
- "eval_accuracy": 0.6956521739130435,
170
- "eval_loss": 0.9996185898780823,
171
- "eval_runtime": 0.6063,
172
- "eval_samples_per_second": 75.874,
173
- "eval_steps_per_second": 3.299,
174
  "step": 84
175
  },
176
  {
177
  "epoch": 13.85,
178
- "learning_rate": 0.00018055555555555557,
179
- "loss": 0.7228,
180
  "step": 90
181
  },
182
  {
183
  "epoch": 14.0,
184
- "eval_accuracy": 0.7608695652173914,
185
- "eval_loss": 0.9706003069877625,
186
- "eval_runtime": 0.6069,
187
- "eval_samples_per_second": 75.799,
188
- "eval_steps_per_second": 3.296,
189
  "step": 91
190
  },
191
  {
192
  "epoch": 14.92,
193
- "eval_accuracy": 0.7391304347826086,
194
- "eval_loss": 0.9295421242713928,
195
- "eval_runtime": 0.6331,
196
- "eval_samples_per_second": 72.657,
197
- "eval_steps_per_second": 3.159,
198
  "step": 97
199
  },
200
  {
201
  "epoch": 15.38,
202
- "learning_rate": 0.00017592592592592595,
203
- "loss": 0.6473,
204
  "step": 100
205
  },
206
  {
207
  "epoch": 16.0,
208
- "eval_accuracy": 0.717391304347826,
209
- "eval_loss": 0.8987626433372498,
210
- "eval_runtime": 0.607,
211
- "eval_samples_per_second": 75.787,
212
- "eval_steps_per_second": 3.295,
213
  "step": 104
214
  },
215
  {
216
  "epoch": 16.92,
217
- "learning_rate": 0.00017129629629629632,
218
- "loss": 0.5696,
219
  "step": 110
220
  },
221
  {
222
  "epoch": 16.92,
223
- "eval_accuracy": 0.6739130434782609,
224
- "eval_loss": 0.9612070322036743,
225
- "eval_runtime": 0.6266,
226
- "eval_samples_per_second": 73.408,
227
- "eval_steps_per_second": 3.192,
228
  "step": 110
229
  },
230
  {
231
  "epoch": 18.0,
232
  "eval_accuracy": 0.7608695652173914,
233
- "eval_loss": 0.8712936043739319,
234
- "eval_runtime": 0.5923,
235
- "eval_samples_per_second": 77.662,
236
- "eval_steps_per_second": 3.377,
237
  "step": 117
238
  },
239
  {
240
  "epoch": 18.46,
241
- "learning_rate": 0.0001666666666666667,
242
- "loss": 0.5546,
243
  "step": 120
244
  },
245
  {
246
  "epoch": 18.92,
247
- "eval_accuracy": 0.7608695652173914,
248
- "eval_loss": 0.8049858212471008,
249
- "eval_runtime": 0.6052,
250
- "eval_samples_per_second": 76.002,
251
- "eval_steps_per_second": 3.304,
252
  "step": 123
253
  },
254
  {
255
  "epoch": 20.0,
256
- "learning_rate": 0.00016203703703703706,
257
- "loss": 0.4747,
258
  "step": 130
259
  },
260
  {
261
  "epoch": 20.0,
262
- "eval_accuracy": 0.7608695652173914,
263
- "eval_loss": 0.7724586129188538,
264
- "eval_runtime": 0.5976,
265
- "eval_samples_per_second": 76.969,
266
- "eval_steps_per_second": 3.346,
267
  "step": 130
268
  },
269
  {
270
  "epoch": 20.92,
271
- "eval_accuracy": 0.6956521739130435,
272
- "eval_loss": 0.7933102250099182,
273
- "eval_runtime": 0.5936,
274
- "eval_samples_per_second": 77.487,
275
- "eval_steps_per_second": 3.369,
276
  "step": 136
277
  },
278
  {
279
  "epoch": 21.54,
280
- "learning_rate": 0.00015740740740740743,
281
- "loss": 0.4393,
282
  "step": 140
283
  },
284
  {
285
  "epoch": 22.0,
286
- "eval_accuracy": 0.6956521739130435,
287
- "eval_loss": 0.7665086984634399,
288
- "eval_runtime": 0.6062,
289
- "eval_samples_per_second": 75.883,
290
- "eval_steps_per_second": 3.299,
291
  "step": 143
292
  },
293
  {
294
  "epoch": 22.92,
295
  "eval_accuracy": 0.717391304347826,
296
- "eval_loss": 0.7885972857475281,
297
- "eval_runtime": 0.644,
298
- "eval_samples_per_second": 71.425,
299
- "eval_steps_per_second": 3.105,
300
  "step": 149
301
  },
302
  {
303
  "epoch": 23.08,
304
- "learning_rate": 0.00015277777777777777,
305
- "loss": 0.4077,
306
  "step": 150
307
  },
308
  {
309
  "epoch": 24.0,
310
- "eval_accuracy": 0.7391304347826086,
311
- "eval_loss": 0.7824200391769409,
312
- "eval_runtime": 0.6187,
313
- "eval_samples_per_second": 74.344,
314
- "eval_steps_per_second": 3.232,
315
  "step": 156
316
  },
317
  {
318
  "epoch": 24.62,
319
- "learning_rate": 0.00014814814814814815,
320
- "loss": 0.3326,
321
  "step": 160
322
  },
323
  {
324
  "epoch": 24.92,
325
- "eval_accuracy": 0.7391304347826086,
326
- "eval_loss": 0.7020803689956665,
327
- "eval_runtime": 0.6226,
328
- "eval_samples_per_second": 73.885,
329
- "eval_steps_per_second": 3.212,
330
  "step": 162
331
  },
332
  {
333
  "epoch": 26.0,
334
- "eval_accuracy": 0.8260869565217391,
335
- "eval_loss": 0.6346074342727661,
336
- "eval_runtime": 0.611,
337
- "eval_samples_per_second": 75.283,
338
- "eval_steps_per_second": 3.273,
339
  "step": 169
340
  },
341
  {
342
  "epoch": 26.15,
343
- "learning_rate": 0.00014351851851851852,
344
- "loss": 0.315,
345
  "step": 170
346
  },
347
  {
348
  "epoch": 26.92,
349
- "eval_accuracy": 0.8695652173913043,
350
- "eval_loss": 0.6163212060928345,
351
- "eval_runtime": 0.6094,
352
- "eval_samples_per_second": 75.49,
353
- "eval_steps_per_second": 3.282,
354
  "step": 175
355
  },
356
  {
357
  "epoch": 27.69,
358
- "learning_rate": 0.0001388888888888889,
359
- "loss": 0.2729,
360
  "step": 180
361
  },
362
  {
363
  "epoch": 28.0,
364
- "eval_accuracy": 0.8043478260869565,
365
- "eval_loss": 0.6938338875770569,
366
- "eval_runtime": 0.6252,
367
- "eval_samples_per_second": 73.577,
368
- "eval_steps_per_second": 3.199,
369
  "step": 182
370
  },
371
  {
372
  "epoch": 28.92,
373
- "eval_accuracy": 0.8043478260869565,
374
- "eval_loss": 0.7417004704475403,
375
- "eval_runtime": 0.6265,
376
- "eval_samples_per_second": 73.429,
377
- "eval_steps_per_second": 3.193,
378
  "step": 188
379
  },
380
  {
381
  "epoch": 29.23,
382
- "learning_rate": 0.00013425925925925926,
383
- "loss": 0.2218,
384
  "step": 190
385
  },
386
  {
387
  "epoch": 30.0,
388
- "eval_accuracy": 0.782608695652174,
389
- "eval_loss": 0.6668894290924072,
390
- "eval_runtime": 0.6201,
391
- "eval_samples_per_second": 74.182,
392
- "eval_steps_per_second": 3.225,
393
  "step": 195
394
  },
395
  {
396
  "epoch": 30.77,
397
- "learning_rate": 0.00012962962962962963,
398
- "loss": 0.2499,
399
  "step": 200
400
  },
401
  {
402
  "epoch": 30.92,
403
- "eval_accuracy": 0.717391304347826,
404
- "eval_loss": 0.7111229300498962,
405
- "eval_runtime": 0.6369,
406
- "eval_samples_per_second": 72.225,
407
- "eval_steps_per_second": 3.14,
408
  "step": 201
409
  },
410
  {
411
  "epoch": 32.0,
412
  "eval_accuracy": 0.782608695652174,
413
- "eval_loss": 0.6729680299758911,
414
- "eval_runtime": 0.6117,
415
- "eval_samples_per_second": 75.197,
416
- "eval_steps_per_second": 3.269,
417
  "step": 208
418
  },
419
  {
420
  "epoch": 32.31,
421
- "learning_rate": 0.000125,
422
- "loss": 0.2218,
423
  "step": 210
424
  },
425
  {
426
  "epoch": 32.92,
427
- "eval_accuracy": 0.8043478260869565,
428
- "eval_loss": 0.6511849164962769,
429
- "eval_runtime": 0.6114,
430
- "eval_samples_per_second": 75.242,
431
- "eval_steps_per_second": 3.271,
432
  "step": 214
433
  },
434
  {
435
  "epoch": 33.85,
436
- "learning_rate": 0.00012037037037037037,
437
- "loss": 0.2037,
438
  "step": 220
439
  },
440
  {
441
  "epoch": 34.0,
442
- "eval_accuracy": 0.7391304347826086,
443
- "eval_loss": 0.7164542078971863,
444
- "eval_runtime": 0.6112,
445
- "eval_samples_per_second": 75.258,
446
- "eval_steps_per_second": 3.272,
447
  "step": 221
448
  },
449
  {
450
  "epoch": 34.92,
451
- "eval_accuracy": 0.8260869565217391,
452
- "eval_loss": 0.6299737691879272,
453
- "eval_runtime": 0.6021,
454
- "eval_samples_per_second": 76.406,
455
- "eval_steps_per_second": 3.322,
456
  "step": 227
457
  },
458
  {
459
  "epoch": 35.38,
460
- "learning_rate": 0.00011574074074074075,
461
- "loss": 0.2367,
462
  "step": 230
463
  },
464
  {
465
  "epoch": 36.0,
466
- "eval_accuracy": 0.782608695652174,
467
- "eval_loss": 0.7421004176139832,
468
- "eval_runtime": 0.6016,
469
- "eval_samples_per_second": 76.464,
470
- "eval_steps_per_second": 3.325,
471
  "step": 234
472
  },
473
  {
474
  "epoch": 36.92,
475
- "learning_rate": 0.00011111111111111112,
476
- "loss": 0.1835,
477
  "step": 240
478
  },
479
  {
480
  "epoch": 36.92,
481
  "eval_accuracy": 0.8043478260869565,
482
- "eval_loss": 0.6643686294555664,
483
- "eval_runtime": 0.6164,
484
- "eval_samples_per_second": 74.628,
485
- "eval_steps_per_second": 3.245,
486
  "step": 240
487
  },
488
  {
489
  "epoch": 38.0,
490
- "eval_accuracy": 0.8260869565217391,
491
- "eval_loss": 0.6250634789466858,
492
- "eval_runtime": 0.6064,
493
- "eval_samples_per_second": 75.856,
494
- "eval_steps_per_second": 3.298,
495
  "step": 247
496
  },
497
  {
498
  "epoch": 38.46,
499
- "learning_rate": 0.00010648148148148149,
500
- "loss": 0.2073,
501
  "step": 250
502
  },
503
  {
504
  "epoch": 38.92,
505
  "eval_accuracy": 0.782608695652174,
506
- "eval_loss": 0.6430649161338806,
507
- "eval_runtime": 0.6133,
508
- "eval_samples_per_second": 75.001,
509
- "eval_steps_per_second": 3.261,
510
  "step": 253
511
  },
512
  {
513
  "epoch": 40.0,
514
- "learning_rate": 0.00010185185185185186,
515
- "loss": 0.1643,
516
  "step": 260
517
  },
518
  {
519
  "epoch": 40.0,
520
- "eval_accuracy": 0.8043478260869565,
521
- "eval_loss": 0.634781539440155,
522
- "eval_runtime": 0.6208,
523
- "eval_samples_per_second": 74.097,
524
- "eval_steps_per_second": 3.222,
525
  "step": 260
526
  },
527
  {
528
  "epoch": 40.92,
529
- "eval_accuracy": 0.8043478260869565,
530
- "eval_loss": 0.6192363500595093,
531
- "eval_runtime": 0.6183,
532
- "eval_samples_per_second": 74.394,
533
- "eval_steps_per_second": 3.235,
534
  "step": 266
535
  },
536
  {
537
  "epoch": 41.54,
538
- "learning_rate": 9.722222222222223e-05,
539
- "loss": 0.1685,
540
  "step": 270
541
  },
542
  {
543
  "epoch": 42.0,
544
- "eval_accuracy": 0.8043478260869565,
545
- "eval_loss": 0.6753482818603516,
546
- "eval_runtime": 0.5997,
547
- "eval_samples_per_second": 76.701,
548
- "eval_steps_per_second": 3.335,
549
  "step": 273
550
  },
551
  {
552
  "epoch": 42.92,
553
- "eval_accuracy": 0.782608695652174,
554
- "eval_loss": 0.7440155148506165,
555
- "eval_runtime": 0.6117,
556
- "eval_samples_per_second": 75.199,
557
- "eval_steps_per_second": 3.27,
558
  "step": 279
559
  },
560
  {
561
  "epoch": 43.08,
562
- "learning_rate": 9.25925925925926e-05,
563
- "loss": 0.1539,
564
  "step": 280
565
  },
566
  {
567
  "epoch": 44.0,
568
- "eval_accuracy": 0.8043478260869565,
569
- "eval_loss": 0.7504969239234924,
570
- "eval_runtime": 0.6044,
571
- "eval_samples_per_second": 76.107,
572
- "eval_steps_per_second": 3.309,
573
  "step": 286
574
  },
575
  {
576
  "epoch": 44.62,
577
- "learning_rate": 8.796296296296297e-05,
578
- "loss": 0.1658,
579
  "step": 290
580
  },
581
  {
582
  "epoch": 44.92,
583
- "eval_accuracy": 0.8260869565217391,
584
- "eval_loss": 0.6331078410148621,
585
- "eval_runtime": 0.6101,
586
- "eval_samples_per_second": 75.401,
587
- "eval_steps_per_second": 3.278,
588
  "step": 292
589
  },
590
  {
591
  "epoch": 46.0,
592
- "eval_accuracy": 0.8043478260869565,
593
- "eval_loss": 0.6549533009529114,
594
- "eval_runtime": 0.8345,
595
- "eval_samples_per_second": 55.12,
596
- "eval_steps_per_second": 2.397,
597
  "step": 299
598
  },
599
  {
600
  "epoch": 46.15,
601
- "learning_rate": 8.333333333333334e-05,
602
- "loss": 0.1596,
603
  "step": 300
604
  },
605
  {
606
  "epoch": 46.92,
607
- "eval_accuracy": 0.8043478260869565,
608
- "eval_loss": 0.6824274063110352,
609
- "eval_runtime": 0.6289,
610
- "eval_samples_per_second": 73.143,
611
- "eval_steps_per_second": 3.18,
612
  "step": 305
613
  },
614
  {
615
  "epoch": 47.69,
616
- "learning_rate": 7.870370370370372e-05,
617
- "loss": 0.1534,
618
  "step": 310
619
  },
620
  {
621
  "epoch": 48.0,
622
- "eval_accuracy": 0.8043478260869565,
623
- "eval_loss": 0.6970986723899841,
624
- "eval_runtime": 0.6878,
625
- "eval_samples_per_second": 66.875,
626
- "eval_steps_per_second": 2.908,
627
  "step": 312
628
  },
629
  {
630
  "epoch": 48.92,
631
- "eval_accuracy": 0.8260869565217391,
632
- "eval_loss": 0.6347023248672485,
633
- "eval_runtime": 0.6022,
634
- "eval_samples_per_second": 76.388,
635
- "eval_steps_per_second": 3.321,
636
  "step": 318
637
  },
638
  {
639
  "epoch": 49.23,
640
- "learning_rate": 7.407407407407407e-05,
641
- "loss": 0.1677,
642
  "step": 320
643
  },
644
  {
645
  "epoch": 50.0,
646
- "eval_accuracy": 0.8260869565217391,
647
- "eval_loss": 0.6392168998718262,
648
- "eval_runtime": 0.6333,
649
- "eval_samples_per_second": 72.632,
650
- "eval_steps_per_second": 3.158,
651
  "step": 325
652
  },
653
  {
654
  "epoch": 50.77,
655
- "learning_rate": 6.944444444444444e-05,
656
- "loss": 0.1453,
657
  "step": 330
658
  },
659
  {
660
  "epoch": 50.92,
661
- "eval_accuracy": 0.8043478260869565,
662
- "eval_loss": 0.6368798017501831,
663
- "eval_runtime": 0.6455,
664
- "eval_samples_per_second": 71.262,
665
- "eval_steps_per_second": 3.098,
666
  "step": 331
667
  },
668
  {
669
  "epoch": 52.0,
670
- "eval_accuracy": 0.782608695652174,
671
- "eval_loss": 0.623034656047821,
672
- "eval_runtime": 0.7388,
673
- "eval_samples_per_second": 62.259,
674
- "eval_steps_per_second": 2.707,
675
  "step": 338
676
  },
677
  {
678
  "epoch": 52.31,
679
- "learning_rate": 6.481481481481482e-05,
680
- "loss": 0.1385,
681
  "step": 340
682
  },
683
  {
684
  "epoch": 52.92,
685
  "eval_accuracy": 0.7608695652173914,
686
- "eval_loss": 0.6432219743728638,
687
- "eval_runtime": 0.6128,
688
- "eval_samples_per_second": 75.06,
689
- "eval_steps_per_second": 3.263,
690
  "step": 344
691
  },
692
  {
693
  "epoch": 53.85,
694
- "learning_rate": 6.018518518518519e-05,
695
- "loss": 0.1221,
696
  "step": 350
697
  },
698
  {
699
  "epoch": 54.0,
700
- "eval_accuracy": 0.7391304347826086,
701
- "eval_loss": 0.6757408380508423,
702
- "eval_runtime": 0.5947,
703
- "eval_samples_per_second": 77.346,
704
- "eval_steps_per_second": 3.363,
705
  "step": 351
706
  },
707
  {
708
  "epoch": 54.92,
709
  "eval_accuracy": 0.7608695652173914,
710
- "eval_loss": 0.7383140921592712,
711
- "eval_runtime": 0.6233,
712
- "eval_samples_per_second": 73.805,
713
- "eval_steps_per_second": 3.209,
714
  "step": 357
715
  },
716
  {
717
  "epoch": 55.38,
718
- "learning_rate": 5.555555555555556e-05,
719
- "loss": 0.1433,
720
  "step": 360
721
  },
722
  {
723
  "epoch": 56.0,
724
- "eval_accuracy": 0.7608695652173914,
725
- "eval_loss": 0.7100470066070557,
726
- "eval_runtime": 0.6051,
727
- "eval_samples_per_second": 76.02,
728
- "eval_steps_per_second": 3.305,
729
  "step": 364
730
  },
731
  {
732
  "epoch": 56.92,
733
- "learning_rate": 5.092592592592593e-05,
734
- "loss": 0.1567,
735
  "step": 370
736
  },
737
  {
738
  "epoch": 56.92,
739
- "eval_accuracy": 0.7608695652173914,
740
- "eval_loss": 0.6861900091171265,
741
- "eval_runtime": 0.6176,
742
- "eval_samples_per_second": 74.482,
743
- "eval_steps_per_second": 3.238,
744
  "step": 370
745
  },
746
  {
747
  "epoch": 58.0,
748
- "eval_accuracy": 0.7608695652173914,
749
- "eval_loss": 0.6653789281845093,
750
- "eval_runtime": 0.6048,
751
- "eval_samples_per_second": 76.061,
752
- "eval_steps_per_second": 3.307,
753
  "step": 377
754
  },
755
  {
756
  "epoch": 58.46,
757
- "learning_rate": 4.62962962962963e-05,
758
- "loss": 0.1361,
759
  "step": 380
760
  },
761
  {
762
  "epoch": 58.92,
763
- "eval_accuracy": 0.782608695652174,
764
- "eval_loss": 0.66651451587677,
765
- "eval_runtime": 0.6057,
766
- "eval_samples_per_second": 75.942,
767
- "eval_steps_per_second": 3.302,
768
  "step": 383
769
  },
770
  {
771
  "epoch": 60.0,
772
- "learning_rate": 4.166666666666667e-05,
773
- "loss": 0.1157,
774
  "step": 390
775
  },
776
  {
777
  "epoch": 60.0,
778
- "eval_accuracy": 0.782608695652174,
779
- "eval_loss": 0.6439006924629211,
780
- "eval_runtime": 0.6871,
781
- "eval_samples_per_second": 66.95,
782
- "eval_steps_per_second": 2.911,
783
  "step": 390
784
  },
785
  {
786
  "epoch": 60.92,
787
- "eval_accuracy": 0.8260869565217391,
788
- "eval_loss": 0.6306149959564209,
789
- "eval_runtime": 0.6235,
790
- "eval_samples_per_second": 73.782,
791
- "eval_steps_per_second": 3.208,
792
  "step": 396
793
  },
794
  {
795
  "epoch": 61.54,
796
- "learning_rate": 3.7037037037037037e-05,
797
- "loss": 0.0934,
798
  "step": 400
799
  },
800
  {
801
  "epoch": 62.0,
802
- "eval_accuracy": 0.8043478260869565,
803
- "eval_loss": 0.6546051502227783,
804
- "eval_runtime": 0.6648,
805
- "eval_samples_per_second": 69.19,
806
- "eval_steps_per_second": 3.008,
807
  "step": 403
808
  },
809
  {
810
  "epoch": 62.92,
811
- "eval_accuracy": 0.8043478260869565,
812
- "eval_loss": 0.665103018283844,
813
- "eval_runtime": 0.6067,
814
- "eval_samples_per_second": 75.824,
815
- "eval_steps_per_second": 3.297,
816
  "step": 409
817
  },
818
  {
819
  "epoch": 63.08,
820
- "learning_rate": 3.240740740740741e-05,
821
- "loss": 0.1123,
822
  "step": 410
823
  },
824
  {
825
  "epoch": 64.0,
826
- "eval_accuracy": 0.7608695652173914,
827
- "eval_loss": 0.6568355560302734,
828
- "eval_runtime": 0.6061,
829
- "eval_samples_per_second": 75.896,
830
- "eval_steps_per_second": 3.3,
831
  "step": 416
832
  },
833
  {
834
  "epoch": 64.62,
835
- "learning_rate": 2.777777777777778e-05,
836
- "loss": 0.0855,
837
  "step": 420
838
  },
839
  {
840
  "epoch": 64.92,
841
  "eval_accuracy": 0.7608695652173914,
842
- "eval_loss": 0.6506606936454773,
843
- "eval_runtime": 0.6256,
844
- "eval_samples_per_second": 73.528,
845
- "eval_steps_per_second": 3.197,
846
  "step": 422
847
  },
848
  {
849
  "epoch": 66.0,
850
- "eval_accuracy": 0.8043478260869565,
851
- "eval_loss": 0.6667113304138184,
852
- "eval_runtime": 0.6184,
853
- "eval_samples_per_second": 74.384,
854
- "eval_steps_per_second": 3.234,
855
  "step": 429
856
  },
857
  {
858
  "epoch": 66.15,
859
- "learning_rate": 2.314814814814815e-05,
860
- "loss": 0.1135,
861
  "step": 430
862
  },
863
  {
864
  "epoch": 66.92,
865
- "eval_accuracy": 0.782608695652174,
866
- "eval_loss": 0.6515849232673645,
867
- "eval_runtime": 0.6641,
868
- "eval_samples_per_second": 69.267,
869
- "eval_steps_per_second": 3.012,
870
  "step": 435
871
  },
872
  {
873
  "epoch": 67.69,
874
- "learning_rate": 1.8518518518518518e-05,
875
- "loss": 0.0932,
876
  "step": 440
877
  },
878
  {
879
  "epoch": 68.0,
880
- "eval_accuracy": 0.8043478260869565,
881
- "eval_loss": 0.6596114039421082,
882
- "eval_runtime": 0.5982,
883
- "eval_samples_per_second": 76.895,
884
- "eval_steps_per_second": 3.343,
885
  "step": 442
886
  },
887
  {
888
  "epoch": 68.92,
889
- "eval_accuracy": 0.8043478260869565,
890
- "eval_loss": 0.677249550819397,
891
- "eval_runtime": 0.6014,
892
- "eval_samples_per_second": 76.488,
893
- "eval_steps_per_second": 3.326,
894
  "step": 448
895
  },
896
  {
897
  "epoch": 69.23,
898
- "learning_rate": 1.388888888888889e-05,
899
- "loss": 0.1228,
900
  "step": 450
901
  },
902
  {
903
  "epoch": 70.0,
904
- "eval_accuracy": 0.7608695652173914,
905
- "eval_loss": 0.6525955200195312,
906
- "eval_runtime": 0.6064,
907
- "eval_samples_per_second": 75.857,
908
- "eval_steps_per_second": 3.298,
909
  "step": 455
910
  },
911
  {
912
  "epoch": 70.77,
913
- "learning_rate": 9.259259259259259e-06,
914
- "loss": 0.0878,
915
  "step": 460
916
  },
917
  {
918
  "epoch": 70.92,
919
- "eval_accuracy": 0.8260869565217391,
920
- "eval_loss": 0.6731473207473755,
921
- "eval_runtime": 0.6319,
922
- "eval_samples_per_second": 72.794,
923
- "eval_steps_per_second": 3.165,
924
  "step": 461
925
  },
926
  {
927
  "epoch": 72.0,
928
- "eval_accuracy": 0.782608695652174,
929
- "eval_loss": 0.6350728869438171,
930
- "eval_runtime": 0.626,
931
- "eval_samples_per_second": 73.479,
932
- "eval_steps_per_second": 3.195,
933
  "step": 468
934
  },
935
  {
936
  "epoch": 72.31,
937
- "learning_rate": 4.6296296296296296e-06,
938
- "loss": 0.1073,
939
  "step": 470
940
  },
941
  {
942
  "epoch": 72.92,
943
- "eval_accuracy": 0.782608695652174,
944
- "eval_loss": 0.6268617510795593,
945
- "eval_runtime": 0.6109,
946
- "eval_samples_per_second": 75.304,
947
- "eval_steps_per_second": 3.274,
948
  "step": 474
949
  },
950
  {
951
  "epoch": 73.85,
952
- "learning_rate": 0.0,
953
- "loss": 0.1028,
954
  "step": 480
955
  },
956
  {
957
- "epoch": 73.85,
958
- "eval_accuracy": 0.782608695652174,
959
- "eval_loss": 0.6742560267448425,
960
- "eval_runtime": 0.6109,
961
- "eval_samples_per_second": 75.3,
962
- "eval_steps_per_second": 3.274,
963
- "step": 480
964
  },
965
  {
966
- "epoch": 73.85,
967
- "step": 480,
968
- "total_flos": 1.6581977329862246e+17,
969
- "train_loss": 0.38355133185784024,
970
- "train_runtime": 282.7663,
971
- "train_samples_per_second": 231.711,
972
- "train_steps_per_second": 1.698
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
973
  }
974
  ],
975
  "logging_steps": 10,
976
- "max_steps": 480,
977
  "num_input_tokens_seen": 0,
978
- "num_train_epochs": 80,
979
  "save_steps": 500,
980
- "total_flos": 1.6581977329862246e+17,
981
  "train_batch_size": 32,
982
  "trial_name": null,
983
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.8478260869565217,
3
  "best_model_checkpoint": "swiftformer-xs-ve-U13-b-80\\checkpoint-175",
4
+ "epoch": 92.3076923076923,
5
  "eval_steps": 500,
6
+ "global_step": 600,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.92,
13
+ "eval_accuracy": 0.2391304347826087,
14
+ "eval_loss": 1.385766863822937,
15
+ "eval_runtime": 0.6016,
16
+ "eval_samples_per_second": 76.463,
17
+ "eval_steps_per_second": 3.324,
18
  "step": 6
19
  },
20
  {
21
  "epoch": 1.54,
22
+ "learning_rate": 4.9999999999999996e-05,
23
+ "loss": 1.3856,
24
  "step": 10
25
  },
26
  {
27
  "epoch": 2.0,
28
+ "eval_accuracy": 0.2826086956521739,
29
+ "eval_loss": 1.382806420326233,
30
+ "eval_runtime": 0.5574,
31
+ "eval_samples_per_second": 82.519,
32
+ "eval_steps_per_second": 3.588,
33
  "step": 13
34
  },
35
  {
36
  "epoch": 2.92,
37
+ "eval_accuracy": 0.1956521739130435,
38
+ "eval_loss": 1.3768857717514038,
39
+ "eval_runtime": 0.5369,
40
+ "eval_samples_per_second": 85.684,
41
+ "eval_steps_per_second": 3.725,
42
  "step": 19
43
  },
44
  {
45
  "epoch": 3.08,
46
+ "learning_rate": 9.999999999999999e-05,
47
+ "loss": 1.3734,
48
  "step": 20
49
  },
50
  {
51
  "epoch": 4.0,
52
+ "eval_accuracy": 0.13043478260869565,
53
+ "eval_loss": 1.3624433279037476,
54
+ "eval_runtime": 0.5605,
55
+ "eval_samples_per_second": 82.064,
56
+ "eval_steps_per_second": 3.568,
57
  "step": 26
58
  },
59
  {
60
  "epoch": 4.62,
61
+ "learning_rate": 0.00015,
62
+ "loss": 1.2978,
63
  "step": 30
64
  },
65
  {
66
  "epoch": 4.92,
67
+ "eval_accuracy": 0.15217391304347827,
68
+ "eval_loss": 1.3553236722946167,
69
+ "eval_runtime": 0.5527,
70
+ "eval_samples_per_second": 83.233,
71
+ "eval_steps_per_second": 3.619,
72
  "step": 32
73
  },
74
  {
75
  "epoch": 6.0,
76
+ "eval_accuracy": 0.08695652173913043,
77
+ "eval_loss": 1.412070870399475,
78
+ "eval_runtime": 0.5682,
79
+ "eval_samples_per_second": 80.963,
80
+ "eval_steps_per_second": 3.52,
81
  "step": 39
82
  },
83
  {
84
  "epoch": 6.15,
85
+ "learning_rate": 0.00019999999999999998,
86
+ "loss": 1.1702,
87
  "step": 40
88
  },
89
  {
90
  "epoch": 6.92,
91
+ "eval_accuracy": 0.2391304347826087,
92
+ "eval_loss": 1.3719676733016968,
93
+ "eval_runtime": 0.5522,
94
+ "eval_samples_per_second": 83.302,
95
+ "eval_steps_per_second": 3.622,
96
  "step": 45
97
  },
98
  {
99
  "epoch": 7.69,
100
+ "learning_rate": 0.00025,
101
+ "loss": 1.0743,
102
  "step": 50
103
  },
104
  {
105
  "epoch": 8.0,
106
+ "eval_accuracy": 0.34782608695652173,
107
+ "eval_loss": 1.3162317276000977,
108
+ "eval_runtime": 0.5693,
109
+ "eval_samples_per_second": 80.8,
110
+ "eval_steps_per_second": 3.513,
111
  "step": 52
112
  },
113
  {
114
  "epoch": 8.92,
115
+ "eval_accuracy": 0.3695652173913043,
116
+ "eval_loss": 1.2251616716384888,
117
+ "eval_runtime": 0.6192,
118
+ "eval_samples_per_second": 74.295,
119
+ "eval_steps_per_second": 3.23,
120
  "step": 58
121
  },
122
  {
123
  "epoch": 9.23,
124
+ "learning_rate": 0.0003,
125
+ "loss": 0.9504,
126
  "step": 60
127
  },
128
  {
129
  "epoch": 10.0,
130
+ "eval_accuracy": 0.43478260869565216,
131
+ "eval_loss": 1.168912649154663,
132
+ "eval_runtime": 0.6842,
133
+ "eval_samples_per_second": 67.23,
134
+ "eval_steps_per_second": 2.923,
135
  "step": 65
136
  },
137
  {
138
  "epoch": 10.77,
139
+ "learning_rate": 0.00029444444444444445,
140
+ "loss": 0.8305,
141
  "step": 70
142
  },
143
  {
144
  "epoch": 10.92,
145
+ "eval_accuracy": 0.5869565217391305,
146
+ "eval_loss": 1.0516129732131958,
147
+ "eval_runtime": 0.5642,
148
+ "eval_samples_per_second": 81.528,
149
+ "eval_steps_per_second": 3.545,
150
  "step": 71
151
  },
152
  {
153
  "epoch": 12.0,
154
+ "eval_accuracy": 0.6739130434782609,
155
+ "eval_loss": 0.9548317193984985,
156
+ "eval_runtime": 0.5661,
157
+ "eval_samples_per_second": 81.254,
158
+ "eval_steps_per_second": 3.533,
159
  "step": 78
160
  },
161
  {
162
  "epoch": 12.31,
163
+ "learning_rate": 0.0002888888888888888,
164
+ "loss": 0.7374,
165
  "step": 80
166
  },
167
  {
168
  "epoch": 12.92,
169
+ "eval_accuracy": 0.717391304347826,
170
+ "eval_loss": 0.9137820601463318,
171
+ "eval_runtime": 0.5482,
172
+ "eval_samples_per_second": 83.916,
173
+ "eval_steps_per_second": 3.649,
174
  "step": 84
175
  },
176
  {
177
  "epoch": 13.85,
178
+ "learning_rate": 0.0002833333333333333,
179
+ "loss": 0.6207,
180
  "step": 90
181
  },
182
  {
183
  "epoch": 14.0,
184
+ "eval_accuracy": 0.6521739130434783,
185
+ "eval_loss": 0.9353066086769104,
186
+ "eval_runtime": 0.5551,
187
+ "eval_samples_per_second": 82.864,
188
+ "eval_steps_per_second": 3.603,
189
  "step": 91
190
  },
191
  {
192
  "epoch": 14.92,
193
+ "eval_accuracy": 0.6739130434782609,
194
+ "eval_loss": 0.8639649152755737,
195
+ "eval_runtime": 0.5605,
196
+ "eval_samples_per_second": 82.076,
197
+ "eval_steps_per_second": 3.569,
198
  "step": 97
199
  },
200
  {
201
  "epoch": 15.38,
202
+ "learning_rate": 0.0002777777777777778,
203
+ "loss": 0.5184,
204
  "step": 100
205
  },
206
  {
207
  "epoch": 16.0,
208
+ "eval_accuracy": 0.782608695652174,
209
+ "eval_loss": 0.8121959567070007,
210
+ "eval_runtime": 0.5575,
211
+ "eval_samples_per_second": 82.517,
212
+ "eval_steps_per_second": 3.588,
213
  "step": 104
214
  },
215
  {
216
  "epoch": 16.92,
217
+ "learning_rate": 0.0002722222222222222,
218
+ "loss": 0.4606,
219
  "step": 110
220
  },
221
  {
222
  "epoch": 16.92,
223
+ "eval_accuracy": 0.8043478260869565,
224
+ "eval_loss": 0.7136393189430237,
225
+ "eval_runtime": 0.5537,
226
+ "eval_samples_per_second": 83.08,
227
+ "eval_steps_per_second": 3.612,
228
  "step": 110
229
  },
230
  {
231
  "epoch": 18.0,
232
  "eval_accuracy": 0.7608695652173914,
233
+ "eval_loss": 0.7955167293548584,
234
+ "eval_runtime": 0.5778,
235
+ "eval_samples_per_second": 79.616,
236
+ "eval_steps_per_second": 3.462,
237
  "step": 117
238
  },
239
  {
240
  "epoch": 18.46,
241
+ "learning_rate": 0.0002666666666666666,
242
+ "loss": 0.4332,
243
  "step": 120
244
  },
245
  {
246
  "epoch": 18.92,
247
+ "eval_accuracy": 0.6956521739130435,
248
+ "eval_loss": 0.7789567112922668,
249
+ "eval_runtime": 0.5435,
250
+ "eval_samples_per_second": 84.632,
251
+ "eval_steps_per_second": 3.68,
252
  "step": 123
253
  },
254
  {
255
  "epoch": 20.0,
256
+ "learning_rate": 0.0002611111111111111,
257
+ "loss": 0.3315,
258
  "step": 130
259
  },
260
  {
261
  "epoch": 20.0,
262
+ "eval_accuracy": 0.7391304347826086,
263
+ "eval_loss": 0.8116961717605591,
264
+ "eval_runtime": 0.6158,
265
+ "eval_samples_per_second": 74.701,
266
+ "eval_steps_per_second": 3.248,
267
  "step": 130
268
  },
269
  {
270
  "epoch": 20.92,
271
+ "eval_accuracy": 0.7608695652173914,
272
+ "eval_loss": 0.8068195581436157,
273
+ "eval_runtime": 0.5397,
274
+ "eval_samples_per_second": 85.24,
275
+ "eval_steps_per_second": 3.706,
276
  "step": 136
277
  },
278
  {
279
  "epoch": 21.54,
280
+ "learning_rate": 0.00025555555555555553,
281
+ "loss": 0.3229,
282
  "step": 140
283
  },
284
  {
285
  "epoch": 22.0,
286
+ "eval_accuracy": 0.782608695652174,
287
+ "eval_loss": 0.8786293268203735,
288
+ "eval_runtime": 0.5594,
289
+ "eval_samples_per_second": 82.233,
290
+ "eval_steps_per_second": 3.575,
291
  "step": 143
292
  },
293
  {
294
  "epoch": 22.92,
295
  "eval_accuracy": 0.717391304347826,
296
+ "eval_loss": 0.9029754996299744,
297
+ "eval_runtime": 0.5517,
298
+ "eval_samples_per_second": 83.385,
299
+ "eval_steps_per_second": 3.625,
300
  "step": 149
301
  },
302
  {
303
  "epoch": 23.08,
304
+ "learning_rate": 0.00025,
305
+ "loss": 0.3065,
306
  "step": 150
307
  },
308
  {
309
  "epoch": 24.0,
310
+ "eval_accuracy": 0.6521739130434783,
311
+ "eval_loss": 0.8253074884414673,
312
+ "eval_runtime": 0.578,
313
+ "eval_samples_per_second": 79.588,
314
+ "eval_steps_per_second": 3.46,
315
  "step": 156
316
  },
317
  {
318
  "epoch": 24.62,
319
+ "learning_rate": 0.00024444444444444443,
320
+ "loss": 0.2315,
321
  "step": 160
322
  },
323
  {
324
  "epoch": 24.92,
325
+ "eval_accuracy": 0.8043478260869565,
326
+ "eval_loss": 0.739804208278656,
327
+ "eval_runtime": 0.5608,
328
+ "eval_samples_per_second": 82.024,
329
+ "eval_steps_per_second": 3.566,
330
  "step": 162
331
  },
332
  {
333
  "epoch": 26.0,
334
+ "eval_accuracy": 0.7608695652173914,
335
+ "eval_loss": 0.7938594222068787,
336
+ "eval_runtime": 0.5501,
337
+ "eval_samples_per_second": 83.615,
338
+ "eval_steps_per_second": 3.635,
339
  "step": 169
340
  },
341
  {
342
  "epoch": 26.15,
343
+ "learning_rate": 0.00023888888888888885,
344
+ "loss": 0.222,
345
  "step": 170
346
  },
347
  {
348
  "epoch": 26.92,
349
+ "eval_accuracy": 0.8478260869565217,
350
+ "eval_loss": 0.6640377044677734,
351
+ "eval_runtime": 0.5452,
352
+ "eval_samples_per_second": 84.37,
353
+ "eval_steps_per_second": 3.668,
354
  "step": 175
355
  },
356
  {
357
  "epoch": 27.69,
358
+ "learning_rate": 0.0002333333333333333,
359
+ "loss": 0.1756,
360
  "step": 180
361
  },
362
  {
363
  "epoch": 28.0,
364
+ "eval_accuracy": 0.7391304347826086,
365
+ "eval_loss": 0.850994348526001,
366
+ "eval_runtime": 0.5442,
367
+ "eval_samples_per_second": 84.525,
368
+ "eval_steps_per_second": 3.675,
369
  "step": 182
370
  },
371
  {
372
  "epoch": 28.92,
373
+ "eval_accuracy": 0.717391304347826,
374
+ "eval_loss": 0.9861403107643127,
375
+ "eval_runtime": 0.5468,
376
+ "eval_samples_per_second": 84.118,
377
+ "eval_steps_per_second": 3.657,
378
  "step": 188
379
  },
380
  {
381
  "epoch": 29.23,
382
+ "learning_rate": 0.00022777777777777778,
383
+ "loss": 0.1702,
384
  "step": 190
385
  },
386
  {
387
  "epoch": 30.0,
388
+ "eval_accuracy": 0.7608695652173914,
389
+ "eval_loss": 1.1060209274291992,
390
+ "eval_runtime": 0.5662,
391
+ "eval_samples_per_second": 81.238,
392
+ "eval_steps_per_second": 3.532,
393
  "step": 195
394
  },
395
  {
396
  "epoch": 30.77,
397
+ "learning_rate": 0.00022222222222222218,
398
+ "loss": 0.202,
399
  "step": 200
400
  },
401
  {
402
  "epoch": 30.92,
403
+ "eval_accuracy": 0.7391304347826086,
404
+ "eval_loss": 1.0928934812545776,
405
+ "eval_runtime": 0.5568,
406
+ "eval_samples_per_second": 82.617,
407
+ "eval_steps_per_second": 3.592,
408
  "step": 201
409
  },
410
  {
411
  "epoch": 32.0,
412
  "eval_accuracy": 0.782608695652174,
413
+ "eval_loss": 0.8670152425765991,
414
+ "eval_runtime": 0.5807,
415
+ "eval_samples_per_second": 79.209,
416
+ "eval_steps_per_second": 3.444,
417
  "step": 208
418
  },
419
  {
420
  "epoch": 32.31,
421
+ "learning_rate": 0.00021666666666666666,
422
+ "loss": 0.1665,
423
  "step": 210
424
  },
425
  {
426
  "epoch": 32.92,
427
+ "eval_accuracy": 0.7608695652173914,
428
+ "eval_loss": 0.8032902479171753,
429
+ "eval_runtime": 0.5451,
430
+ "eval_samples_per_second": 84.381,
431
+ "eval_steps_per_second": 3.669,
432
  "step": 214
433
  },
434
  {
435
  "epoch": 33.85,
436
+ "learning_rate": 0.0002111111111111111,
437
+ "loss": 0.1695,
438
  "step": 220
439
  },
440
  {
441
  "epoch": 34.0,
442
+ "eval_accuracy": 0.782608695652174,
443
+ "eval_loss": 0.7234715819358826,
444
+ "eval_runtime": 0.5632,
445
+ "eval_samples_per_second": 81.673,
446
+ "eval_steps_per_second": 3.551,
447
  "step": 221
448
  },
449
  {
450
  "epoch": 34.92,
451
+ "eval_accuracy": 0.7608695652173914,
452
+ "eval_loss": 0.8916966915130615,
453
+ "eval_runtime": 0.5518,
454
+ "eval_samples_per_second": 83.363,
455
+ "eval_steps_per_second": 3.624,
456
  "step": 227
457
  },
458
  {
459
  "epoch": 35.38,
460
+ "learning_rate": 0.00020555555555555556,
461
+ "loss": 0.1807,
462
  "step": 230
463
  },
464
  {
465
  "epoch": 36.0,
466
+ "eval_accuracy": 0.7391304347826086,
467
+ "eval_loss": 0.9215289950370789,
468
+ "eval_runtime": 0.5536,
469
+ "eval_samples_per_second": 83.099,
470
+ "eval_steps_per_second": 3.613,
471
  "step": 234
472
  },
473
  {
474
  "epoch": 36.92,
475
+ "learning_rate": 0.00019999999999999998,
476
+ "loss": 0.1289,
477
  "step": 240
478
  },
479
  {
480
  "epoch": 36.92,
481
  "eval_accuracy": 0.8043478260869565,
482
+ "eval_loss": 0.8231099247932434,
483
+ "eval_runtime": 0.5624,
484
+ "eval_samples_per_second": 81.787,
485
+ "eval_steps_per_second": 3.556,
486
  "step": 240
487
  },
488
  {
489
  "epoch": 38.0,
490
+ "eval_accuracy": 0.782608695652174,
491
+ "eval_loss": 0.9255649447441101,
492
+ "eval_runtime": 0.5579,
493
+ "eval_samples_per_second": 82.459,
494
+ "eval_steps_per_second": 3.585,
495
  "step": 247
496
  },
497
  {
498
  "epoch": 38.46,
499
+ "learning_rate": 0.00019444444444444443,
500
+ "loss": 0.145,
501
  "step": 250
502
  },
503
  {
504
  "epoch": 38.92,
505
  "eval_accuracy": 0.782608695652174,
506
+ "eval_loss": 0.8866252303123474,
507
+ "eval_runtime": 0.5578,
508
+ "eval_samples_per_second": 82.469,
509
+ "eval_steps_per_second": 3.586,
510
  "step": 253
511
  },
512
  {
513
  "epoch": 40.0,
514
+ "learning_rate": 0.00018888888888888888,
515
+ "loss": 0.1422,
516
  "step": 260
517
  },
518
  {
519
  "epoch": 40.0,
520
+ "eval_accuracy": 0.8260869565217391,
521
+ "eval_loss": 0.8511058688163757,
522
+ "eval_runtime": 0.5658,
523
+ "eval_samples_per_second": 81.305,
524
+ "eval_steps_per_second": 3.535,
525
  "step": 260
526
  },
527
  {
528
  "epoch": 40.92,
529
+ "eval_accuracy": 0.7391304347826086,
530
+ "eval_loss": 0.9955706596374512,
531
+ "eval_runtime": 0.5459,
532
+ "eval_samples_per_second": 84.265,
533
+ "eval_steps_per_second": 3.664,
534
  "step": 266
535
  },
536
  {
537
  "epoch": 41.54,
538
+ "learning_rate": 0.00018333333333333334,
539
+ "loss": 0.1313,
540
  "step": 270
541
  },
542
  {
543
  "epoch": 42.0,
544
+ "eval_accuracy": 0.7391304347826086,
545
+ "eval_loss": 1.3004724979400635,
546
+ "eval_runtime": 0.5565,
547
+ "eval_samples_per_second": 82.654,
548
+ "eval_steps_per_second": 3.594,
549
  "step": 273
550
  },
551
  {
552
  "epoch": 42.92,
553
+ "eval_accuracy": 0.6739130434782609,
554
+ "eval_loss": 1.1531860828399658,
555
+ "eval_runtime": 0.5577,
556
+ "eval_samples_per_second": 82.479,
557
+ "eval_steps_per_second": 3.586,
558
  "step": 279
559
  },
560
  {
561
  "epoch": 43.08,
562
+ "learning_rate": 0.00017777777777777776,
563
+ "loss": 0.1128,
564
  "step": 280
565
  },
566
  {
567
  "epoch": 44.0,
568
+ "eval_accuracy": 0.7391304347826086,
569
+ "eval_loss": 1.0891470909118652,
570
+ "eval_runtime": 0.5553,
571
+ "eval_samples_per_second": 82.842,
572
+ "eval_steps_per_second": 3.602,
573
  "step": 286
574
  },
575
  {
576
  "epoch": 44.62,
577
+ "learning_rate": 0.0001722222222222222,
578
+ "loss": 0.1213,
579
  "step": 290
580
  },
581
  {
582
  "epoch": 44.92,
583
+ "eval_accuracy": 0.7391304347826086,
584
+ "eval_loss": 1.0764859914779663,
585
+ "eval_runtime": 0.5645,
586
+ "eval_samples_per_second": 81.483,
587
+ "eval_steps_per_second": 3.543,
588
  "step": 292
589
  },
590
  {
591
  "epoch": 46.0,
592
+ "eval_accuracy": 0.7391304347826086,
593
+ "eval_loss": 0.9141591191291809,
594
+ "eval_runtime": 0.5767,
595
+ "eval_samples_per_second": 79.768,
596
+ "eval_steps_per_second": 3.468,
597
  "step": 299
598
  },
599
  {
600
  "epoch": 46.15,
601
+ "learning_rate": 0.00016666666666666666,
602
+ "loss": 0.1161,
603
  "step": 300
604
  },
605
  {
606
  "epoch": 46.92,
607
+ "eval_accuracy": 0.717391304347826,
608
+ "eval_loss": 0.909971296787262,
609
+ "eval_runtime": 0.5553,
610
+ "eval_samples_per_second": 82.834,
611
+ "eval_steps_per_second": 3.601,
612
  "step": 305
613
  },
614
  {
615
  "epoch": 47.69,
616
+ "learning_rate": 0.0001611111111111111,
617
+ "loss": 0.1123,
618
  "step": 310
619
  },
620
  {
621
  "epoch": 48.0,
622
+ "eval_accuracy": 0.782608695652174,
623
+ "eval_loss": 0.8906996250152588,
624
+ "eval_runtime": 0.545,
625
+ "eval_samples_per_second": 84.409,
626
+ "eval_steps_per_second": 3.67,
627
  "step": 312
628
  },
629
  {
630
  "epoch": 48.92,
631
+ "eval_accuracy": 0.7608695652173914,
632
+ "eval_loss": 0.9461659789085388,
633
+ "eval_runtime": 0.5617,
634
+ "eval_samples_per_second": 81.888,
635
+ "eval_steps_per_second": 3.56,
636
  "step": 318
637
  },
638
  {
639
  "epoch": 49.23,
640
+ "learning_rate": 0.00015555555555555554,
641
+ "loss": 0.1107,
642
  "step": 320
643
  },
644
  {
645
  "epoch": 50.0,
646
+ "eval_accuracy": 0.7391304347826086,
647
+ "eval_loss": 0.8591524958610535,
648
+ "eval_runtime": 0.5836,
649
+ "eval_samples_per_second": 78.822,
650
+ "eval_steps_per_second": 3.427,
651
  "step": 325
652
  },
653
  {
654
  "epoch": 50.77,
655
+ "learning_rate": 0.00015,
656
+ "loss": 0.0915,
657
  "step": 330
658
  },
659
  {
660
  "epoch": 50.92,
661
+ "eval_accuracy": 0.7608695652173914,
662
+ "eval_loss": 0.9894090890884399,
663
+ "eval_runtime": 0.5586,
664
+ "eval_samples_per_second": 82.348,
665
+ "eval_steps_per_second": 3.58,
666
  "step": 331
667
  },
668
  {
669
  "epoch": 52.0,
670
+ "eval_accuracy": 0.7608695652173914,
671
+ "eval_loss": 1.1094378232955933,
672
+ "eval_runtime": 0.5369,
673
+ "eval_samples_per_second": 85.67,
674
+ "eval_steps_per_second": 3.725,
675
  "step": 338
676
  },
677
  {
678
  "epoch": 52.31,
679
+ "learning_rate": 0.0001444444444444444,
680
+ "loss": 0.0981,
681
  "step": 340
682
  },
683
  {
684
  "epoch": 52.92,
685
  "eval_accuracy": 0.7608695652173914,
686
+ "eval_loss": 1.1955540180206299,
687
+ "eval_runtime": 0.5652,
688
+ "eval_samples_per_second": 81.388,
689
+ "eval_steps_per_second": 3.539,
690
  "step": 344
691
  },
692
  {
693
  "epoch": 53.85,
694
+ "learning_rate": 0.0001388888888888889,
695
+ "loss": 0.0762,
696
  "step": 350
697
  },
698
  {
699
  "epoch": 54.0,
700
+ "eval_accuracy": 0.782608695652174,
701
+ "eval_loss": 1.0078867673873901,
702
+ "eval_runtime": 0.5486,
703
+ "eval_samples_per_second": 83.844,
704
+ "eval_steps_per_second": 3.645,
705
  "step": 351
706
  },
707
  {
708
  "epoch": 54.92,
709
  "eval_accuracy": 0.7608695652173914,
710
+ "eval_loss": 0.9898611307144165,
711
+ "eval_runtime": 0.5586,
712
+ "eval_samples_per_second": 82.345,
713
+ "eval_steps_per_second": 3.58,
714
  "step": 357
715
  },
716
  {
717
  "epoch": 55.38,
718
+ "learning_rate": 0.0001333333333333333,
719
+ "loss": 0.1083,
720
  "step": 360
721
  },
722
  {
723
  "epoch": 56.0,
724
+ "eval_accuracy": 0.782608695652174,
725
+ "eval_loss": 0.9164445996284485,
726
+ "eval_runtime": 0.5694,
727
+ "eval_samples_per_second": 80.79,
728
+ "eval_steps_per_second": 3.513,
729
  "step": 364
730
  },
731
  {
732
  "epoch": 56.92,
733
+ "learning_rate": 0.00012777777777777776,
734
+ "loss": 0.1087,
735
  "step": 370
736
  },
737
  {
738
  "epoch": 56.92,
739
+ "eval_accuracy": 0.782608695652174,
740
+ "eval_loss": 0.9263260960578918,
741
+ "eval_runtime": 0.567,
742
+ "eval_samples_per_second": 81.133,
743
+ "eval_steps_per_second": 3.528,
744
  "step": 370
745
  },
746
  {
747
  "epoch": 58.0,
748
+ "eval_accuracy": 0.7391304347826086,
749
+ "eval_loss": 0.9160030484199524,
750
+ "eval_runtime": 0.6023,
751
+ "eval_samples_per_second": 76.371,
752
+ "eval_steps_per_second": 3.32,
753
  "step": 377
754
  },
755
  {
756
  "epoch": 58.46,
757
+ "learning_rate": 0.00012222222222222221,
758
+ "loss": 0.0871,
759
  "step": 380
760
  },
761
  {
762
  "epoch": 58.92,
763
+ "eval_accuracy": 0.717391304347826,
764
+ "eval_loss": 1.0178961753845215,
765
+ "eval_runtime": 0.5547,
766
+ "eval_samples_per_second": 82.923,
767
+ "eval_steps_per_second": 3.605,
768
  "step": 383
769
  },
770
  {
771
  "epoch": 60.0,
772
+ "learning_rate": 0.00011666666666666665,
773
+ "loss": 0.0852,
774
  "step": 390
775
  },
776
  {
777
  "epoch": 60.0,
778
+ "eval_accuracy": 0.7391304347826086,
779
+ "eval_loss": 0.9246215224266052,
780
+ "eval_runtime": 0.5641,
781
+ "eval_samples_per_second": 81.551,
782
+ "eval_steps_per_second": 3.546,
783
  "step": 390
784
  },
785
  {
786
  "epoch": 60.92,
787
+ "eval_accuracy": 0.8043478260869565,
788
+ "eval_loss": 0.8928539156913757,
789
+ "eval_runtime": 0.5647,
790
+ "eval_samples_per_second": 81.457,
791
+ "eval_steps_per_second": 3.542,
792
  "step": 396
793
  },
794
  {
795
  "epoch": 61.54,
796
+ "learning_rate": 0.00011111111111111109,
797
+ "loss": 0.0613,
798
  "step": 400
799
  },
800
  {
801
  "epoch": 62.0,
802
+ "eval_accuracy": 0.717391304347826,
803
+ "eval_loss": 0.9988526105880737,
804
+ "eval_runtime": 0.5653,
805
+ "eval_samples_per_second": 81.37,
806
+ "eval_steps_per_second": 3.538,
807
  "step": 403
808
  },
809
  {
810
  "epoch": 62.92,
811
+ "eval_accuracy": 0.717391304347826,
812
+ "eval_loss": 1.0367439985275269,
813
+ "eval_runtime": 0.5583,
814
+ "eval_samples_per_second": 82.397,
815
+ "eval_steps_per_second": 3.582,
816
  "step": 409
817
  },
818
  {
819
  "epoch": 63.08,
820
+ "learning_rate": 0.00010555555555555555,
821
+ "loss": 0.0899,
822
  "step": 410
823
  },
824
  {
825
  "epoch": 64.0,
826
+ "eval_accuracy": 0.6956521739130435,
827
+ "eval_loss": 1.1212767362594604,
828
+ "eval_runtime": 0.6169,
829
+ "eval_samples_per_second": 74.561,
830
+ "eval_steps_per_second": 3.242,
831
  "step": 416
832
  },
833
  {
834
  "epoch": 64.62,
835
+ "learning_rate": 9.999999999999999e-05,
836
+ "loss": 0.0669,
837
  "step": 420
838
  },
839
  {
840
  "epoch": 64.92,
841
  "eval_accuracy": 0.7608695652173914,
842
+ "eval_loss": 1.0092602968215942,
843
+ "eval_runtime": 0.5458,
844
+ "eval_samples_per_second": 84.281,
845
+ "eval_steps_per_second": 3.664,
846
  "step": 422
847
  },
848
  {
849
  "epoch": 66.0,
850
+ "eval_accuracy": 0.7391304347826086,
851
+ "eval_loss": 1.0128601789474487,
852
+ "eval_runtime": 0.5653,
853
+ "eval_samples_per_second": 81.375,
854
+ "eval_steps_per_second": 3.538,
855
  "step": 429
856
  },
857
  {
858
  "epoch": 66.15,
859
+ "learning_rate": 9.444444444444444e-05,
860
+ "loss": 0.0791,
861
  "step": 430
862
  },
863
  {
864
  "epoch": 66.92,
865
+ "eval_accuracy": 0.717391304347826,
866
+ "eval_loss": 0.9979057908058167,
867
+ "eval_runtime": 0.5593,
868
+ "eval_samples_per_second": 82.241,
869
+ "eval_steps_per_second": 3.576,
870
  "step": 435
871
  },
872
  {
873
  "epoch": 67.69,
874
+ "learning_rate": 8.888888888888888e-05,
875
+ "loss": 0.0848,
876
  "step": 440
877
  },
878
  {
879
  "epoch": 68.0,
880
+ "eval_accuracy": 0.7391304347826086,
881
+ "eval_loss": 1.0137319564819336,
882
+ "eval_runtime": 0.5679,
883
+ "eval_samples_per_second": 81.003,
884
+ "eval_steps_per_second": 3.522,
885
  "step": 442
886
  },
887
  {
888
  "epoch": 68.92,
889
+ "eval_accuracy": 0.6956521739130435,
890
+ "eval_loss": 1.0760867595672607,
891
+ "eval_runtime": 0.5572,
892
+ "eval_samples_per_second": 82.55,
893
+ "eval_steps_per_second": 3.589,
894
  "step": 448
895
  },
896
  {
897
  "epoch": 69.23,
898
+ "learning_rate": 8.333333333333333e-05,
899
+ "loss": 0.0799,
900
  "step": 450
901
  },
902
  {
903
  "epoch": 70.0,
904
+ "eval_accuracy": 0.6956521739130435,
905
+ "eval_loss": 1.0152040719985962,
906
+ "eval_runtime": 0.56,
907
+ "eval_samples_per_second": 82.145,
908
+ "eval_steps_per_second": 3.572,
909
  "step": 455
910
  },
911
  {
912
  "epoch": 70.77,
913
+ "learning_rate": 7.777777777777777e-05,
914
+ "loss": 0.0727,
915
  "step": 460
916
  },
917
  {
918
  "epoch": 70.92,
919
+ "eval_accuracy": 0.6956521739130435,
920
+ "eval_loss": 1.130175232887268,
921
+ "eval_runtime": 0.5554,
922
+ "eval_samples_per_second": 82.818,
923
+ "eval_steps_per_second": 3.601,
924
  "step": 461
925
  },
926
  {
927
  "epoch": 72.0,
928
+ "eval_accuracy": 0.717391304347826,
929
+ "eval_loss": 1.046817421913147,
930
+ "eval_runtime": 0.5592,
931
+ "eval_samples_per_second": 82.253,
932
+ "eval_steps_per_second": 3.576,
933
  "step": 468
934
  },
935
  {
936
  "epoch": 72.31,
937
+ "learning_rate": 7.22222222222222e-05,
938
+ "loss": 0.0763,
939
  "step": 470
940
  },
941
  {
942
  "epoch": 72.92,
943
+ "eval_accuracy": 0.6739130434782609,
944
+ "eval_loss": 1.0759297609329224,
945
+ "eval_runtime": 0.5607,
946
+ "eval_samples_per_second": 82.046,
947
+ "eval_steps_per_second": 3.567,
948
  "step": 474
949
  },
950
  {
951
  "epoch": 73.85,
952
+ "learning_rate": 6.666666666666666e-05,
953
+ "loss": 0.06,
954
  "step": 480
955
  },
956
  {
957
+ "epoch": 74.0,
958
+ "eval_accuracy": 0.717391304347826,
959
+ "eval_loss": 1.0803240537643433,
960
+ "eval_runtime": 0.553,
961
+ "eval_samples_per_second": 83.183,
962
+ "eval_steps_per_second": 3.617,
963
+ "step": 481
964
  },
965
  {
966
+ "epoch": 74.92,
967
+ "eval_accuracy": 0.6956521739130435,
968
+ "eval_loss": 1.0483999252319336,
969
+ "eval_runtime": 0.5749,
970
+ "eval_samples_per_second": 80.02,
971
+ "eval_steps_per_second": 3.479,
972
+ "step": 487
973
+ },
974
+ {
975
+ "epoch": 75.38,
976
+ "learning_rate": 6.111111111111111e-05,
977
+ "loss": 0.0746,
978
+ "step": 490
979
+ },
980
+ {
981
+ "epoch": 76.0,
982
+ "eval_accuracy": 0.717391304347826,
983
+ "eval_loss": 0.999907374382019,
984
+ "eval_runtime": 0.6218,
985
+ "eval_samples_per_second": 73.983,
986
+ "eval_steps_per_second": 3.217,
987
+ "step": 494
988
+ },
989
+ {
990
+ "epoch": 76.92,
991
+ "learning_rate": 5.5555555555555545e-05,
992
+ "loss": 0.0687,
993
+ "step": 500
994
+ },
995
+ {
996
+ "epoch": 76.92,
997
+ "eval_accuracy": 0.717391304347826,
998
+ "eval_loss": 0.993706226348877,
999
+ "eval_runtime": 0.5757,
1000
+ "eval_samples_per_second": 79.899,
1001
+ "eval_steps_per_second": 3.474,
1002
+ "step": 500
1003
+ },
1004
+ {
1005
+ "epoch": 78.0,
1006
+ "eval_accuracy": 0.6956521739130435,
1007
+ "eval_loss": 1.1188522577285767,
1008
+ "eval_runtime": 0.5575,
1009
+ "eval_samples_per_second": 82.513,
1010
+ "eval_steps_per_second": 3.588,
1011
+ "step": 507
1012
+ },
1013
+ {
1014
+ "epoch": 78.46,
1015
+ "learning_rate": 4.9999999999999996e-05,
1016
+ "loss": 0.0761,
1017
+ "step": 510
1018
+ },
1019
+ {
1020
+ "epoch": 78.92,
1021
+ "eval_accuracy": 0.6956521739130435,
1022
+ "eval_loss": 1.1013031005859375,
1023
+ "eval_runtime": 0.6076,
1024
+ "eval_samples_per_second": 75.709,
1025
+ "eval_steps_per_second": 3.292,
1026
+ "step": 513
1027
+ },
1028
+ {
1029
+ "epoch": 80.0,
1030
+ "learning_rate": 4.444444444444444e-05,
1031
+ "loss": 0.0729,
1032
+ "step": 520
1033
+ },
1034
+ {
1035
+ "epoch": 80.0,
1036
+ "eval_accuracy": 0.6956521739130435,
1037
+ "eval_loss": 1.029362678527832,
1038
+ "eval_runtime": 0.5744,
1039
+ "eval_samples_per_second": 80.09,
1040
+ "eval_steps_per_second": 3.482,
1041
+ "step": 520
1042
+ },
1043
+ {
1044
+ "epoch": 80.92,
1045
+ "eval_accuracy": 0.717391304347826,
1046
+ "eval_loss": 1.0859981775283813,
1047
+ "eval_runtime": 0.5737,
1048
+ "eval_samples_per_second": 80.176,
1049
+ "eval_steps_per_second": 3.486,
1050
+ "step": 526
1051
+ },
1052
+ {
1053
+ "epoch": 81.54,
1054
+ "learning_rate": 3.8888888888888884e-05,
1055
+ "loss": 0.0472,
1056
+ "step": 530
1057
+ },
1058
+ {
1059
+ "epoch": 82.0,
1060
+ "eval_accuracy": 0.717391304347826,
1061
+ "eval_loss": 1.03272545337677,
1062
+ "eval_runtime": 0.5693,
1063
+ "eval_samples_per_second": 80.807,
1064
+ "eval_steps_per_second": 3.513,
1065
+ "step": 533
1066
+ },
1067
+ {
1068
+ "epoch": 82.92,
1069
+ "eval_accuracy": 0.717391304347826,
1070
+ "eval_loss": 1.0224952697753906,
1071
+ "eval_runtime": 0.5731,
1072
+ "eval_samples_per_second": 80.269,
1073
+ "eval_steps_per_second": 3.49,
1074
+ "step": 539
1075
+ },
1076
+ {
1077
+ "epoch": 83.08,
1078
+ "learning_rate": 3.333333333333333e-05,
1079
+ "loss": 0.0519,
1080
+ "step": 540
1081
+ },
1082
+ {
1083
+ "epoch": 84.0,
1084
+ "eval_accuracy": 0.6956521739130435,
1085
+ "eval_loss": 1.1345374584197998,
1086
+ "eval_runtime": 0.5627,
1087
+ "eval_samples_per_second": 81.748,
1088
+ "eval_steps_per_second": 3.554,
1089
+ "step": 546
1090
+ },
1091
+ {
1092
+ "epoch": 84.62,
1093
+ "learning_rate": 2.7777777777777772e-05,
1094
+ "loss": 0.0688,
1095
+ "step": 550
1096
+ },
1097
+ {
1098
+ "epoch": 84.92,
1099
+ "eval_accuracy": 0.6956521739130435,
1100
+ "eval_loss": 1.092256784439087,
1101
+ "eval_runtime": 0.577,
1102
+ "eval_samples_per_second": 79.728,
1103
+ "eval_steps_per_second": 3.466,
1104
+ "step": 552
1105
+ },
1106
+ {
1107
+ "epoch": 86.0,
1108
+ "eval_accuracy": 0.717391304347826,
1109
+ "eval_loss": 1.087559700012207,
1110
+ "eval_runtime": 0.5661,
1111
+ "eval_samples_per_second": 81.256,
1112
+ "eval_steps_per_second": 3.533,
1113
+ "step": 559
1114
+ },
1115
+ {
1116
+ "epoch": 86.15,
1117
+ "learning_rate": 2.222222222222222e-05,
1118
+ "loss": 0.0462,
1119
+ "step": 560
1120
+ },
1121
+ {
1122
+ "epoch": 86.92,
1123
+ "eval_accuracy": 0.6956521739130435,
1124
+ "eval_loss": 1.0740270614624023,
1125
+ "eval_runtime": 0.6474,
1126
+ "eval_samples_per_second": 71.049,
1127
+ "eval_steps_per_second": 3.089,
1128
+ "step": 565
1129
+ },
1130
+ {
1131
+ "epoch": 87.69,
1132
+ "learning_rate": 1.6666666666666664e-05,
1133
+ "loss": 0.0457,
1134
+ "step": 570
1135
+ },
1136
+ {
1137
+ "epoch": 88.0,
1138
+ "eval_accuracy": 0.6956521739130435,
1139
+ "eval_loss": 1.1073704957962036,
1140
+ "eval_runtime": 0.5747,
1141
+ "eval_samples_per_second": 80.036,
1142
+ "eval_steps_per_second": 3.48,
1143
+ "step": 572
1144
+ },
1145
+ {
1146
+ "epoch": 88.92,
1147
+ "eval_accuracy": 0.6956521739130435,
1148
+ "eval_loss": 1.077661395072937,
1149
+ "eval_runtime": 0.5678,
1150
+ "eval_samples_per_second": 81.01,
1151
+ "eval_steps_per_second": 3.522,
1152
+ "step": 578
1153
+ },
1154
+ {
1155
+ "epoch": 89.23,
1156
+ "learning_rate": 1.111111111111111e-05,
1157
+ "loss": 0.0482,
1158
+ "step": 580
1159
+ },
1160
+ {
1161
+ "epoch": 90.0,
1162
+ "eval_accuracy": 0.7391304347826086,
1163
+ "eval_loss": 1.0495113134384155,
1164
+ "eval_runtime": 0.5757,
1165
+ "eval_samples_per_second": 79.909,
1166
+ "eval_steps_per_second": 3.474,
1167
+ "step": 585
1168
+ },
1169
+ {
1170
+ "epoch": 90.77,
1171
+ "learning_rate": 5.555555555555555e-06,
1172
+ "loss": 0.0464,
1173
+ "step": 590
1174
+ },
1175
+ {
1176
+ "epoch": 90.92,
1177
+ "eval_accuracy": 0.717391304347826,
1178
+ "eval_loss": 1.039515495300293,
1179
+ "eval_runtime": 0.5732,
1180
+ "eval_samples_per_second": 80.246,
1181
+ "eval_steps_per_second": 3.489,
1182
+ "step": 591
1183
+ },
1184
+ {
1185
+ "epoch": 92.0,
1186
+ "eval_accuracy": 0.717391304347826,
1187
+ "eval_loss": 1.1445815563201904,
1188
+ "eval_runtime": 0.5643,
1189
+ "eval_samples_per_second": 81.522,
1190
+ "eval_steps_per_second": 3.544,
1191
+ "step": 598
1192
+ },
1193
+ {
1194
+ "epoch": 92.31,
1195
+ "learning_rate": 0.0,
1196
+ "loss": 0.0578,
1197
+ "step": 600
1198
+ },
1199
+ {
1200
+ "epoch": 92.31,
1201
+ "eval_accuracy": 0.6956521739130435,
1202
+ "eval_loss": 1.0596117973327637,
1203
+ "eval_runtime": 0.6055,
1204
+ "eval_samples_per_second": 75.965,
1205
+ "eval_steps_per_second": 3.303,
1206
+ "step": 600
1207
+ },
1208
+ {
1209
+ "epoch": 92.31,
1210
+ "step": 600,
1211
+ "total_flos": 2.072479896260444e+17,
1212
+ "train_loss": 0.2747100231051445,
1213
+ "train_runtime": 347.4081,
1214
+ "train_samples_per_second": 235.746,
1215
+ "train_steps_per_second": 1.727
1216
  }
1217
  ],
1218
  "logging_steps": 10,
1219
+ "max_steps": 600,
1220
  "num_input_tokens_seen": 0,
1221
+ "num_train_epochs": 100,
1222
  "save_steps": 500,
1223
+ "total_flos": 2.072479896260444e+17,
1224
  "train_batch_size": 32,
1225
  "trial_name": null,
1226
  "trial_params": null