napolitan commited on
Commit
ed612e1
1 Parent(s): cb2290b

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +1033 -0
trainer_state.json ADDED
@@ -0,0 +1,1033 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.03530377894639969,
3
+ "best_model_checkpoint": "my_awesome_ds_model/checkpoint-22473",
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 74910,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.07,
13
+ "learning_rate": 9.933253237217995e-05,
14
+ "loss": 0.2746,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.13,
19
+ "learning_rate": 9.86650647443599e-05,
20
+ "loss": 0.0897,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.2,
25
+ "learning_rate": 9.799759711653985e-05,
26
+ "loss": 0.0718,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.27,
31
+ "learning_rate": 9.73301294887198e-05,
32
+ "loss": 0.0616,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.33,
37
+ "learning_rate": 9.666266186089975e-05,
38
+ "loss": 0.0568,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.4,
43
+ "learning_rate": 9.59951942330797e-05,
44
+ "loss": 0.0525,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.47,
49
+ "learning_rate": 9.532772660525965e-05,
50
+ "loss": 0.0495,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.53,
55
+ "learning_rate": 9.46602589774396e-05,
56
+ "loss": 0.0488,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.6,
61
+ "learning_rate": 9.399279134961955e-05,
62
+ "loss": 0.0457,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.67,
67
+ "learning_rate": 9.33253237217995e-05,
68
+ "loss": 0.0422,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.73,
73
+ "learning_rate": 9.265785609397944e-05,
74
+ "loss": 0.0431,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.8,
79
+ "learning_rate": 9.199038846615939e-05,
80
+ "loss": 0.0467,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.87,
85
+ "learning_rate": 9.132292083833934e-05,
86
+ "loss": 0.0473,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.93,
91
+ "learning_rate": 9.065545321051929e-05,
92
+ "loss": 0.0458,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 1.0,
97
+ "eval_accuracy": 0.9874301209567,
98
+ "eval_f1": 0.9027943197434722,
99
+ "eval_loss": 0.04262871295213699,
100
+ "eval_precision": 0.9134223210975158,
101
+ "eval_recall": 0.8924107951458069,
102
+ "eval_runtime": 108.0836,
103
+ "eval_samples_per_second": 76.441,
104
+ "eval_steps_per_second": 9.557,
105
+ "step": 7491
106
+ },
107
+ {
108
+ "epoch": 1.0,
109
+ "learning_rate": 8.998798558269924e-05,
110
+ "loss": 0.0423,
111
+ "step": 7500
112
+ },
113
+ {
114
+ "epoch": 1.07,
115
+ "learning_rate": 8.932051795487919e-05,
116
+ "loss": 0.0369,
117
+ "step": 8000
118
+ },
119
+ {
120
+ "epoch": 1.13,
121
+ "learning_rate": 8.865305032705914e-05,
122
+ "loss": 0.0363,
123
+ "step": 8500
124
+ },
125
+ {
126
+ "epoch": 1.2,
127
+ "learning_rate": 8.798558269923909e-05,
128
+ "loss": 0.0333,
129
+ "step": 9000
130
+ },
131
+ {
132
+ "epoch": 1.27,
133
+ "learning_rate": 8.731811507141903e-05,
134
+ "loss": 0.0356,
135
+ "step": 9500
136
+ },
137
+ {
138
+ "epoch": 1.33,
139
+ "learning_rate": 8.665064744359898e-05,
140
+ "loss": 0.036,
141
+ "step": 10000
142
+ },
143
+ {
144
+ "epoch": 1.4,
145
+ "learning_rate": 8.598317981577893e-05,
146
+ "loss": 0.0372,
147
+ "step": 10500
148
+ },
149
+ {
150
+ "epoch": 1.47,
151
+ "learning_rate": 8.531571218795888e-05,
152
+ "loss": 0.0354,
153
+ "step": 11000
154
+ },
155
+ {
156
+ "epoch": 1.54,
157
+ "learning_rate": 8.464824456013883e-05,
158
+ "loss": 0.0356,
159
+ "step": 11500
160
+ },
161
+ {
162
+ "epoch": 1.6,
163
+ "learning_rate": 8.398077693231878e-05,
164
+ "loss": 0.0323,
165
+ "step": 12000
166
+ },
167
+ {
168
+ "epoch": 1.67,
169
+ "learning_rate": 8.331330930449873e-05,
170
+ "loss": 0.037,
171
+ "step": 12500
172
+ },
173
+ {
174
+ "epoch": 1.74,
175
+ "learning_rate": 8.264584167667868e-05,
176
+ "loss": 0.0361,
177
+ "step": 13000
178
+ },
179
+ {
180
+ "epoch": 1.8,
181
+ "learning_rate": 8.197837404885863e-05,
182
+ "loss": 0.0346,
183
+ "step": 13500
184
+ },
185
+ {
186
+ "epoch": 1.87,
187
+ "learning_rate": 8.131090642103857e-05,
188
+ "loss": 0.0351,
189
+ "step": 14000
190
+ },
191
+ {
192
+ "epoch": 1.94,
193
+ "learning_rate": 8.064343879321852e-05,
194
+ "loss": 0.0326,
195
+ "step": 14500
196
+ },
197
+ {
198
+ "epoch": 2.0,
199
+ "eval_accuracy": 0.9869225169145689,
200
+ "eval_f1": 0.8974685334464715,
201
+ "eval_loss": 0.03918388485908508,
202
+ "eval_precision": 0.935896175400649,
203
+ "eval_recall": 0.8620720883897844,
204
+ "eval_runtime": 107.1166,
205
+ "eval_samples_per_second": 77.131,
206
+ "eval_steps_per_second": 9.644,
207
+ "step": 14982
208
+ },
209
+ {
210
+ "epoch": 2.0,
211
+ "learning_rate": 7.997597116539847e-05,
212
+ "loss": 0.0342,
213
+ "step": 15000
214
+ },
215
+ {
216
+ "epoch": 2.07,
217
+ "learning_rate": 7.930850353757842e-05,
218
+ "loss": 0.0269,
219
+ "step": 15500
220
+ },
221
+ {
222
+ "epoch": 2.14,
223
+ "learning_rate": 7.864103590975837e-05,
224
+ "loss": 0.0273,
225
+ "step": 16000
226
+ },
227
+ {
228
+ "epoch": 2.2,
229
+ "learning_rate": 7.797356828193832e-05,
230
+ "loss": 0.0247,
231
+ "step": 16500
232
+ },
233
+ {
234
+ "epoch": 2.27,
235
+ "learning_rate": 7.730610065411827e-05,
236
+ "loss": 0.0262,
237
+ "step": 17000
238
+ },
239
+ {
240
+ "epoch": 2.34,
241
+ "learning_rate": 7.663863302629822e-05,
242
+ "loss": 0.0274,
243
+ "step": 17500
244
+ },
245
+ {
246
+ "epoch": 2.4,
247
+ "learning_rate": 7.597116539847817e-05,
248
+ "loss": 0.0262,
249
+ "step": 18000
250
+ },
251
+ {
252
+ "epoch": 2.47,
253
+ "learning_rate": 7.530369777065813e-05,
254
+ "loss": 0.0272,
255
+ "step": 18500
256
+ },
257
+ {
258
+ "epoch": 2.54,
259
+ "learning_rate": 7.463623014283808e-05,
260
+ "loss": 0.0278,
261
+ "step": 19000
262
+ },
263
+ {
264
+ "epoch": 2.6,
265
+ "learning_rate": 7.396876251501803e-05,
266
+ "loss": 0.0291,
267
+ "step": 19500
268
+ },
269
+ {
270
+ "epoch": 2.67,
271
+ "learning_rate": 7.330129488719797e-05,
272
+ "loss": 0.0265,
273
+ "step": 20000
274
+ },
275
+ {
276
+ "epoch": 2.74,
277
+ "learning_rate": 7.263382725937792e-05,
278
+ "loss": 0.0245,
279
+ "step": 20500
280
+ },
281
+ {
282
+ "epoch": 2.8,
283
+ "learning_rate": 7.196635963155787e-05,
284
+ "loss": 0.0271,
285
+ "step": 21000
286
+ },
287
+ {
288
+ "epoch": 2.87,
289
+ "learning_rate": 7.129889200373782e-05,
290
+ "loss": 0.0283,
291
+ "step": 21500
292
+ },
293
+ {
294
+ "epoch": 2.94,
295
+ "learning_rate": 7.063142437591777e-05,
296
+ "loss": 0.0277,
297
+ "step": 22000
298
+ },
299
+ {
300
+ "epoch": 3.0,
301
+ "eval_accuracy": 0.9892468091074852,
302
+ "eval_f1": 0.9156408399710355,
303
+ "eval_loss": 0.03530377894639969,
304
+ "eval_precision": 0.9151438393341776,
305
+ "eval_recall": 0.9161383807281289,
306
+ "eval_runtime": 107.1556,
307
+ "eval_samples_per_second": 77.103,
308
+ "eval_steps_per_second": 9.64,
309
+ "step": 22473
310
+ },
311
+ {
312
+ "epoch": 3.0,
313
+ "learning_rate": 6.996395674809772e-05,
314
+ "loss": 0.0232,
315
+ "step": 22500
316
+ },
317
+ {
318
+ "epoch": 3.07,
319
+ "learning_rate": 6.929648912027767e-05,
320
+ "loss": 0.0195,
321
+ "step": 23000
322
+ },
323
+ {
324
+ "epoch": 3.14,
325
+ "learning_rate": 6.862902149245762e-05,
326
+ "loss": 0.0198,
327
+ "step": 23500
328
+ },
329
+ {
330
+ "epoch": 3.2,
331
+ "learning_rate": 6.796155386463757e-05,
332
+ "loss": 0.0202,
333
+ "step": 24000
334
+ },
335
+ {
336
+ "epoch": 3.27,
337
+ "learning_rate": 6.729408623681752e-05,
338
+ "loss": 0.0181,
339
+ "step": 24500
340
+ },
341
+ {
342
+ "epoch": 3.34,
343
+ "learning_rate": 6.662661860899748e-05,
344
+ "loss": 0.0193,
345
+ "step": 25000
346
+ },
347
+ {
348
+ "epoch": 3.4,
349
+ "learning_rate": 6.595915098117743e-05,
350
+ "loss": 0.0198,
351
+ "step": 25500
352
+ },
353
+ {
354
+ "epoch": 3.47,
355
+ "learning_rate": 6.529168335335738e-05,
356
+ "loss": 0.0198,
357
+ "step": 26000
358
+ },
359
+ {
360
+ "epoch": 3.54,
361
+ "learning_rate": 6.462421572553732e-05,
362
+ "loss": 0.0197,
363
+ "step": 26500
364
+ },
365
+ {
366
+ "epoch": 3.6,
367
+ "learning_rate": 6.395674809771727e-05,
368
+ "loss": 0.0194,
369
+ "step": 27000
370
+ },
371
+ {
372
+ "epoch": 3.67,
373
+ "learning_rate": 6.328928046989722e-05,
374
+ "loss": 0.0203,
375
+ "step": 27500
376
+ },
377
+ {
378
+ "epoch": 3.74,
379
+ "learning_rate": 6.262181284207717e-05,
380
+ "loss": 0.0203,
381
+ "step": 28000
382
+ },
383
+ {
384
+ "epoch": 3.8,
385
+ "learning_rate": 6.195434521425712e-05,
386
+ "loss": 0.0211,
387
+ "step": 28500
388
+ },
389
+ {
390
+ "epoch": 3.87,
391
+ "learning_rate": 6.128687758643707e-05,
392
+ "loss": 0.0201,
393
+ "step": 29000
394
+ },
395
+ {
396
+ "epoch": 3.94,
397
+ "learning_rate": 6.061940995861701e-05,
398
+ "loss": 0.0209,
399
+ "step": 29500
400
+ },
401
+ {
402
+ "epoch": 4.0,
403
+ "eval_accuracy": 0.9885855880526038,
404
+ "eval_f1": 0.9106117466202469,
405
+ "eval_loss": 0.04323778301477432,
406
+ "eval_precision": 0.909255079006772,
407
+ "eval_recall": 0.9119724687556602,
408
+ "eval_runtime": 107.2535,
409
+ "eval_samples_per_second": 77.032,
410
+ "eval_steps_per_second": 9.631,
411
+ "step": 29964
412
+ },
413
+ {
414
+ "epoch": 4.0,
415
+ "learning_rate": 5.995194233079696e-05,
416
+ "loss": 0.0192,
417
+ "step": 30000
418
+ },
419
+ {
420
+ "epoch": 4.07,
421
+ "learning_rate": 5.928447470297691e-05,
422
+ "loss": 0.0115,
423
+ "step": 30500
424
+ },
425
+ {
426
+ "epoch": 4.14,
427
+ "learning_rate": 5.861700707515686e-05,
428
+ "loss": 0.0131,
429
+ "step": 31000
430
+ },
431
+ {
432
+ "epoch": 4.21,
433
+ "learning_rate": 5.794953944733681e-05,
434
+ "loss": 0.014,
435
+ "step": 31500
436
+ },
437
+ {
438
+ "epoch": 4.27,
439
+ "learning_rate": 5.7282071819516756e-05,
440
+ "loss": 0.0134,
441
+ "step": 32000
442
+ },
443
+ {
444
+ "epoch": 4.34,
445
+ "learning_rate": 5.6614604191696705e-05,
446
+ "loss": 0.0136,
447
+ "step": 32500
448
+ },
449
+ {
450
+ "epoch": 4.41,
451
+ "learning_rate": 5.5947136563876653e-05,
452
+ "loss": 0.0152,
453
+ "step": 33000
454
+ },
455
+ {
456
+ "epoch": 4.47,
457
+ "learning_rate": 5.52796689360566e-05,
458
+ "loss": 0.0135,
459
+ "step": 33500
460
+ },
461
+ {
462
+ "epoch": 4.54,
463
+ "learning_rate": 5.461220130823655e-05,
464
+ "loss": 0.0152,
465
+ "step": 34000
466
+ },
467
+ {
468
+ "epoch": 4.61,
469
+ "learning_rate": 5.394473368041651e-05,
470
+ "loss": 0.0138,
471
+ "step": 34500
472
+ },
473
+ {
474
+ "epoch": 4.67,
475
+ "learning_rate": 5.3277266052596456e-05,
476
+ "loss": 0.0154,
477
+ "step": 35000
478
+ },
479
+ {
480
+ "epoch": 4.74,
481
+ "learning_rate": 5.2609798424776405e-05,
482
+ "loss": 0.0139,
483
+ "step": 35500
484
+ },
485
+ {
486
+ "epoch": 4.81,
487
+ "learning_rate": 5.1942330796956354e-05,
488
+ "loss": 0.015,
489
+ "step": 36000
490
+ },
491
+ {
492
+ "epoch": 4.87,
493
+ "learning_rate": 5.12748631691363e-05,
494
+ "loss": 0.0131,
495
+ "step": 36500
496
+ },
497
+ {
498
+ "epoch": 4.94,
499
+ "learning_rate": 5.060739554131625e-05,
500
+ "loss": 0.0141,
501
+ "step": 37000
502
+ },
503
+ {
504
+ "epoch": 5.0,
505
+ "eval_accuracy": 0.9885588720503864,
506
+ "eval_f1": 0.9082037484609421,
507
+ "eval_loss": 0.049268897622823715,
508
+ "eval_precision": 0.9146688711307064,
509
+ "eval_recall": 0.9018293787357363,
510
+ "eval_runtime": 107.4361,
511
+ "eval_samples_per_second": 76.902,
512
+ "eval_steps_per_second": 9.615,
513
+ "step": 37455
514
+ },
515
+ {
516
+ "epoch": 5.01,
517
+ "learning_rate": 4.99399279134962e-05,
518
+ "loss": 0.0126,
519
+ "step": 37500
520
+ },
521
+ {
522
+ "epoch": 5.07,
523
+ "learning_rate": 4.927246028567615e-05,
524
+ "loss": 0.0073,
525
+ "step": 38000
526
+ },
527
+ {
528
+ "epoch": 5.14,
529
+ "learning_rate": 4.86049926578561e-05,
530
+ "loss": 0.0076,
531
+ "step": 38500
532
+ },
533
+ {
534
+ "epoch": 5.21,
535
+ "learning_rate": 4.793752503003605e-05,
536
+ "loss": 0.0076,
537
+ "step": 39000
538
+ },
539
+ {
540
+ "epoch": 5.27,
541
+ "learning_rate": 4.7270057402215996e-05,
542
+ "loss": 0.0086,
543
+ "step": 39500
544
+ },
545
+ {
546
+ "epoch": 5.34,
547
+ "learning_rate": 4.6602589774395945e-05,
548
+ "loss": 0.0092,
549
+ "step": 40000
550
+ },
551
+ {
552
+ "epoch": 5.41,
553
+ "learning_rate": 4.5935122146575894e-05,
554
+ "loss": 0.0087,
555
+ "step": 40500
556
+ },
557
+ {
558
+ "epoch": 5.47,
559
+ "learning_rate": 4.526765451875584e-05,
560
+ "loss": 0.0085,
561
+ "step": 41000
562
+ },
563
+ {
564
+ "epoch": 5.54,
565
+ "learning_rate": 4.460018689093579e-05,
566
+ "loss": 0.0086,
567
+ "step": 41500
568
+ },
569
+ {
570
+ "epoch": 5.61,
571
+ "learning_rate": 4.393271926311574e-05,
572
+ "loss": 0.0084,
573
+ "step": 42000
574
+ },
575
+ {
576
+ "epoch": 5.67,
577
+ "learning_rate": 4.326525163529569e-05,
578
+ "loss": 0.0091,
579
+ "step": 42500
580
+ },
581
+ {
582
+ "epoch": 5.74,
583
+ "learning_rate": 4.259778400747564e-05,
584
+ "loss": 0.0083,
585
+ "step": 43000
586
+ },
587
+ {
588
+ "epoch": 5.81,
589
+ "learning_rate": 4.193031637965559e-05,
590
+ "loss": 0.0088,
591
+ "step": 43500
592
+ },
593
+ {
594
+ "epoch": 5.87,
595
+ "learning_rate": 4.1262848751835536e-05,
596
+ "loss": 0.0088,
597
+ "step": 44000
598
+ },
599
+ {
600
+ "epoch": 5.94,
601
+ "learning_rate": 4.0595381124015485e-05,
602
+ "loss": 0.0081,
603
+ "step": 44500
604
+ },
605
+ {
606
+ "epoch": 6.0,
607
+ "eval_accuracy": 0.9886724150598104,
608
+ "eval_f1": 0.9117419940949353,
609
+ "eval_loss": 0.058003462851047516,
610
+ "eval_precision": 0.9146085847079194,
611
+ "eval_recall": 0.9088933164281833,
612
+ "eval_runtime": 107.2097,
613
+ "eval_samples_per_second": 77.064,
614
+ "eval_steps_per_second": 9.635,
615
+ "step": 44946
616
+ },
617
+ {
618
+ "epoch": 6.01,
619
+ "learning_rate": 3.9927913496195434e-05,
620
+ "loss": 0.0091,
621
+ "step": 45000
622
+ },
623
+ {
624
+ "epoch": 6.07,
625
+ "learning_rate": 3.926044586837538e-05,
626
+ "loss": 0.004,
627
+ "step": 45500
628
+ },
629
+ {
630
+ "epoch": 6.14,
631
+ "learning_rate": 3.859297824055533e-05,
632
+ "loss": 0.0044,
633
+ "step": 46000
634
+ },
635
+ {
636
+ "epoch": 6.21,
637
+ "learning_rate": 3.792551061273528e-05,
638
+ "loss": 0.0042,
639
+ "step": 46500
640
+ },
641
+ {
642
+ "epoch": 6.27,
643
+ "learning_rate": 3.725804298491523e-05,
644
+ "loss": 0.0042,
645
+ "step": 47000
646
+ },
647
+ {
648
+ "epoch": 6.34,
649
+ "learning_rate": 3.659057535709518e-05,
650
+ "loss": 0.005,
651
+ "step": 47500
652
+ },
653
+ {
654
+ "epoch": 6.41,
655
+ "learning_rate": 3.592310772927513e-05,
656
+ "loss": 0.0036,
657
+ "step": 48000
658
+ },
659
+ {
660
+ "epoch": 6.47,
661
+ "learning_rate": 3.525564010145508e-05,
662
+ "loss": 0.0052,
663
+ "step": 48500
664
+ },
665
+ {
666
+ "epoch": 6.54,
667
+ "learning_rate": 3.4588172473635026e-05,
668
+ "loss": 0.0046,
669
+ "step": 49000
670
+ },
671
+ {
672
+ "epoch": 6.61,
673
+ "learning_rate": 3.392070484581498e-05,
674
+ "loss": 0.0035,
675
+ "step": 49500
676
+ },
677
+ {
678
+ "epoch": 6.67,
679
+ "learning_rate": 3.325323721799493e-05,
680
+ "loss": 0.0051,
681
+ "step": 50000
682
+ },
683
+ {
684
+ "epoch": 6.74,
685
+ "learning_rate": 3.258576959017488e-05,
686
+ "loss": 0.0051,
687
+ "step": 50500
688
+ },
689
+ {
690
+ "epoch": 6.81,
691
+ "learning_rate": 3.191830196235483e-05,
692
+ "loss": 0.0049,
693
+ "step": 51000
694
+ },
695
+ {
696
+ "epoch": 6.87,
697
+ "learning_rate": 3.125083433453478e-05,
698
+ "loss": 0.0051,
699
+ "step": 51500
700
+ },
701
+ {
702
+ "epoch": 6.94,
703
+ "learning_rate": 3.0583366706714726e-05,
704
+ "loss": 0.0048,
705
+ "step": 52000
706
+ },
707
+ {
708
+ "epoch": 7.0,
709
+ "eval_accuracy": 0.9884319710398536,
710
+ "eval_f1": 0.9093145069460259,
711
+ "eval_loss": 0.07726183533668518,
712
+ "eval_precision": 0.9146889031430404,
713
+ "eval_recall": 0.90400289802572,
714
+ "eval_runtime": 107.8277,
715
+ "eval_samples_per_second": 76.622,
716
+ "eval_steps_per_second": 9.58,
717
+ "step": 52437
718
+ },
719
+ {
720
+ "epoch": 7.01,
721
+ "learning_rate": 2.9915899078894678e-05,
722
+ "loss": 0.0045,
723
+ "step": 52500
724
+ },
725
+ {
726
+ "epoch": 7.08,
727
+ "learning_rate": 2.9248431451074627e-05,
728
+ "loss": 0.0021,
729
+ "step": 53000
730
+ },
731
+ {
732
+ "epoch": 7.14,
733
+ "learning_rate": 2.8580963823254576e-05,
734
+ "loss": 0.0022,
735
+ "step": 53500
736
+ },
737
+ {
738
+ "epoch": 7.21,
739
+ "learning_rate": 2.7913496195434525e-05,
740
+ "loss": 0.0024,
741
+ "step": 54000
742
+ },
743
+ {
744
+ "epoch": 7.28,
745
+ "learning_rate": 2.7246028567614474e-05,
746
+ "loss": 0.0025,
747
+ "step": 54500
748
+ },
749
+ {
750
+ "epoch": 7.34,
751
+ "learning_rate": 2.6578560939794423e-05,
752
+ "loss": 0.0018,
753
+ "step": 55000
754
+ },
755
+ {
756
+ "epoch": 7.41,
757
+ "learning_rate": 2.591109331197437e-05,
758
+ "loss": 0.0025,
759
+ "step": 55500
760
+ },
761
+ {
762
+ "epoch": 7.48,
763
+ "learning_rate": 2.524362568415432e-05,
764
+ "loss": 0.0027,
765
+ "step": 56000
766
+ },
767
+ {
768
+ "epoch": 7.54,
769
+ "learning_rate": 2.457615805633427e-05,
770
+ "loss": 0.0024,
771
+ "step": 56500
772
+ },
773
+ {
774
+ "epoch": 7.61,
775
+ "learning_rate": 2.390869042851422e-05,
776
+ "loss": 0.0021,
777
+ "step": 57000
778
+ },
779
+ {
780
+ "epoch": 7.68,
781
+ "learning_rate": 2.3241222800694167e-05,
782
+ "loss": 0.0022,
783
+ "step": 57500
784
+ },
785
+ {
786
+ "epoch": 7.74,
787
+ "learning_rate": 2.2573755172874116e-05,
788
+ "loss": 0.0025,
789
+ "step": 58000
790
+ },
791
+ {
792
+ "epoch": 7.81,
793
+ "learning_rate": 2.1906287545054065e-05,
794
+ "loss": 0.0031,
795
+ "step": 58500
796
+ },
797
+ {
798
+ "epoch": 7.88,
799
+ "learning_rate": 2.1238819917234014e-05,
800
+ "loss": 0.0022,
801
+ "step": 59000
802
+ },
803
+ {
804
+ "epoch": 7.94,
805
+ "learning_rate": 2.0571352289413963e-05,
806
+ "loss": 0.0023,
807
+ "step": 59500
808
+ },
809
+ {
810
+ "epoch": 8.0,
811
+ "eval_accuracy": 0.9890330810897457,
812
+ "eval_f1": 0.9139605257215121,
813
+ "eval_loss": 0.0873148962855339,
814
+ "eval_precision": 0.9116136588881881,
815
+ "eval_recall": 0.9163195073356276,
816
+ "eval_runtime": 107.6976,
817
+ "eval_samples_per_second": 76.715,
818
+ "eval_steps_per_second": 9.592,
819
+ "step": 59928
820
+ },
821
+ {
822
+ "epoch": 8.01,
823
+ "learning_rate": 1.9903884661593912e-05,
824
+ "loss": 0.0015,
825
+ "step": 60000
826
+ },
827
+ {
828
+ "epoch": 8.08,
829
+ "learning_rate": 1.9236417033773864e-05,
830
+ "loss": 0.001,
831
+ "step": 60500
832
+ },
833
+ {
834
+ "epoch": 8.14,
835
+ "learning_rate": 1.8568949405953813e-05,
836
+ "loss": 0.0008,
837
+ "step": 61000
838
+ },
839
+ {
840
+ "epoch": 8.21,
841
+ "learning_rate": 1.7901481778133762e-05,
842
+ "loss": 0.0011,
843
+ "step": 61500
844
+ },
845
+ {
846
+ "epoch": 8.28,
847
+ "learning_rate": 1.723401415031371e-05,
848
+ "loss": 0.0014,
849
+ "step": 62000
850
+ },
851
+ {
852
+ "epoch": 8.34,
853
+ "learning_rate": 1.656654652249366e-05,
854
+ "loss": 0.0012,
855
+ "step": 62500
856
+ },
857
+ {
858
+ "epoch": 8.41,
859
+ "learning_rate": 1.589907889467361e-05,
860
+ "loss": 0.0012,
861
+ "step": 63000
862
+ },
863
+ {
864
+ "epoch": 8.48,
865
+ "learning_rate": 1.5231611266853558e-05,
866
+ "loss": 0.0008,
867
+ "step": 63500
868
+ },
869
+ {
870
+ "epoch": 8.54,
871
+ "learning_rate": 1.4564143639033507e-05,
872
+ "loss": 0.0013,
873
+ "step": 64000
874
+ },
875
+ {
876
+ "epoch": 8.61,
877
+ "learning_rate": 1.3896676011213459e-05,
878
+ "loss": 0.0012,
879
+ "step": 64500
880
+ },
881
+ {
882
+ "epoch": 8.68,
883
+ "learning_rate": 1.3229208383393408e-05,
884
+ "loss": 0.0008,
885
+ "step": 65000
886
+ },
887
+ {
888
+ "epoch": 8.74,
889
+ "learning_rate": 1.2561740755573357e-05,
890
+ "loss": 0.0011,
891
+ "step": 65500
892
+ },
893
+ {
894
+ "epoch": 8.81,
895
+ "learning_rate": 1.1894273127753304e-05,
896
+ "loss": 0.0013,
897
+ "step": 66000
898
+ },
899
+ {
900
+ "epoch": 8.88,
901
+ "learning_rate": 1.1226805499933253e-05,
902
+ "loss": 0.0012,
903
+ "step": 66500
904
+ },
905
+ {
906
+ "epoch": 8.94,
907
+ "learning_rate": 1.0559337872113203e-05,
908
+ "loss": 0.0013,
909
+ "step": 67000
910
+ },
911
+ {
912
+ "epoch": 9.0,
913
+ "eval_accuracy": 0.9886724150598104,
914
+ "eval_f1": 0.9117727148951621,
915
+ "eval_loss": 0.09737657755613327,
916
+ "eval_precision": 0.9138464337700145,
917
+ "eval_recall": 0.9097083861619272,
918
+ "eval_runtime": 107.7722,
919
+ "eval_samples_per_second": 76.662,
920
+ "eval_steps_per_second": 9.585,
921
+ "step": 67419
922
+ },
923
+ {
924
+ "epoch": 9.01,
925
+ "learning_rate": 9.891870244293152e-06,
926
+ "loss": 0.0011,
927
+ "step": 67500
928
+ },
929
+ {
930
+ "epoch": 9.08,
931
+ "learning_rate": 9.224402616473101e-06,
932
+ "loss": 0.0003,
933
+ "step": 68000
934
+ },
935
+ {
936
+ "epoch": 9.14,
937
+ "learning_rate": 8.55693498865305e-06,
938
+ "loss": 0.0005,
939
+ "step": 68500
940
+ },
941
+ {
942
+ "epoch": 9.21,
943
+ "learning_rate": 7.889467360833e-06,
944
+ "loss": 0.0003,
945
+ "step": 69000
946
+ },
947
+ {
948
+ "epoch": 9.28,
949
+ "learning_rate": 7.221999733012949e-06,
950
+ "loss": 0.0006,
951
+ "step": 69500
952
+ },
953
+ {
954
+ "epoch": 9.34,
955
+ "learning_rate": 6.554532105192898e-06,
956
+ "loss": 0.0007,
957
+ "step": 70000
958
+ },
959
+ {
960
+ "epoch": 9.41,
961
+ "learning_rate": 5.887064477372848e-06,
962
+ "loss": 0.0005,
963
+ "step": 70500
964
+ },
965
+ {
966
+ "epoch": 9.48,
967
+ "learning_rate": 5.219596849552797e-06,
968
+ "loss": 0.0005,
969
+ "step": 71000
970
+ },
971
+ {
972
+ "epoch": 9.54,
973
+ "learning_rate": 4.552129221732746e-06,
974
+ "loss": 0.0007,
975
+ "step": 71500
976
+ },
977
+ {
978
+ "epoch": 9.61,
979
+ "learning_rate": 3.884661593912695e-06,
980
+ "loss": 0.0007,
981
+ "step": 72000
982
+ },
983
+ {
984
+ "epoch": 9.68,
985
+ "learning_rate": 3.2171939660926445e-06,
986
+ "loss": 0.0006,
987
+ "step": 72500
988
+ },
989
+ {
990
+ "epoch": 9.75,
991
+ "learning_rate": 2.549726338272594e-06,
992
+ "loss": 0.0006,
993
+ "step": 73000
994
+ },
995
+ {
996
+ "epoch": 9.81,
997
+ "learning_rate": 1.8822587104525433e-06,
998
+ "loss": 0.0006,
999
+ "step": 73500
1000
+ },
1001
+ {
1002
+ "epoch": 9.88,
1003
+ "learning_rate": 1.2147910826324925e-06,
1004
+ "loss": 0.0002,
1005
+ "step": 74000
1006
+ },
1007
+ {
1008
+ "epoch": 9.95,
1009
+ "learning_rate": 5.473234548124416e-07,
1010
+ "loss": 0.0005,
1011
+ "step": 74500
1012
+ },
1013
+ {
1014
+ "epoch": 10.0,
1015
+ "eval_accuracy": 0.9888661060758868,
1016
+ "eval_f1": 0.9123397290170047,
1017
+ "eval_loss": 0.10618099570274353,
1018
+ "eval_precision": 0.9160883856829802,
1019
+ "eval_recall": 0.9086216265169353,
1020
+ "eval_runtime": 107.8685,
1021
+ "eval_samples_per_second": 76.593,
1022
+ "eval_steps_per_second": 9.576,
1023
+ "step": 74910
1024
+ }
1025
+ ],
1026
+ "logging_steps": 500,
1027
+ "max_steps": 74910,
1028
+ "num_train_epochs": 10,
1029
+ "save_steps": 500,
1030
+ "total_flos": 1.4917264852008653e+18,
1031
+ "trial_name": null,
1032
+ "trial_params": null
1033
+ }