Theoreticallyhugo commited on
Commit
e1080b1
1 Parent(s): 6e2aae3

Training in progress, epoch 40, checkpoint

Browse files
checkpoint-3240/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:88958053a1296e0a83469e9628b4feac2ea27e717c1255179eb5717e3012a1a0
3
  size 592318676
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d904a40e8bf2435609ca0452c3e99c5835ec68eb87254e43cc838f8334784402
3
  size 592318676
checkpoint-3240/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed48b09797e44da41b3eb98e96180641cc735aa5b260b805fa3ac8ff0f5deb06
3
  size 1014645498
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df5dcde98dba2b05409426a76c56bb8c9d465ef0f34115afbf6be64255512232
3
  size 1014645498
checkpoint-3240/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ff8bca015f0e06a2bc98277c349f4bacc177ea700f7f38b06cd961ef69354bab
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8c5c99c3bad1ce3feb0cbd346df74269719c44b16be7c0f6afdd1817489144f
3
  size 14244
checkpoint-3240/trainer_state.json CHANGED
@@ -11,1602 +11,1602 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_B": {
14
- "f1-score": 0.7905866302864938,
15
- "precision": 0.7768096514745308,
16
- "recall": 0.8048611111111111,
17
- "support": 1440.0
18
  },
19
  "eval_I": {
20
- "f1-score": 0.9005331227169513,
21
- "precision": 0.9637593111099372,
22
- "recall": 0.8450919534905267,
23
- "support": 21587.0
24
  },
25
  "eval_O": {
26
- "f1-score": 0.8288892663043479,
27
- "precision": 0.7463108800367001,
28
- "recall": 0.9320156593144275,
29
- "support": 10473.0
30
  },
31
- "eval_accuracy": 0.8705373134328358,
32
- "eval_loss": 0.32930415868759155,
33
  "eval_macro avg": {
34
- "f1-score": 0.8400030064359308,
35
- "precision": 0.8289599475403894,
36
- "recall": 0.8606562413053552,
37
- "support": 33500.0
38
- },
39
- "eval_runtime": 4.7349,
40
- "eval_samples_per_second": 17.107,
41
- "eval_steps_per_second": 2.323,
42
  "eval_weighted avg": {
43
- "f1-score": 0.8734092702599646,
44
- "precision": 0.8877430445874,
45
- "recall": 0.8705373134328358,
46
- "support": 33500.0
47
  },
48
  "step": 81
49
  },
50
  {
51
  "epoch": 2.0,
52
  "eval_B": {
53
- "f1-score": 0.8656218739579861,
54
- "precision": 0.8325849903784477,
55
- "recall": 0.9013888888888889,
56
- "support": 1440.0
57
  },
58
  "eval_I": {
59
- "f1-score": 0.9547168948727912,
60
- "precision": 0.9423736462093862,
61
- "recall": 0.9673877796822161,
62
- "support": 21587.0
63
  },
64
  "eval_O": {
65
- "f1-score": 0.9013528191962081,
66
- "precision": 0.9332379102341274,
67
- "recall": 0.8715745249689678,
68
- "support": 10473.0
69
  },
70
- "eval_accuracy": 0.9345970149253732,
71
- "eval_loss": 0.2162298709154129,
72
  "eval_macro avg": {
73
- "f1-score": 0.9072305293423284,
74
- "precision": 0.902732182273987,
75
- "recall": 0.9134503978466909,
76
- "support": 33500.0
77
- },
78
- "eval_runtime": 4.753,
79
- "eval_samples_per_second": 17.042,
80
- "eval_steps_per_second": 2.314,
81
  "eval_weighted avg": {
82
- "f1-score": 0.9342040950316517,
83
- "precision": 0.9347982961417612,
84
- "recall": 0.9345970149253732,
85
- "support": 33500.0
86
  },
87
  "step": 162
88
  },
89
  {
90
  "epoch": 3.0,
91
  "eval_B": {
92
- "f1-score": 0.8603587759409076,
93
- "precision": 0.8717034925160371,
94
- "recall": 0.8493055555555555,
95
- "support": 1440.0
96
  },
97
  "eval_I": {
98
- "f1-score": 0.9434195218082959,
99
- "precision": 0.9677057963955188,
100
- "recall": 0.9203224162690509,
101
- "support": 21587.0
102
  },
103
  "eval_O": {
104
- "f1-score": 0.8879310344827586,
105
- "precision": 0.8459410391631365,
106
- "recall": 0.9343072663038289,
107
- "support": 10473.0
108
  },
109
- "eval_accuracy": 0.9216417910447762,
110
- "eval_loss": 0.24082203209400177,
111
  "eval_macro avg": {
112
- "f1-score": 0.8972364440773206,
113
- "precision": 0.8951167760248975,
114
- "recall": 0.9013117460428117,
115
- "support": 33500.0
116
- },
117
- "eval_runtime": 4.7554,
118
- "eval_samples_per_second": 17.033,
119
- "eval_steps_per_second": 2.313,
120
  "eval_weighted avg": {
121
- "f1-score": 0.9225019575751797,
122
- "precision": 0.9255121957960801,
123
- "recall": 0.9216417910447762,
124
- "support": 33500.0
125
  },
126
  "step": 243
127
  },
128
  {
129
  "epoch": 4.0,
130
  "eval_B": {
131
- "f1-score": 0.8804020100502512,
132
- "precision": 0.8504854368932039,
133
- "recall": 0.9125,
134
- "support": 1440.0
135
  },
136
  "eval_I": {
137
- "f1-score": 0.9589184992350025,
138
- "precision": 0.945595388218339,
139
- "recall": 0.9726224116366332,
140
- "support": 21587.0
141
  },
142
  "eval_O": {
143
- "f1-score": 0.9091178797468354,
144
- "precision": 0.9427750999897446,
145
- "recall": 0.877780960565263,
146
- "support": 10473.0
147
  },
148
- "eval_accuracy": 0.9403880597014925,
149
- "eval_loss": 0.22083505988121033,
150
  "eval_macro avg": {
151
- "f1-score": 0.9161461296773631,
152
- "precision": 0.9129519750337626,
153
- "recall": 0.9209677907339654,
154
- "support": 33500.0
155
- },
156
- "eval_runtime": 4.7449,
157
- "eval_samples_per_second": 17.071,
158
- "eval_steps_per_second": 2.318,
159
  "eval_weighted avg": {
160
- "f1-score": 0.9399744505088647,
161
- "precision": 0.9406253819936745,
162
- "recall": 0.9403880597014925,
163
- "support": 33500.0
164
  },
165
  "step": 324
166
  },
167
  {
168
  "epoch": 5.0,
169
  "eval_B": {
170
- "f1-score": 0.8879781420765027,
171
- "precision": 0.8736559139784946,
172
- "recall": 0.9027777777777778,
173
- "support": 1440.0
174
  },
175
  "eval_I": {
176
- "f1-score": 0.9613201259451634,
177
- "precision": 0.9539317642765919,
178
- "recall": 0.9688238291564367,
179
- "support": 21587.0
180
  },
181
  "eval_O": {
182
- "f1-score": 0.916006030835076,
183
- "precision": 0.9334853291038858,
184
- "recall": 0.899169292466342,
185
- "support": 10473.0
186
  },
187
- "eval_accuracy": 0.9442089552238806,
188
- "eval_loss": 0.2257838249206543,
189
  "eval_macro avg": {
190
- "f1-score": 0.921768099618914,
191
- "precision": 0.9203576691196576,
192
- "recall": 0.9235902998001855,
193
- "support": 33500.0
194
- },
195
- "eval_runtime": 4.7845,
196
- "eval_samples_per_second": 16.93,
197
- "eval_steps_per_second": 2.299,
198
  "eval_weighted avg": {
199
- "f1-score": 0.9440011117702735,
200
- "precision": 0.9440889959185916,
201
- "recall": 0.9442089552238806,
202
- "support": 33500.0
203
  },
204
  "step": 405
205
  },
206
  {
207
  "epoch": 6.0,
208
  "eval_B": {
209
- "f1-score": 0.8873579056148813,
210
- "precision": 0.8803827751196173,
211
- "recall": 0.8944444444444445,
212
- "support": 1440.0
213
  },
214
  "eval_I": {
215
- "f1-score": 0.9622060984570168,
216
- "precision": 0.9539267015706806,
217
- "recall": 0.9706304720433594,
218
- "support": 21587.0
219
  },
220
  "eval_O": {
221
- "f1-score": 0.9182769530299342,
222
- "precision": 0.9365567911040509,
223
- "recall": 0.9006970304592762,
224
- "support": 10473.0
225
  },
226
- "eval_accuracy": 0.9454925373134329,
227
- "eval_loss": 0.2242102175951004,
228
  "eval_macro avg": {
229
- "f1-score": 0.9226136523672773,
230
- "precision": 0.923622089264783,
231
- "recall": 0.9219239823156933,
232
- "support": 33500.0
233
- },
234
- "eval_runtime": 4.7715,
235
- "eval_samples_per_second": 16.976,
236
- "eval_steps_per_second": 2.305,
237
  "eval_weighted avg": {
238
- "f1-score": 0.945255312255509,
239
- "precision": 0.9453351097376493,
240
- "recall": 0.9454925373134329,
241
- "support": 33500.0
242
  },
243
  "step": 486
244
  },
245
  {
246
  "epoch": 6.17,
247
- "grad_norm": 1.8705081939697266,
248
  "learning_rate": 1.7530864197530865e-05,
249
- "loss": 0.1601,
250
  "step": 500
251
  },
252
  {
253
  "epoch": 7.0,
254
  "eval_B": {
255
- "f1-score": 0.8876749743939911,
256
- "precision": 0.8730691739422431,
257
- "recall": 0.9027777777777778,
258
- "support": 1440.0
259
  },
260
  "eval_I": {
261
- "f1-score": 0.9611415113943341,
262
- "precision": 0.9581108451482232,
263
- "recall": 0.9641914114976606,
264
- "support": 21587.0
265
  },
266
  "eval_O": {
267
- "f1-score": 0.9164739884393064,
268
- "precision": 0.9247594050743657,
269
- "recall": 0.9083357204239473,
270
- "support": 10473.0
271
  },
272
- "eval_accuracy": 0.944089552238806,
273
- "eval_loss": 0.22517083585262299,
274
  "eval_macro avg": {
275
- "f1-score": 0.9217634914092105,
276
- "precision": 0.9186464747216107,
277
- "recall": 0.9251016365664619,
278
- "support": 33500.0
279
- },
280
- "eval_runtime": 4.7926,
281
- "eval_samples_per_second": 16.901,
282
- "eval_steps_per_second": 2.295,
283
  "eval_weighted avg": {
284
- "f1-score": 0.9440192791200506,
285
- "precision": 0.9440287663891153,
286
- "recall": 0.944089552238806,
287
- "support": 33500.0
288
  },
289
  "step": 567
290
  },
291
  {
292
  "epoch": 8.0,
293
  "eval_B": {
294
- "f1-score": 0.894540522210919,
295
- "precision": 0.8740888005301524,
296
- "recall": 0.9159722222222222,
297
- "support": 1440.0
298
  },
299
  "eval_I": {
300
- "f1-score": 0.9638018385639943,
301
- "precision": 0.9540255968049378,
302
- "recall": 0.9737805160513272,
303
- "support": 21587.0
304
  },
305
  "eval_O": {
306
- "f1-score": 0.9203132648066569,
307
- "precision": 0.9441598875163202,
308
- "recall": 0.8976415544734078,
309
- "support": 10473.0
310
  },
311
- "eval_accuracy": 0.9474925373134329,
312
- "eval_loss": 0.24869142472743988,
313
  "eval_macro avg": {
314
- "f1-score": 0.9262185418605234,
315
- "precision": 0.9240914282838034,
316
- "recall": 0.9291314309156524,
317
- "support": 33500.0
318
- },
319
- "eval_runtime": 4.7743,
320
- "eval_samples_per_second": 16.966,
321
- "eval_steps_per_second": 2.304,
322
  "eval_weighted avg": {
323
- "f1-score": 0.947228939205516,
324
- "precision": 0.9475052218791054,
325
- "recall": 0.9474925373134329,
326
- "support": 33500.0
327
  },
328
  "step": 648
329
  },
330
  {
331
  "epoch": 9.0,
332
  "eval_B": {
333
- "f1-score": 0.8851236868858015,
334
- "precision": 0.8643282594308405,
335
- "recall": 0.9069444444444444,
336
- "support": 1440.0
337
  },
338
  "eval_I": {
339
- "f1-score": 0.9617001723147618,
340
- "precision": 0.9540067462849849,
341
- "recall": 0.9695186918052532,
342
- "support": 21587.0
343
  },
344
  "eval_O": {
345
- "f1-score": 0.9165854609237966,
346
- "precision": 0.9358272808675754,
347
- "recall": 0.8981189725961998,
348
- "support": 10473.0
349
  },
350
- "eval_accuracy": 0.9445074626865672,
351
- "eval_loss": 0.3045330345630646,
352
  "eval_macro avg": {
353
- "f1-score": 0.9211364400414532,
354
- "precision": 0.9180540955278002,
355
- "recall": 0.9248607029486324,
356
- "support": 33500.0
357
- },
358
- "eval_runtime": 4.7599,
359
- "eval_samples_per_second": 17.017,
360
- "eval_steps_per_second": 2.311,
361
  "eval_weighted avg": {
362
- "f1-score": 0.9443044555560967,
363
- "precision": 0.9444685205421044,
364
- "recall": 0.9445074626865672,
365
- "support": 33500.0
366
  },
367
  "step": 729
368
  },
369
  {
370
  "epoch": 10.0,
371
  "eval_B": {
372
- "f1-score": 0.8815068493150685,
373
- "precision": 0.8695945945945946,
374
- "recall": 0.89375,
375
- "support": 1440.0
376
  },
377
  "eval_I": {
378
- "f1-score": 0.956297959947595,
379
- "precision": 0.9660159757999717,
380
- "recall": 0.9467735211006625,
381
- "support": 21587.0
382
  },
383
  "eval_O": {
384
- "f1-score": 0.9101049868766404,
385
- "precision": 0.8937678357728068,
386
- "recall": 0.9270505108373914,
387
- "support": 10473.0
388
  },
389
- "eval_accuracy": 0.9383283582089552,
390
- "eval_loss": 0.32627996802330017,
391
  "eval_macro avg": {
392
- "f1-score": 0.9159699320464346,
393
- "precision": 0.909792802055791,
394
- "recall": 0.9225246773126846,
395
- "support": 33500.0
396
- },
397
- "eval_runtime": 4.7534,
398
- "eval_samples_per_second": 17.04,
399
- "eval_steps_per_second": 2.314,
400
  "eval_weighted avg": {
401
- "f1-score": 0.9386418940884026,
402
- "precision": 0.9392845859659347,
403
- "recall": 0.9383283582089552,
404
- "support": 33500.0
405
  },
406
  "step": 810
407
  },
408
  {
409
  "epoch": 11.0,
410
  "eval_B": {
411
- "f1-score": 0.8901136755080951,
412
- "precision": 0.8831168831168831,
413
- "recall": 0.8972222222222223,
414
- "support": 1440.0
415
  },
416
  "eval_I": {
417
- "f1-score": 0.9620073001078947,
418
- "precision": 0.9535359970874671,
419
- "recall": 0.9706304720433594,
420
- "support": 21587.0
421
  },
422
  "eval_O": {
423
- "f1-score": 0.917413322945072,
424
- "precision": 0.9361025539103647,
425
- "recall": 0.8994557433400172,
426
- "support": 10473.0
427
  },
428
- "eval_accuracy": 0.9452238805970149,
429
- "eval_loss": 0.3123624622821808,
430
  "eval_macro avg": {
431
- "f1-score": 0.9231780995203539,
432
- "precision": 0.9242518113715716,
433
- "recall": 0.922436145868533,
434
- "support": 33500.0
435
- },
436
- "eval_runtime": 4.7539,
437
- "eval_samples_per_second": 17.039,
438
- "eval_steps_per_second": 2.314,
439
  "eval_weighted avg": {
440
- "f1-score": 0.9449756719810304,
441
- "precision": 0.9450588635199616,
442
- "recall": 0.9452238805970149,
443
- "support": 33500.0
444
  },
445
  "step": 891
446
  },
447
  {
448
  "epoch": 12.0,
449
  "eval_B": {
450
- "f1-score": 0.8800819952169457,
451
- "precision": 0.8661735036987223,
452
- "recall": 0.8944444444444445,
453
- "support": 1440.0
454
  },
455
  "eval_I": {
456
- "f1-score": 0.9595463311598861,
457
- "precision": 0.9512039692293687,
458
- "recall": 0.9680363181544448,
459
- "support": 21587.0
460
  },
461
  "eval_O": {
462
- "f1-score": 0.9115367743822197,
463
- "precision": 0.9310035842293907,
464
- "recall": 0.8928673732454884,
465
- "support": 10473.0
466
  },
467
- "eval_accuracy": 0.9413731343283582,
468
- "eval_loss": 0.3427491784095764,
469
  "eval_macro avg": {
470
- "f1-score": 0.9170550335863505,
471
- "precision": 0.9161270190524938,
472
- "recall": 0.9184493786147926,
473
- "support": 33500.0
474
- },
475
- "eval_runtime": 4.7455,
476
- "eval_samples_per_second": 17.069,
477
- "eval_steps_per_second": 2.318,
478
  "eval_weighted avg": {
479
- "f1-score": 0.9411214734915179,
480
- "precision": 0.9412337452750731,
481
- "recall": 0.9413731343283582,
482
- "support": 33500.0
483
  },
484
  "step": 972
485
  },
486
  {
487
  "epoch": 12.35,
488
- "grad_norm": 0.47551417350769043,
489
  "learning_rate": 1.506172839506173e-05,
490
- "loss": 0.0222,
491
  "step": 1000
492
  },
493
  {
494
  "epoch": 13.0,
495
  "eval_B": {
496
- "f1-score": 0.8896434634974534,
497
- "precision": 0.8704318936877077,
498
- "recall": 0.9097222222222222,
499
- "support": 1440.0
500
  },
501
  "eval_I": {
502
- "f1-score": 0.9620136667970457,
503
- "precision": 0.9556591698665204,
504
- "recall": 0.9684532357437347,
505
- "support": 21587.0
506
  },
507
  "eval_O": {
508
- "f1-score": 0.9183177933177933,
509
- "precision": 0.9343808676746714,
510
- "recall": 0.9027976701995608,
511
- "support": 10473.0
512
  },
513
- "eval_accuracy": 0.9454029850746268,
514
- "eval_loss": 0.3816670775413513,
515
  "eval_macro avg": {
516
- "f1-score": 0.9233249745374308,
517
- "precision": 0.9201573104096331,
518
- "recall": 0.9269910427218392,
519
- "support": 33500.0
520
- },
521
- "eval_runtime": 4.7247,
522
- "eval_samples_per_second": 17.144,
523
- "eval_steps_per_second": 2.328,
524
  "eval_weighted avg": {
525
- "f1-score": 0.945242324238848,
526
- "precision": 0.9453435001186778,
527
- "recall": 0.9454029850746268,
528
- "support": 33500.0
529
  },
530
  "step": 1053
531
  },
532
  {
533
  "epoch": 14.0,
534
  "eval_B": {
535
- "f1-score": 0.8865272540281112,
536
- "precision": 0.8754231550440081,
537
- "recall": 0.8979166666666667,
538
- "support": 1440.0
539
  },
540
  "eval_I": {
541
- "f1-score": 0.9599206715247671,
542
- "precision": 0.9557331129172981,
543
- "recall": 0.9641450873210728,
544
- "support": 21587.0
545
  },
546
  "eval_O": {
547
- "f1-score": 0.9141367826632559,
548
- "precision": 0.9242631270739801,
549
- "recall": 0.9042299245679366,
550
- "support": 10473.0
551
  },
552
- "eval_accuracy": 0.9425671641791045,
553
- "eval_loss": 0.3827010989189148,
554
  "eval_macro avg": {
555
- "f1-score": 0.9201949027387114,
556
- "precision": 0.9184731316784287,
557
- "recall": 0.9220972261852255,
558
- "support": 33500.0
559
- },
560
- "eval_runtime": 4.7319,
561
- "eval_samples_per_second": 17.118,
562
- "eval_steps_per_second": 2.325,
563
  "eval_weighted avg": {
564
- "f1-score": 0.9424525763235196,
565
- "precision": 0.9424426203479068,
566
- "recall": 0.9425671641791045,
567
- "support": 33500.0
568
  },
569
  "step": 1134
570
  },
571
  {
572
  "epoch": 15.0,
573
  "eval_B": {
574
- "f1-score": 0.8871919472405415,
575
- "precision": 0.8868841082581541,
576
- "recall": 0.8875,
577
- "support": 1440.0
578
  },
579
  "eval_I": {
580
- "f1-score": 0.9616856471022177,
581
- "precision": 0.9571841585975862,
582
- "recall": 0.9662296752675221,
583
- "support": 21587.0
584
  },
585
  "eval_O": {
586
- "f1-score": 0.9181813798756087,
587
- "precision": 0.9273470977795092,
588
- "recall": 0.9091950730449728,
589
- "support": 10473.0
590
  },
591
- "eval_accuracy": 0.9450149253731344,
592
- "eval_loss": 0.39016807079315186,
593
  "eval_macro avg": {
594
- "f1-score": 0.9223529914061226,
595
- "precision": 0.9238051215450831,
596
- "recall": 0.920974916104165,
597
- "support": 33500.0
598
- },
599
- "eval_runtime": 4.7376,
600
- "eval_samples_per_second": 17.097,
601
- "eval_steps_per_second": 2.322,
602
  "eval_weighted avg": {
603
- "f1-score": 0.944882927148036,
604
- "precision": 0.9448344388830637,
605
- "recall": 0.9450149253731344,
606
- "support": 33500.0
607
  },
608
  "step": 1215
609
  },
610
  {
611
  "epoch": 16.0,
612
  "eval_B": {
613
- "f1-score": 0.8897338403041825,
614
- "precision": 0.8857536132140399,
615
- "recall": 0.89375,
616
- "support": 1440.0
617
  },
618
  "eval_I": {
619
- "f1-score": 0.9608010520729992,
620
- "precision": 0.9570693142121713,
621
- "recall": 0.9645620049103627,
622
- "support": 21587.0
623
  },
624
  "eval_O": {
625
- "f1-score": 0.9159121556540166,
626
- "precision": 0.9240112719852298,
627
- "recall": 0.9079537859257137,
628
- "support": 10473.0
629
  },
630
- "eval_accuracy": 0.943820895522388,
631
- "eval_loss": 0.40087994933128357,
632
  "eval_macro avg": {
633
- "f1-score": 0.9221490160103993,
634
- "precision": 0.9222780664704805,
635
- "recall": 0.9220885969453588,
636
- "support": 33500.0
637
- },
638
- "eval_runtime": 4.8145,
639
- "eval_samples_per_second": 16.824,
640
- "eval_steps_per_second": 2.285,
641
  "eval_weighted avg": {
642
- "f1-score": 0.9437127476806678,
643
- "precision": 0.9436689713560499,
644
- "recall": 0.943820895522388,
645
- "support": 33500.0
646
  },
647
  "step": 1296
648
  },
649
  {
650
  "epoch": 17.0,
651
  "eval_B": {
652
- "f1-score": 0.8904347826086957,
653
- "precision": 0.89198606271777,
654
- "recall": 0.8888888888888888,
655
- "support": 1440.0
656
  },
657
  "eval_I": {
658
- "f1-score": 0.9615171394252979,
659
- "precision": 0.9589034784611841,
660
- "recall": 0.9641450873210728,
661
- "support": 21587.0
662
  },
663
  "eval_O": {
664
- "f1-score": 0.9182546920750733,
665
- "precision": 0.9232625482625483,
666
- "recall": 0.9133008689009835,
667
- "support": 10473.0
668
  },
669
- "eval_accuracy": 0.9450149253731344,
670
- "eval_loss": 0.42787015438079834,
671
  "eval_macro avg": {
672
- "f1-score": 0.9234022047030224,
673
- "precision": 0.9247173631471673,
674
- "recall": 0.9221116150369818,
675
- "support": 33500.0
676
- },
677
- "eval_runtime": 4.7453,
678
- "eval_samples_per_second": 17.07,
679
- "eval_steps_per_second": 2.318,
680
  "eval_weighted avg": {
681
- "f1-score": 0.9449366556964975,
682
- "precision": 0.9448847160539952,
683
- "recall": 0.9450149253731344,
684
- "support": 33500.0
685
  },
686
  "step": 1377
687
  },
688
  {
689
  "epoch": 18.0,
690
  "eval_B": {
691
- "f1-score": 0.8895027624309393,
692
- "precision": 0.8846153846153846,
693
- "recall": 0.8944444444444445,
694
- "support": 1440.0
695
  },
696
  "eval_I": {
697
- "f1-score": 0.9607206210979795,
698
- "precision": 0.9555912007332722,
699
- "recall": 0.9659054060314078,
700
- "support": 21587.0
701
  },
702
  "eval_O": {
703
- "f1-score": 0.9159781610861478,
704
- "precision": 0.9271322378716745,
705
- "recall": 0.905089277188962,
706
- "support": 10473.0
707
  },
708
- "eval_accuracy": 0.943820895522388,
709
- "eval_loss": 0.422620564699173,
710
  "eval_macro avg": {
711
- "f1-score": 0.9220671815383555,
712
- "precision": 0.9224462744067772,
713
- "recall": 0.9218130425549381,
714
- "support": 33500.0
715
- },
716
- "eval_runtime": 4.7703,
717
- "eval_samples_per_second": 16.98,
718
- "eval_steps_per_second": 2.306,
719
  "eval_weighted avg": {
720
- "f1-score": 0.9436716210924735,
721
- "precision": 0.9436432636210552,
722
- "recall": 0.943820895522388,
723
- "support": 33500.0
724
  },
725
  "step": 1458
726
  },
727
  {
728
  "epoch": 18.52,
729
- "grad_norm": 0.1027584820985794,
730
  "learning_rate": 1.2592592592592593e-05,
731
- "loss": 0.0054,
732
  "step": 1500
733
  },
734
  {
735
  "epoch": 19.0,
736
  "eval_B": {
737
- "f1-score": 0.8924247665167762,
738
- "precision": 0.8890420399724328,
739
- "recall": 0.8958333333333334,
740
- "support": 1440.0
741
  },
742
  "eval_I": {
743
- "f1-score": 0.9599648245857633,
744
- "precision": 0.959121387283237,
745
- "recall": 0.9608097466067541,
746
- "support": 21587.0
747
  },
748
  "eval_O": {
749
- "f1-score": 0.9158252380724505,
750
- "precision": 0.9179777436684574,
751
- "recall": 0.913682803399217,
752
- "support": 10473.0
753
  },
754
- "eval_accuracy": 0.9432835820895522,
755
- "eval_loss": 0.4375981390476227,
756
  "eval_macro avg": {
757
- "f1-score": 0.9227382763916633,
758
- "precision": 0.9220470569747091,
759
- "recall": 0.9234419611131015,
760
- "support": 33500.0
761
- },
762
- "eval_runtime": 4.7487,
763
- "eval_samples_per_second": 17.057,
764
- "eval_steps_per_second": 2.316,
765
  "eval_weighted avg": {
766
- "f1-score": 0.9432623895656658,
767
- "precision": 0.9432464129636805,
768
- "recall": 0.9432835820895522,
769
- "support": 33500.0
770
  },
771
  "step": 1539
772
  },
773
  {
774
  "epoch": 20.0,
775
  "eval_B": {
776
- "f1-score": 0.8907273354015857,
777
- "precision": 0.8843258042436687,
778
- "recall": 0.8972222222222223,
779
- "support": 1440.0
780
  },
781
  "eval_I": {
782
- "f1-score": 0.9620682542229149,
783
- "precision": 0.9520958083832335,
784
- "recall": 0.972251818223931,
785
- "support": 21587.0
786
  },
787
  "eval_O": {
788
- "f1-score": 0.9177252296267345,
789
- "precision": 0.9396698349174587,
790
- "recall": 0.8967822018523823,
791
- "support": 10473.0
792
  },
793
- "eval_accuracy": 0.9454328358208955,
794
- "eval_loss": 0.44444534182548523,
795
  "eval_macro avg": {
796
- "f1-score": 0.9235069397504118,
797
- "precision": 0.9253638158481202,
798
- "recall": 0.922085414099512,
799
- "support": 33500.0
800
- },
801
- "eval_runtime": 4.7631,
802
- "eval_samples_per_second": 17.006,
803
- "eval_steps_per_second": 2.309,
804
  "eval_weighted avg": {
805
- "f1-score": 0.9451388387095264,
806
- "precision": 0.9452980165603072,
807
- "recall": 0.9454328358208955,
808
- "support": 33500.0
809
  },
810
  "step": 1620
811
  },
812
  {
813
  "epoch": 21.0,
814
  "eval_B": {
815
- "f1-score": 0.890790825059911,
816
- "precision": 0.8784604996623903,
817
- "recall": 0.9034722222222222,
818
- "support": 1440.0
819
  },
820
  "eval_I": {
821
- "f1-score": 0.9615375783600082,
822
- "precision": 0.9533284764593388,
823
- "recall": 0.9698892852179553,
824
- "support": 21587.0
825
  },
826
  "eval_O": {
827
- "f1-score": 0.9162201656113005,
828
- "precision": 0.9351695336581486,
829
- "recall": 0.8980234889716414,
830
- "support": 10473.0
831
  },
832
- "eval_accuracy": 0.9445671641791045,
833
- "eval_loss": 0.4363422691822052,
834
  "eval_macro avg": {
835
- "f1-score": 0.9228495230104065,
836
- "precision": 0.9223195032599593,
837
- "recall": 0.9237949988039397,
838
- "support": 33500.0
839
- },
840
- "eval_runtime": 4.7902,
841
- "eval_samples_per_second": 16.91,
842
- "eval_steps_per_second": 2.296,
843
  "eval_weighted avg": {
844
- "f1-score": 0.9443290831818184,
845
- "precision": 0.9444332975177129,
846
- "recall": 0.9445671641791045,
847
- "support": 33500.0
848
  },
849
  "step": 1701
850
  },
851
  {
852
  "epoch": 22.0,
853
  "eval_B": {
854
- "f1-score": 0.8907216494845361,
855
- "precision": 0.8816326530612245,
856
- "recall": 0.9,
857
- "support": 1440.0
858
  },
859
  "eval_I": {
860
- "f1-score": 0.961758728985775,
861
- "precision": 0.9588801399825022,
862
- "recall": 0.9646546532635383,
863
- "support": 21587.0
864
  },
865
  "eval_O": {
866
- "f1-score": 0.9186952756663139,
867
- "precision": 0.9258217783380199,
868
- "recall": 0.9116776472834909,
869
- "support": 10473.0
870
  },
871
- "eval_accuracy": 0.9453134328358209,
872
- "eval_loss": 0.3998025953769684,
873
  "eval_macro avg": {
874
- "f1-score": 0.9237252180455416,
875
- "precision": 0.9221115237939155,
876
- "recall": 0.925444100182343,
877
- "support": 33500.0
878
- },
879
- "eval_runtime": 4.7313,
880
- "eval_samples_per_second": 17.12,
881
- "eval_steps_per_second": 2.325,
882
  "eval_weighted avg": {
883
- "f1-score": 0.9452424023858794,
884
- "precision": 0.9452247190072991,
885
- "recall": 0.9453134328358209,
886
- "support": 33500.0
887
  },
888
  "step": 1782
889
  },
890
  {
891
  "epoch": 23.0,
892
  "eval_B": {
893
- "f1-score": 0.8938053097345132,
894
- "precision": 0.8765020026702269,
895
- "recall": 0.9118055555555555,
896
- "support": 1440.0
897
  },
898
  "eval_I": {
899
- "f1-score": 0.9620159689850926,
900
- "precision": 0.9584770313146641,
901
- "recall": 0.9655811367952934,
902
- "support": 21587.0
903
  },
904
  "eval_O": {
905
- "f1-score": 0.918660748745658,
906
- "precision": 0.9284251584592882,
907
- "recall": 0.9090995894204144,
908
- "support": 10473.0
909
  },
910
- "eval_accuracy": 0.9456119402985075,
911
- "eval_loss": 0.4420088231563568,
912
  "eval_macro avg": {
913
- "f1-score": 0.9248273424884212,
914
- "precision": 0.9211347308147264,
915
- "recall": 0.9288287605904211,
916
- "support": 33500.0
917
- },
918
- "eval_runtime": 4.7375,
919
- "eval_samples_per_second": 17.098,
920
- "eval_steps_per_second": 2.322,
921
  "eval_weighted avg": {
922
- "f1-score": 0.9455299220929005,
923
- "precision": 0.9455583057725045,
924
- "recall": 0.9456119402985075,
925
- "support": 33500.0
926
  },
927
  "step": 1863
928
  },
929
  {
930
  "epoch": 24.0,
931
  "eval_B": {
932
- "f1-score": 0.89482699554642,
933
- "precision": 0.8830290736984449,
934
- "recall": 0.9069444444444444,
935
- "support": 1440.0
936
  },
937
  "eval_I": {
938
- "f1-score": 0.962135833103575,
939
- "precision": 0.9544605005242285,
940
- "recall": 0.969935609394543,
941
- "support": 21587.0
942
  },
943
  "eval_O": {
944
- "f1-score": 0.9185192391885975,
945
- "precision": 0.9362356207854026,
946
- "recall": 0.9014608994557434,
947
- "support": 10473.0
948
  },
949
- "eval_accuracy": 0.945820895522388,
950
- "eval_loss": 0.4822453558444977,
951
  "eval_macro avg": {
952
- "f1-score": 0.9251606892795309,
953
- "precision": 0.924575065002692,
954
- "recall": 0.9261136510982436,
955
- "support": 33500.0
956
- },
957
- "eval_runtime": 4.7345,
958
- "eval_samples_per_second": 17.108,
959
- "eval_steps_per_second": 2.323,
960
  "eval_weighted avg": {
961
- "f1-score": 0.9456068386512209,
962
- "precision": 0.945692428281427,
963
- "recall": 0.945820895522388,
964
- "support": 33500.0
965
  },
966
  "step": 1944
967
  },
968
  {
969
  "epoch": 24.69,
970
- "grad_norm": 0.10262901335954666,
971
  "learning_rate": 1.0123456790123458e-05,
972
- "loss": 0.0037,
973
  "step": 2000
974
  },
975
  {
976
  "epoch": 25.0,
977
  "eval_B": {
978
- "f1-score": 0.8864342423196411,
979
- "precision": 0.8812628689087165,
980
- "recall": 0.8916666666666667,
981
- "support": 1440.0
982
  },
983
  "eval_I": {
984
- "f1-score": 0.9595285234126801,
985
- "precision": 0.9537320946409775,
986
- "recall": 0.9653958400889424,
987
- "support": 21587.0
988
  },
989
  "eval_O": {
990
- "f1-score": 0.9129445923058311,
991
- "precision": 0.9255298273155416,
992
- "recall": 0.9006970304592762,
993
- "support": 10473.0
994
  },
995
- "eval_accuracy": 0.942,
996
- "eval_loss": 0.46240004897117615,
997
  "eval_macro avg": {
998
- "f1-score": 0.9196357860127174,
999
- "precision": 0.9201749302884119,
1000
- "recall": 0.9192531790716284,
1001
- "support": 33500.0
1002
- },
1003
- "eval_runtime": 4.7408,
1004
- "eval_samples_per_second": 17.086,
1005
- "eval_steps_per_second": 2.32,
1006
  "eval_weighted avg": {
1007
- "f1-score": 0.941823171912501,
1008
- "precision": 0.9418002131259403,
1009
- "recall": 0.942,
1010
- "support": 33500.0
1011
  },
1012
  "step": 2025
1013
  },
1014
  {
1015
  "epoch": 26.0,
1016
  "eval_B": {
1017
- "f1-score": 0.892392049348869,
1018
- "precision": 0.8809201623815968,
1019
- "recall": 0.9041666666666667,
1020
- "support": 1440.0
1021
  },
1022
  "eval_I": {
1023
- "f1-score": 0.9613188432492171,
1024
- "precision": 0.9556420233463035,
1025
- "recall": 0.9670635104461018,
1026
- "support": 21587.0
1027
  },
1028
  "eval_O": {
1029
- "f1-score": 0.9172881355932202,
1030
- "precision": 0.9306278864105335,
1031
- "recall": 0.904325408192495,
1032
- "support": 10473.0
1033
  },
1034
- "eval_accuracy": 0.9447462686567164,
1035
- "eval_loss": 0.4685353934764862,
1036
  "eval_macro avg": {
1037
- "f1-score": 0.9236663427304355,
1038
- "precision": 0.9223966907128114,
1039
- "recall": 0.9251851951017546,
1040
- "support": 33500.0
1041
- },
1042
- "eval_runtime": 4.7522,
1043
- "eval_samples_per_second": 17.045,
1044
- "eval_steps_per_second": 2.315,
1045
  "eval_weighted avg": {
1046
- "f1-score": 0.9445908377418215,
1047
- "precision": 0.9446100073487664,
1048
- "recall": 0.9447462686567164,
1049
- "support": 33500.0
1050
  },
1051
  "step": 2106
1052
  },
1053
  {
1054
  "epoch": 27.0,
1055
  "eval_B": {
1056
- "f1-score": 0.8878923766816142,
1057
- "precision": 0.882111034955449,
1058
- "recall": 0.89375,
1059
- "support": 1440.0
1060
  },
1061
  "eval_I": {
1062
- "f1-score": 0.9603469923524712,
1063
- "precision": 0.9467098748762265,
1064
- "recall": 0.9743827303469681,
1065
- "support": 21587.0
1066
  },
1067
  "eval_O": {
1068
- "f1-score": 0.9131848640126133,
1069
- "precision": 0.943398147205538,
1070
- "recall": 0.8848467487825837,
1071
- "support": 10473.0
1072
  },
1073
- "eval_accuracy": 0.9429253731343283,
1074
- "eval_loss": 0.5011060833930969,
1075
  "eval_macro avg": {
1076
- "f1-score": 0.9204747443488995,
1077
- "precision": 0.9240730190124045,
1078
- "recall": 0.9176598263765173,
1079
- "support": 33500.0
1080
- },
1081
- "eval_runtime": 4.7877,
1082
- "eval_samples_per_second": 16.918,
1083
- "eval_steps_per_second": 2.298,
1084
  "eval_weighted avg": {
1085
- "f1-score": 0.942488376929505,
1086
- "precision": 0.9428977538797776,
1087
- "recall": 0.9429253731343283,
1088
- "support": 33500.0
1089
  },
1090
  "step": 2187
1091
  },
1092
  {
1093
  "epoch": 28.0,
1094
  "eval_B": {
1095
- "f1-score": 0.8914835164835164,
1096
- "precision": 0.8817934782608695,
1097
- "recall": 0.9013888888888889,
1098
- "support": 1440.0
1099
  },
1100
  "eval_I": {
1101
- "f1-score": 0.9599201652393882,
1102
- "precision": 0.9618175053483397,
1103
- "recall": 0.9580302960114884,
1104
- "support": 21587.0
1105
  },
1106
  "eval_O": {
1107
- "f1-score": 0.9164245916472212,
1108
- "precision": 0.9141174235227056,
1109
- "recall": 0.9187434355008116,
1110
- "support": 10473.0
1111
  },
1112
- "eval_accuracy": 0.9433134328358209,
1113
- "eval_loss": 0.4657354950904846,
1114
  "eval_macro avg": {
1115
- "f1-score": 0.9226094244567086,
1116
- "precision": 0.919242802377305,
1117
- "recall": 0.9260542068003962,
1118
- "support": 33500.0
1119
- },
1120
- "eval_runtime": 4.7715,
1121
- "eval_samples_per_second": 16.976,
1122
- "eval_steps_per_second": 2.305,
1123
  "eval_weighted avg": {
1124
- "f1-score": 0.9433805259426951,
1125
- "precision": 0.9434653394986137,
1126
- "recall": 0.9433134328358209,
1127
- "support": 33500.0
1128
  },
1129
  "step": 2268
1130
  },
1131
  {
1132
  "epoch": 29.0,
1133
  "eval_B": {
1134
- "f1-score": 0.8932038834951457,
1135
- "precision": 0.8623141564318035,
1136
- "recall": 0.9263888888888889,
1137
- "support": 1440.0
1138
  },
1139
  "eval_I": {
1140
- "f1-score": 0.9610192139335576,
1141
- "precision": 0.9559099867088318,
1142
- "recall": 0.9661833510909343,
1143
- "support": 21587.0
1144
  },
1145
  "eval_O": {
1146
- "f1-score": 0.9164846896685592,
1147
- "precision": 0.9318136964673377,
1148
- "recall": 0.9016518667048601,
1149
- "support": 10473.0
1150
  },
1151
- "eval_accuracy": 0.9442985074626866,
1152
- "eval_loss": 0.45431333780288696,
1153
  "eval_macro avg": {
1154
- "f1-score": 0.9235692623657542,
1155
- "precision": 0.9166792798693243,
1156
- "recall": 0.931408035561561,
1157
- "support": 33500.0
1158
- },
1159
- "eval_runtime": 4.7759,
1160
- "eval_samples_per_second": 16.96,
1161
- "eval_steps_per_second": 2.303,
1162
  "eval_weighted avg": {
1163
- "f1-score": 0.9441814781586727,
1164
- "precision": 0.9443536152670977,
1165
- "recall": 0.9442985074626866,
1166
- "support": 33500.0
1167
  },
1168
  "step": 2349
1169
  },
1170
  {
1171
  "epoch": 30.0,
1172
  "eval_B": {
1173
- "f1-score": 0.8953290146607569,
1174
- "precision": 0.8794373744139317,
1175
- "recall": 0.9118055555555555,
1176
- "support": 1440.0
1177
  },
1178
  "eval_I": {
1179
- "f1-score": 0.9620589989448088,
1180
- "precision": 0.9528786295269687,
1181
- "recall": 0.9714179830453513,
1182
- "support": 21587.0
1183
  },
1184
  "eval_O": {
1185
- "f1-score": 0.9176964782884776,
1186
- "precision": 0.9394,
1187
- "recall": 0.896973169101499,
1188
- "support": 10473.0
1189
  },
1190
- "eval_accuracy": 0.9455820895522388,
1191
- "eval_loss": 0.4872027635574341,
1192
  "eval_macro avg": {
1193
- "f1-score": 0.9250281639646811,
1194
- "precision": 0.9239053346469669,
1195
- "recall": 0.926732235900802,
1196
- "support": 33500.0
1197
- },
1198
- "eval_runtime": 4.7685,
1199
- "eval_samples_per_second": 16.987,
1200
- "eval_steps_per_second": 2.307,
1201
  "eval_weighted avg": {
1202
- "f1-score": 0.945321689804427,
1203
- "precision": 0.9455079699926786,
1204
- "recall": 0.9455820895522388,
1205
- "support": 33500.0
1206
  },
1207
  "step": 2430
1208
  },
1209
  {
1210
  "epoch": 30.86,
1211
- "grad_norm": 8.135757446289062,
1212
  "learning_rate": 7.654320987654322e-06,
1213
- "loss": 0.0025,
1214
  "step": 2500
1215
  },
1216
  {
1217
  "epoch": 31.0,
1218
  "eval_B": {
1219
- "f1-score": 0.8891177480260899,
1220
- "precision": 0.8791581805838425,
1221
- "recall": 0.8993055555555556,
1222
- "support": 1440.0
1223
  },
1224
  "eval_I": {
1225
- "f1-score": 0.9598950783461034,
1226
- "precision": 0.9535978787601719,
1227
- "recall": 0.9662759994441099,
1228
- "support": 21587.0
1229
  },
1230
  "eval_O": {
1231
- "f1-score": 0.9137011538834482,
1232
- "precision": 0.9281000689451394,
1233
- "recall": 0.8997421942136924,
1234
- "support": 10473.0
1235
  },
1236
- "eval_accuracy": 0.9425970149253732,
1237
- "eval_loss": 0.4678420424461365,
1238
  "eval_macro avg": {
1239
- "f1-score": 0.9209046600852139,
1240
- "precision": 0.9202853760963846,
1241
- "recall": 0.9217745830711194,
1242
- "support": 33500.0
1243
- },
1244
- "eval_runtime": 4.7618,
1245
- "eval_samples_per_second": 17.01,
1246
- "eval_steps_per_second": 2.31,
1247
  "eval_weighted avg": {
1248
- "f1-score": 0.9424112477025748,
1249
- "precision": 0.9424267824148957,
1250
- "recall": 0.9425970149253732,
1251
- "support": 33500.0
1252
  },
1253
  "step": 2511
1254
  },
1255
  {
1256
  "epoch": 32.0,
1257
  "eval_B": {
1258
- "f1-score": 0.8871355060034306,
1259
- "precision": 0.8766101694915254,
1260
- "recall": 0.8979166666666667,
1261
- "support": 1440.0
1262
  },
1263
  "eval_I": {
1264
- "f1-score": 0.9603147215128719,
1265
- "precision": 0.9538848263254114,
1266
- "recall": 0.966831889563163,
1267
- "support": 21587.0
1268
  },
1269
  "eval_O": {
1270
- "f1-score": 0.9144436899796294,
1271
- "precision": 0.9292262198127156,
1272
- "recall": 0.9001241287119259,
1273
- "support": 10473.0
1274
  },
1275
- "eval_accuracy": 0.9430149253731344,
1276
- "eval_loss": 0.4877029359340668,
1277
  "eval_macro avg": {
1278
- "f1-score": 0.9206313058319773,
1279
- "precision": 0.9199070718765507,
1280
- "recall": 0.9216242283139185,
1281
- "support": 33500.0
1282
- },
1283
- "eval_runtime": 4.759,
1284
- "eval_samples_per_second": 17.02,
1285
- "eval_steps_per_second": 2.311,
1286
  "eval_weighted avg": {
1287
- "f1-score": 0.9428285906597003,
1288
- "precision": 0.942854226568747,
1289
- "recall": 0.9430149253731344,
1290
- "support": 33500.0
1291
  },
1292
  "step": 2592
1293
  },
1294
  {
1295
  "epoch": 33.0,
1296
  "eval_B": {
1297
- "f1-score": 0.8910756701730572,
1298
- "precision": 0.8712674187126742,
1299
- "recall": 0.9118055555555555,
1300
- "support": 1440.0
1301
  },
1302
  "eval_I": {
1303
- "f1-score": 0.9623726096415894,
1304
- "precision": 0.9516713470423046,
1305
- "recall": 0.9733172742854496,
1306
- "support": 21587.0
1307
  },
1308
  "eval_O": {
1309
- "f1-score": 0.9176966843241121,
1310
- "precision": 0.9435199193141705,
1311
- "recall": 0.8932493077437219,
1312
- "support": 10473.0
1313
  },
1314
- "eval_accuracy": 0.9456417910447761,
1315
- "eval_loss": 0.49512049555778503,
1316
  "eval_macro avg": {
1317
- "f1-score": 0.9237149880462528,
1318
- "precision": 0.9221528950230496,
1319
- "recall": 0.9261240458615757,
1320
- "support": 33500.0
1321
- },
1322
- "eval_runtime": 4.765,
1323
- "eval_samples_per_second": 16.999,
1324
- "eval_steps_per_second": 2.308,
1325
  "eval_weighted avg": {
1326
- "f1-score": 0.9453410108748842,
1327
- "precision": 0.9456668228813668,
1328
- "recall": 0.9456417910447761,
1329
- "support": 33500.0
1330
  },
1331
  "step": 2673
1332
  },
1333
  {
1334
  "epoch": 34.0,
1335
  "eval_B": {
1336
- "f1-score": 0.8933423913043478,
1337
- "precision": 0.8743351063829787,
1338
- "recall": 0.9131944444444444,
1339
- "support": 1440.0
1340
  },
1341
  "eval_I": {
1342
- "f1-score": 0.9620096067661051,
1343
- "precision": 0.9546159459952563,
1344
- "recall": 0.9695186918052532,
1345
- "support": 21587.0
1346
  },
1347
  "eval_O": {
1348
- "f1-score": 0.9179849111706012,
1349
- "precision": 0.9362589356632248,
1350
- "recall": 0.9004105795856011,
1351
- "support": 10473.0
1352
  },
1353
- "eval_accuracy": 0.9454925373134329,
1354
- "eval_loss": 0.47959965467453003,
1355
  "eval_macro avg": {
1356
- "f1-score": 0.9244456364136847,
1357
- "precision": 0.9217366626804866,
1358
- "recall": 0.9277079052784329,
1359
- "support": 33500.0
1360
- },
1361
- "eval_runtime": 4.7437,
1362
- "eval_samples_per_second": 17.075,
1363
- "eval_steps_per_second": 2.319,
1364
  "eval_weighted avg": {
1365
- "f1-score": 0.9452946387888919,
1366
- "precision": 0.9454261735102102,
1367
- "recall": 0.9454925373134329,
1368
- "support": 33500.0
1369
  },
1370
  "step": 2754
1371
  },
1372
  {
1373
  "epoch": 35.0,
1374
  "eval_B": {
1375
- "f1-score": 0.8896504455106238,
1376
- "precision": 0.878213802435724,
1377
- "recall": 0.9013888888888889,
1378
- "support": 1440.0
1379
  },
1380
  "eval_I": {
1381
- "f1-score": 0.9619017054832203,
1382
- "precision": 0.9520805917320869,
1383
- "recall": 0.9719275489878167,
1384
- "support": 21587.0
1385
  },
1386
  "eval_O": {
1387
- "f1-score": 0.917587251930785,
1388
- "precision": 0.9400100150225338,
1389
- "recall": 0.896209300105032,
1390
- "support": 10473.0
1391
  },
1392
- "eval_accuracy": 0.9452238805970149,
1393
- "eval_loss": 0.4806292951107025,
1394
  "eval_macro avg": {
1395
- "f1-score": 0.923046467641543,
1396
- "precision": 0.9234348030634482,
1397
- "recall": 0.9231752459939125,
1398
- "support": 33500.0
1399
- },
1400
- "eval_runtime": 4.73,
1401
- "eval_samples_per_second": 17.125,
1402
- "eval_steps_per_second": 2.326,
1403
  "eval_weighted avg": {
1404
- "f1-score": 0.9449420909633638,
1405
- "precision": 0.9451318357181789,
1406
- "recall": 0.9452238805970149,
1407
- "support": 33500.0
1408
  },
1409
  "step": 2835
1410
  },
1411
  {
1412
  "epoch": 36.0,
1413
  "eval_B": {
1414
- "f1-score": 0.8926979773740144,
1415
- "precision": 0.8815165876777251,
1416
- "recall": 0.9041666666666667,
1417
- "support": 1440.0
1418
  },
1419
  "eval_I": {
1420
- "f1-score": 0.9623375129414473,
1421
- "precision": 0.9559374714324893,
1422
- "recall": 0.9688238291564367,
1423
- "support": 21587.0
1424
  },
1425
  "eval_O": {
1426
- "f1-score": 0.9190998156950239,
1427
- "precision": 0.9339576145884673,
1428
- "recall": 0.9047073426907285,
1429
- "support": 10473.0
1430
  },
1431
- "eval_accuracy": 0.946,
1432
- "eval_loss": 0.4832863509654999,
1433
  "eval_macro avg": {
1434
- "f1-score": 0.9247117686701619,
1435
- "precision": 0.923803891232894,
1436
- "recall": 0.9258992795046107,
1437
- "support": 33500.0
1438
- },
1439
- "eval_runtime": 4.7435,
1440
- "eval_samples_per_second": 17.076,
1441
- "eval_steps_per_second": 2.319,
1442
  "eval_weighted avg": {
1443
- "f1-score": 0.9458267865390921,
1444
- "precision": 0.9458669904374355,
1445
- "recall": 0.946,
1446
- "support": 33500.0
1447
  },
1448
  "step": 2916
1449
  },
1450
  {
1451
  "epoch": 37.0,
1452
  "eval_B": {
1453
- "f1-score": 0.8904958677685951,
1454
- "precision": 0.8831967213114754,
1455
- "recall": 0.8979166666666667,
1456
- "support": 1440.0
1457
  },
1458
  "eval_I": {
1459
- "f1-score": 0.9614206416841329,
1460
- "precision": 0.9560696289509849,
1461
- "recall": 0.966831889563163,
1462
- "support": 21587.0
1463
  },
1464
  "eval_O": {
1465
- "f1-score": 0.9176459209826394,
1466
- "precision": 0.9296492259455222,
1467
- "recall": 0.9059486298099876,
1468
- "support": 10473.0
1469
  },
1470
- "eval_accuracy": 0.9448358208955224,
1471
- "eval_loss": 0.4824429750442505,
1472
  "eval_macro avg": {
1473
- "f1-score": 0.9231874768117891,
1474
- "precision": 0.9229718587359942,
1475
- "recall": 0.9235657286799391,
1476
- "support": 33500.0
1477
- },
1478
- "eval_runtime": 4.7583,
1479
- "eval_samples_per_second": 17.023,
1480
- "eval_steps_per_second": 2.312,
1481
  "eval_weighted avg": {
1482
- "f1-score": 0.9446867812559204,
1483
- "precision": 0.9446774537964444,
1484
- "recall": 0.9448358208955224,
1485
- "support": 33500.0
1486
  },
1487
  "step": 2997
1488
  },
1489
  {
1490
  "epoch": 37.04,
1491
- "grad_norm": 0.2996862828731537,
1492
  "learning_rate": 5.185185185185185e-06,
1493
- "loss": 0.0016,
1494
  "step": 3000
1495
  },
1496
  {
1497
  "epoch": 38.0,
1498
  "eval_B": {
1499
- "f1-score": 0.8921302578018996,
1500
- "precision": 0.8720159151193634,
1501
- "recall": 0.9131944444444444,
1502
- "support": 1440.0
1503
  },
1504
  "eval_I": {
1505
- "f1-score": 0.9625111778598124,
1506
- "precision": 0.9529192772178334,
1507
- "recall": 0.9722981424005188,
1508
- "support": 21587.0
1509
  },
1510
  "eval_O": {
1511
- "f1-score": 0.9186359410930084,
1512
- "precision": 0.9420028095524784,
1513
- "recall": 0.8964002673541488,
1514
- "support": 10473.0
1515
  },
1516
- "eval_accuracy": 0.9460298507462687,
1517
- "eval_loss": 0.49525222182273865,
1518
  "eval_macro avg": {
1519
- "f1-score": 0.9244257922515735,
1520
- "precision": 0.9223126672965584,
1521
- "recall": 0.9272976180663707,
1522
- "support": 33500.0
1523
- },
1524
- "eval_runtime": 4.7487,
1525
- "eval_samples_per_second": 17.057,
1526
- "eval_steps_per_second": 2.316,
1527
  "eval_weighted avg": {
1528
- "f1-score": 0.9457692710078085,
1529
- "precision": 0.9460288590900406,
1530
- "recall": 0.9460298507462687,
1531
- "support": 33500.0
1532
  },
1533
  "step": 3078
1534
  },
1535
  {
1536
  "epoch": 39.0,
1537
  "eval_B": {
1538
- "f1-score": 0.8914011647824598,
1539
- "precision": 0.8796484110885734,
1540
- "recall": 0.9034722222222222,
1541
- "support": 1440.0
1542
  },
1543
  "eval_I": {
1544
- "f1-score": 0.9630648600316389,
1545
- "precision": 0.9533817521561507,
1546
- "recall": 0.9729466808727475,
1547
- "support": 21587.0
1548
  },
1549
  "eval_O": {
1550
- "f1-score": 0.9199569976544176,
1551
- "precision": 0.9421479331398258,
1552
- "recall": 0.8987873579681085,
1553
- "support": 10473.0
1554
  },
1555
- "eval_accuracy": 0.9467761194029851,
1556
- "eval_loss": 0.4956269860267639,
1557
  "eval_macro avg": {
1558
- "f1-score": 0.9248076741561722,
1559
- "precision": 0.9250593654615167,
1560
- "recall": 0.9250687536876927,
1561
- "support": 33500.0
1562
- },
1563
- "eval_runtime": 4.7588,
1564
- "eval_samples_per_second": 17.021,
1565
- "eval_steps_per_second": 2.312,
1566
  "eval_weighted avg": {
1567
- "f1-score": 0.9465077148425208,
1568
- "precision": 0.9467003253592766,
1569
- "recall": 0.9467761194029851,
1570
- "support": 33500.0
1571
  },
1572
  "step": 3159
1573
  },
1574
  {
1575
  "epoch": 40.0,
1576
  "eval_B": {
1577
- "f1-score": 0.8935439560439561,
1578
- "precision": 0.8838315217391305,
1579
- "recall": 0.9034722222222222,
1580
- "support": 1440.0
1581
  },
1582
  "eval_I": {
1583
- "f1-score": 0.962365961745999,
1584
- "precision": 0.9541067200874158,
1585
- "recall": 0.9707694445731228,
1586
- "support": 21587.0
1587
  },
1588
  "eval_O": {
1589
- "f1-score": 0.9186346593952379,
1590
- "precision": 0.9373012718600954,
1591
- "recall": 0.9006970304592762,
1592
- "support": 10473.0
1593
  },
1594
- "eval_accuracy": 0.9459701492537314,
1595
- "eval_loss": 0.48434382677078247,
1596
  "eval_macro avg": {
1597
- "f1-score": 0.9248481923950643,
1598
- "precision": 0.9250798378955473,
1599
- "recall": 0.9249795657515404,
1600
- "support": 33500.0
1601
- },
1602
- "eval_runtime": 4.7973,
1603
- "eval_samples_per_second": 16.884,
1604
- "eval_steps_per_second": 2.293,
1605
  "eval_weighted avg": {
1606
- "f1-score": 0.9457360627092687,
1607
- "precision": 0.9458321008364828,
1608
- "recall": 0.9459701492537314,
1609
- "support": 33500.0
1610
  },
1611
  "step": 3240
1612
  }
@@ -1616,7 +1616,7 @@
1616
  "num_input_tokens_seen": 0,
1617
  "num_train_epochs": 50,
1618
  "save_steps": 500,
1619
- "total_flos": 5733604413432000.0,
1620
  "train_batch_size": 4,
1621
  "trial_name": null,
1622
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_B": {
14
+ "f1-score": 0.8331822302810515,
15
+ "precision": 0.7901977644024075,
16
+ "recall": 0.8811121764141898,
17
+ "support": 1043.0
18
  },
19
  "eval_I": {
20
+ "f1-score": 0.9471294753175105,
21
+ "precision": 0.9443616777446711,
22
+ "recall": 0.9499135446685879,
23
+ "support": 17350.0
24
  },
25
  "eval_O": {
26
+ "f1-score": 0.8992868897421833,
27
+ "precision": 0.9103731674811195,
28
+ "recall": 0.8884673748103187,
29
+ "support": 9226.0
30
  },
31
+ "eval_accuracy": 0.9267895289474637,
32
+ "eval_loss": 0.20478281378746033,
33
  "eval_macro avg": {
34
+ "f1-score": 0.8931995317802484,
35
+ "precision": 0.8816442032093993,
36
+ "recall": 0.9064976986310321,
37
+ "support": 27619.0
38
+ },
39
+ "eval_runtime": 4.6128,
40
+ "eval_samples_per_second": 17.343,
41
+ "eval_steps_per_second": 2.168,
42
  "eval_weighted avg": {
43
+ "f1-score": 0.9268447919078652,
44
+ "precision": 0.9271861479533134,
45
+ "recall": 0.9267895289474637,
46
+ "support": 27619.0
47
  },
48
  "step": 81
49
  },
50
  {
51
  "epoch": 2.0,
52
  "eval_B": {
53
+ "f1-score": 0.8682547399124939,
54
+ "precision": 0.8806706114398422,
55
+ "recall": 0.8561840843720039,
56
+ "support": 1043.0
57
  },
58
  "eval_I": {
59
+ "f1-score": 0.9590983771942466,
60
+ "precision": 0.9592919333448654,
61
+ "recall": 0.9589048991354466,
62
+ "support": 17350.0
63
  },
64
  "eval_O": {
65
+ "f1-score": 0.9226525313717006,
66
+ "precision": 0.9208594256100194,
67
+ "recall": 0.9244526338608281,
68
+ "support": 9226.0
69
  },
70
+ "eval_accuracy": 0.9435171439950758,
71
+ "eval_loss": 0.16574698686599731,
72
  "eval_macro avg": {
73
+ "f1-score": 0.9166685494928136,
74
+ "precision": 0.9202739901315756,
75
+ "recall": 0.9131805391227594,
76
+ "support": 27619.0
77
+ },
78
+ "eval_runtime": 4.6238,
79
+ "eval_samples_per_second": 17.302,
80
+ "eval_steps_per_second": 2.163,
81
  "eval_weighted avg": {
82
+ "f1-score": 0.9434932036816766,
83
+ "precision": 0.9434846863370583,
84
+ "recall": 0.9435171439950758,
85
+ "support": 27619.0
86
  },
87
  "step": 162
88
  },
89
  {
90
  "epoch": 3.0,
91
  "eval_B": {
92
+ "f1-score": 0.8791725434884815,
93
+ "precision": 0.8625461254612546,
94
+ "recall": 0.8964525407478428,
95
+ "support": 1043.0
96
  },
97
  "eval_I": {
98
+ "f1-score": 0.9588372159825352,
99
+ "precision": 0.9557349825345015,
100
+ "recall": 0.9619596541786744,
101
+ "support": 17350.0
102
  },
103
  "eval_O": {
104
+ "f1-score": 0.9204284621270084,
105
+ "precision": 0.9282407407407407,
106
+ "recall": 0.9127465857359636,
107
+ "support": 9226.0
108
  },
109
+ "eval_accuracy": 0.9430464535283681,
110
+ "eval_loss": 0.15394224226474762,
111
  "eval_macro avg": {
112
+ "f1-score": 0.9194794071993417,
113
+ "precision": 0.9155072829121655,
114
+ "recall": 0.9237195935541602,
115
+ "support": 27619.0
116
+ },
117
+ "eval_runtime": 4.6233,
118
+ "eval_samples_per_second": 17.303,
119
+ "eval_steps_per_second": 2.163,
120
  "eval_weighted avg": {
121
+ "f1-score": 0.9429985029052194,
122
+ "precision": 0.9430314866542512,
123
+ "recall": 0.9430464535283681,
124
+ "support": 27619.0
125
  },
126
  "step": 243
127
  },
128
  {
129
  "epoch": 4.0,
130
  "eval_B": {
131
+ "f1-score": 0.8800746616892207,
132
+ "precision": 0.8572727272727273,
133
+ "recall": 0.9041227229146692,
134
+ "support": 1043.0
135
  },
136
  "eval_I": {
137
+ "f1-score": 0.9537844036697248,
138
+ "precision": 0.9488876212207644,
139
+ "recall": 0.9587319884726225,
140
+ "support": 17350.0
141
  },
142
  "eval_O": {
143
+ "f1-score": 0.9100192149327477,
144
+ "precision": 0.9220157970853265,
145
+ "recall": 0.8983308042488619,
146
+ "support": 9226.0
147
  },
148
+ "eval_accuracy": 0.9364929939534379,
149
+ "eval_loss": 0.19801080226898193,
150
  "eval_macro avg": {
151
+ "f1-score": 0.9146260934305644,
152
+ "precision": 0.9093920485262728,
153
+ "recall": 0.9203951718787179,
154
+ "support": 27619.0
155
+ },
156
+ "eval_runtime": 4.6132,
157
+ "eval_samples_per_second": 17.342,
158
+ "eval_steps_per_second": 2.168,
159
  "eval_weighted avg": {
160
+ "f1-score": 0.9363812792925563,
161
+ "precision": 0.9364514800186444,
162
+ "recall": 0.9364929939534379,
163
+ "support": 27619.0
164
  },
165
  "step": 324
166
  },
167
  {
168
  "epoch": 5.0,
169
  "eval_B": {
170
+ "f1-score": 0.883877159309021,
171
+ "precision": 0.8847262247838616,
172
+ "recall": 0.8830297219558965,
173
+ "support": 1043.0
174
  },
175
  "eval_I": {
176
+ "f1-score": 0.9591500482927107,
177
+ "precision": 0.9456643513331839,
178
+ "recall": 0.9730259365994236,
179
+ "support": 17350.0
180
  },
181
  "eval_O": {
182
+ "f1-score": 0.9191176470588236,
183
+ "precision": 0.9454503781801513,
184
+ "recall": 0.8942120095382614,
185
+ "support": 9226.0
186
  },
187
+ "eval_accuracy": 0.9432999022412107,
188
+ "eval_loss": 0.20074255764484406,
189
  "eval_macro avg": {
190
+ "f1-score": 0.9207149515535183,
191
+ "precision": 0.9252803180990656,
192
+ "recall": 0.9167558893645271,
193
+ "support": 27619.0
194
+ },
195
+ "eval_runtime": 4.6092,
196
+ "eval_samples_per_second": 17.356,
197
+ "eval_steps_per_second": 2.17,
198
  "eval_weighted avg": {
199
+ "f1-score": 0.9429348139614955,
200
+ "precision": 0.9432916158141273,
201
+ "recall": 0.9432999022412107,
202
+ "support": 27619.0
203
  },
204
  "step": 405
205
  },
206
  {
207
  "epoch": 6.0,
208
  "eval_B": {
209
+ "f1-score": 0.8897338403041826,
210
+ "precision": 0.882186616399623,
211
+ "recall": 0.8974113135186961,
212
+ "support": 1043.0
213
  },
214
  "eval_I": {
215
+ "f1-score": 0.9624160683661308,
216
+ "precision": 0.9707969272268809,
217
+ "recall": 0.954178674351585,
218
+ "support": 17350.0
219
  },
220
  "eval_O": {
221
+ "f1-score": 0.9297955261331482,
222
+ "precision": 0.9161493950552341,
223
+ "recall": 0.9438543247344461,
224
+ "support": 9226.0
225
  },
226
+ "eval_accuracy": 0.9485861182519281,
227
+ "eval_loss": 0.16565027832984924,
228
  "eval_macro avg": {
229
+ "f1-score": 0.9273151449344872,
230
+ "precision": 0.9230443128939126,
231
+ "recall": 0.9318147708682424,
232
+ "support": 27619.0
233
+ },
234
+ "eval_runtime": 4.6219,
235
+ "eval_samples_per_second": 17.309,
236
+ "eval_steps_per_second": 2.164,
237
  "eval_weighted avg": {
238
+ "f1-score": 0.9487745648174827,
239
+ "precision": 0.9491959030765336,
240
+ "recall": 0.9485861182519281,
241
+ "support": 27619.0
242
  },
243
  "step": 486
244
  },
245
  {
246
  "epoch": 6.17,
247
+ "grad_norm": 4.767655849456787,
248
  "learning_rate": 1.7530864197530865e-05,
249
+ "loss": 0.1585,
250
  "step": 500
251
  },
252
  {
253
  "epoch": 7.0,
254
  "eval_B": {
255
+ "f1-score": 0.8802588996763754,
256
+ "precision": 0.85,
257
+ "recall": 0.912751677852349,
258
+ "support": 1043.0
259
  },
260
  "eval_I": {
261
+ "f1-score": 0.958423523458803,
262
+ "precision": 0.9354661691269275,
263
+ "recall": 0.9825360230547551,
264
+ "support": 17350.0
265
  },
266
  "eval_O": {
267
+ "f1-score": 0.9139526911210146,
268
+ "precision": 0.9664088931851136,
269
+ "recall": 0.8668978972469109,
270
+ "support": 9226.0
271
  },
272
+ "eval_accuracy": 0.9412723125384699,
273
+ "eval_loss": 0.2591131329536438,
274
  "eval_macro avg": {
275
+ "f1-score": 0.9175450380853977,
276
+ "precision": 0.917291687437347,
277
+ "recall": 0.9207285327180049,
278
+ "support": 27619.0
279
+ },
280
+ "eval_runtime": 4.617,
281
+ "eval_samples_per_second": 17.327,
282
+ "eval_steps_per_second": 2.166,
283
  "eval_weighted avg": {
284
+ "f1-score": 0.9406164485555297,
285
+ "precision": 0.9425749115781907,
286
+ "recall": 0.9412723125384699,
287
+ "support": 27619.0
288
  },
289
  "step": 567
290
  },
291
  {
292
  "epoch": 8.0,
293
  "eval_B": {
294
+ "f1-score": 0.8687910028116214,
295
+ "precision": 0.84967919340055,
296
+ "recall": 0.8887823585810163,
297
+ "support": 1043.0
298
  },
299
  "eval_I": {
300
+ "f1-score": 0.9595010203788336,
301
+ "precision": 0.956997878562009,
302
+ "recall": 0.9620172910662824,
303
+ "support": 17350.0
304
  },
305
  "eval_O": {
306
+ "f1-score": 0.9212035166275324,
307
+ "precision": 0.9282491471332673,
308
+ "recall": 0.9142640364188164,
309
+ "support": 9226.0
310
  },
311
+ "eval_accuracy": 0.9432999022412107,
312
+ "eval_loss": 0.2693409323692322,
313
  "eval_macro avg": {
314
+ "f1-score": 0.9164985132726624,
315
+ "precision": 0.9116420730319422,
316
+ "recall": 0.9216878953553717,
317
+ "support": 27619.0
318
+ },
319
+ "eval_runtime": 4.6153,
320
+ "eval_samples_per_second": 17.333,
321
+ "eval_steps_per_second": 2.167,
322
  "eval_weighted avg": {
323
+ "f1-score": 0.9432823550422136,
324
+ "precision": 0.9433417293609165,
325
+ "recall": 0.9432999022412107,
326
+ "support": 27619.0
327
  },
328
  "step": 648
329
  },
330
  {
331
  "epoch": 9.0,
332
  "eval_B": {
333
+ "f1-score": 0.8645343367826904,
334
+ "precision": 0.8485687903970452,
335
+ "recall": 0.8811121764141898,
336
+ "support": 1043.0
337
  },
338
  "eval_I": {
339
+ "f1-score": 0.9567003348057282,
340
+ "precision": 0.934535260814599,
341
+ "recall": 0.9799423631123919,
342
+ "support": 17350.0
343
  },
344
  "eval_O": {
345
+ "f1-score": 0.9104673003585861,
346
+ "precision": 0.9586479683567062,
347
+ "recall": 0.8668978972469109,
348
+ "support": 9226.0
349
  },
350
+ "eval_accuracy": 0.9384481697382236,
351
+ "eval_loss": 0.3087053894996643,
352
  "eval_macro avg": {
353
+ "f1-score": 0.910567323982335,
354
+ "precision": 0.9139173398561168,
355
+ "recall": 0.9093174789244975,
356
+ "support": 27619.0
357
+ },
358
+ "eval_runtime": 4.6156,
359
+ "eval_samples_per_second": 17.333,
360
+ "eval_steps_per_second": 2.167,
361
  "eval_weighted avg": {
362
+ "f1-score": 0.9377758584761231,
363
+ "precision": 0.9393435743356523,
364
+ "recall": 0.9384481697382236,
365
+ "support": 27619.0
366
  },
367
  "step": 729
368
  },
369
  {
370
  "epoch": 10.0,
371
  "eval_B": {
372
+ "f1-score": 0.8803126526624329,
373
+ "precision": 0.897410358565737,
374
+ "recall": 0.8638542665388304,
375
+ "support": 1043.0
376
  },
377
  "eval_I": {
378
+ "f1-score": 0.9590983771942466,
379
+ "precision": 0.9592919333448654,
380
+ "recall": 0.9589048991354466,
381
+ "support": 17350.0
382
  },
383
  "eval_O": {
384
+ "f1-score": 0.9220456265542221,
385
+ "precision": 0.9197584124245038,
386
+ "recall": 0.9243442445263386,
387
+ "support": 9226.0
388
  },
389
+ "eval_accuracy": 0.9437705927079184,
390
+ "eval_loss": 0.2796257436275482,
391
  "eval_macro avg": {
392
+ "f1-score": 0.9204855521369671,
393
+ "precision": 0.9254869014450354,
394
+ "recall": 0.9157011367335386,
395
+ "support": 27619.0
396
+ },
397
+ "eval_runtime": 4.641,
398
+ "eval_samples_per_second": 17.238,
399
+ "eval_steps_per_second": 2.155,
400
  "eval_weighted avg": {
401
+ "f1-score": 0.9437458232244597,
402
+ "precision": 0.9437490553802075,
403
+ "recall": 0.9437705927079184,
404
+ "support": 27619.0
405
  },
406
  "step": 810
407
  },
408
  {
409
  "epoch": 11.0,
410
  "eval_B": {
411
+ "f1-score": 0.8830917874396135,
412
+ "precision": 0.8899707887049659,
413
+ "recall": 0.8763183125599233,
414
+ "support": 1043.0
415
  },
416
  "eval_I": {
417
+ "f1-score": 0.9574634882843123,
418
+ "precision": 0.9647182727751448,
419
+ "recall": 0.9503170028818444,
420
+ "support": 17350.0
421
  },
422
  "eval_O": {
423
+ "f1-score": 0.9200619426496501,
424
+ "precision": 0.9067466582465004,
425
+ "recall": 0.9337741166269239,
426
+ "support": 9226.0
427
  },
428
+ "eval_accuracy": 0.9419964517180202,
429
+ "eval_loss": 0.3198166787624359,
430
  "eval_macro avg": {
431
+ "f1-score": 0.9202057394578587,
432
+ "precision": 0.9204785732422037,
433
+ "recall": 0.9201364773562305,
434
+ "support": 27619.0
435
+ },
436
+ "eval_runtime": 4.6345,
437
+ "eval_samples_per_second": 17.262,
438
+ "eval_steps_per_second": 2.158,
439
  "eval_weighted avg": {
440
+ "f1-score": 0.942161111514465,
441
+ "precision": 0.9425303680165918,
442
+ "recall": 0.9419964517180202,
443
+ "support": 27619.0
444
  },
445
  "step": 891
446
  },
447
  {
448
  "epoch": 12.0,
449
  "eval_B": {
450
+ "f1-score": 0.871306005719733,
451
+ "precision": 0.8663507109004739,
452
+ "recall": 0.8763183125599233,
453
+ "support": 1043.0
454
  },
455
  "eval_I": {
456
+ "f1-score": 0.9576855173984852,
457
+ "precision": 0.9464205312922107,
458
+ "recall": 0.9692219020172911,
459
+ "support": 17350.0
460
  },
461
  "eval_O": {
462
+ "f1-score": 0.9162135168127844,
463
+ "precision": 0.9386084583901774,
464
+ "recall": 0.8948623455451984,
465
+ "support": 9226.0
466
  },
467
+ "eval_accuracy": 0.9408740359897172,
468
+ "eval_loss": 0.35264310240745544,
469
  "eval_macro avg": {
470
+ "f1-score": 0.9150683466436677,
471
+ "precision": 0.917126566860954,
472
+ "recall": 0.9134675200408043,
473
+ "support": 27619.0
474
+ },
475
+ "eval_runtime": 4.6326,
476
+ "eval_samples_per_second": 17.269,
477
+ "eval_steps_per_second": 2.159,
478
  "eval_weighted avg": {
479
+ "f1-score": 0.9405699625961891,
480
+ "precision": 0.9407871989028143,
481
+ "recall": 0.9408740359897172,
482
+ "support": 27619.0
483
  },
484
  "step": 972
485
  },
486
  {
487
  "epoch": 12.35,
488
+ "grad_norm": 0.3602512776851654,
489
  "learning_rate": 1.506172839506173e-05,
490
+ "loss": 0.021,
491
  "step": 1000
492
  },
493
  {
494
  "epoch": 13.0,
495
  "eval_B": {
496
+ "f1-score": 0.877092300334768,
497
+ "precision": 0.875,
498
+ "recall": 0.8791946308724832,
499
+ "support": 1043.0
500
  },
501
  "eval_I": {
502
+ "f1-score": 0.9597242793665263,
503
+ "precision": 0.9486991778353419,
504
+ "recall": 0.9710086455331413,
505
+ "support": 17350.0
506
  },
507
  "eval_O": {
508
+ "f1-score": 0.9203392649259936,
509
+ "precision": 0.9419040054464994,
510
+ "recall": 0.8997398655972252,
511
+ "support": 9226.0
512
  },
513
+ "eval_accuracy": 0.943734385748941,
514
+ "eval_loss": 0.3594338297843933,
515
  "eval_macro avg": {
516
+ "f1-score": 0.9190519482090961,
517
+ "precision": 0.9218677277606138,
518
+ "recall": 0.9166477140009498,
519
+ "support": 27619.0
520
+ },
521
+ "eval_runtime": 4.6042,
522
+ "eval_samples_per_second": 17.376,
523
+ "eval_steps_per_second": 2.172,
524
  "eval_weighted avg": {
525
+ "f1-score": 0.943447393984779,
526
+ "precision": 0.9436461164304495,
527
+ "recall": 0.943734385748941,
528
+ "support": 27619.0
529
  },
530
  "step": 1053
531
  },
532
  {
533
  "epoch": 14.0,
534
  "eval_B": {
535
+ "f1-score": 0.8754052802223252,
536
+ "precision": 0.8467741935483871,
537
+ "recall": 0.9060402684563759,
538
+ "support": 1043.0
539
  },
540
  "eval_I": {
541
+ "f1-score": 0.9574973153224439,
542
+ "precision": 0.9392880904856953,
543
+ "recall": 0.9764265129682997,
544
+ "support": 17350.0
545
  },
546
  "eval_O": {
547
+ "f1-score": 0.9130164471825015,
548
+ "precision": 0.9539388213062477,
549
+ "recall": 0.8754606546715803,
550
+ "support": 9226.0
551
  },
552
+ "eval_accuracy": 0.9400412759332344,
553
+ "eval_loss": 0.3840242028236389,
554
  "eval_macro avg": {
555
+ "f1-score": 0.9153063475757568,
556
+ "precision": 0.9133337017801101,
557
+ "recall": 0.9193091453654186,
558
+ "support": 27619.0
559
+ },
560
+ "eval_runtime": 4.6182,
561
+ "eval_samples_per_second": 17.323,
562
+ "eval_steps_per_second": 2.165,
563
  "eval_weighted avg": {
564
+ "f1-score": 0.9395385738014427,
565
+ "precision": 0.9406884180878825,
566
+ "recall": 0.9400412759332344,
567
+ "support": 27619.0
568
  },
569
  "step": 1134
570
  },
571
  {
572
  "epoch": 15.0,
573
  "eval_B": {
574
+ "f1-score": 0.8797709923664121,
575
+ "precision": 0.8755935422602089,
576
+ "recall": 0.8839884947267498,
577
+ "support": 1043.0
578
  },
579
  "eval_I": {
580
+ "f1-score": 0.9603688219460511,
581
+ "precision": 0.954302299112224,
582
+ "recall": 0.9665129682997118,
583
+ "support": 17350.0
584
  },
585
  "eval_O": {
586
+ "f1-score": 0.922722283205269,
587
+ "precision": 0.9346230820547031,
588
+ "recall": 0.9111207457186213,
589
+ "support": 9226.0
590
  },
591
+ "eval_accuracy": 0.9448930084362215,
592
+ "eval_loss": 0.3762665390968323,
593
  "eval_macro avg": {
594
+ "f1-score": 0.9209540325059108,
595
+ "precision": 0.9215063078090453,
596
+ "recall": 0.920540736248361,
597
+ "support": 27619.0
598
+ },
599
+ "eval_runtime": 4.586,
600
+ "eval_samples_per_second": 17.444,
601
+ "eval_steps_per_second": 2.181,
602
  "eval_weighted avg": {
603
+ "f1-score": 0.944749483712443,
604
+ "precision": 0.9447562007752335,
605
+ "recall": 0.9448930084362215,
606
+ "support": 27619.0
607
  },
608
  "step": 1215
609
  },
610
  {
611
  "epoch": 16.0,
612
  "eval_B": {
613
+ "f1-score": 0.8858669256920835,
614
+ "precision": 0.8976377952755905,
615
+ "recall": 0.8744007670182167,
616
+ "support": 1043.0
617
  },
618
  "eval_I": {
619
+ "f1-score": 0.9586212929752427,
620
+ "precision": 0.9701351590627397,
621
+ "recall": 0.9473775216138328,
622
+ "support": 17350.0
623
  },
624
  "eval_O": {
625
+ "f1-score": 0.9235412474849094,
626
+ "precision": 0.9027950310559006,
627
+ "recall": 0.9452633860828095,
628
+ "support": 9226.0
629
  },
630
+ "eval_accuracy": 0.9439154205438285,
631
+ "eval_loss": 0.36898088455200195,
632
  "eval_macro avg": {
633
+ "f1-score": 0.9226764887174118,
634
+ "precision": 0.923522661798077,
635
+ "recall": 0.922347224904953,
636
+ "support": 27619.0
637
+ },
638
+ "eval_runtime": 4.6024,
639
+ "eval_samples_per_second": 17.382,
640
+ "eval_steps_per_second": 2.173,
641
  "eval_weighted avg": {
642
+ "f1-score": 0.9441554794131966,
643
+ "precision": 0.9449027186622512,
644
+ "recall": 0.9439154205438285,
645
+ "support": 27619.0
646
  },
647
  "step": 1296
648
  },
649
  {
650
  "epoch": 17.0,
651
  "eval_B": {
652
+ "f1-score": 0.8863309352517986,
653
+ "precision": 0.8867562380038387,
654
+ "recall": 0.8859060402684564,
655
+ "support": 1043.0
656
  },
657
  "eval_I": {
658
+ "f1-score": 0.9630614486386496,
659
+ "precision": 0.9659090909090909,
660
+ "recall": 0.9602305475504322,
661
+ "support": 17350.0
662
  },
663
  "eval_O": {
664
+ "f1-score": 0.9296685529506872,
665
+ "precision": 0.9245363918962375,
666
+ "recall": 0.9348580099718188,
667
+ "support": 9226.0
668
  },
669
+ "eval_accuracy": 0.9489481878417032,
670
+ "eval_loss": 0.3569599986076355,
671
  "eval_macro avg": {
672
+ "f1-score": 0.9263536456137119,
673
+ "precision": 0.9257339069363891,
674
+ "recall": 0.9269981992635691,
675
+ "support": 27619.0
676
+ },
677
+ "eval_runtime": 4.6143,
678
+ "eval_samples_per_second": 17.337,
679
+ "eval_steps_per_second": 2.167,
680
  "eval_weighted avg": {
681
+ "f1-score": 0.9490090650954502,
682
+ "precision": 0.9490996138580476,
683
+ "recall": 0.9489481878417032,
684
+ "support": 27619.0
685
  },
686
  "step": 1377
687
  },
688
  {
689
  "epoch": 18.0,
690
  "eval_B": {
691
+ "f1-score": 0.8778409090909091,
692
+ "precision": 0.8671655753040225,
693
+ "recall": 0.8887823585810163,
694
+ "support": 1043.0
695
  },
696
  "eval_I": {
697
+ "f1-score": 0.9557028055829692,
698
+ "precision": 0.9355230471984544,
699
+ "recall": 0.9767723342939482,
700
+ "support": 17350.0
701
  },
702
  "eval_O": {
703
+ "f1-score": 0.9098012570069645,
704
+ "precision": 0.9524599881446354,
705
+ "recall": 0.8707999132885325,
706
+ "support": 9226.0
707
  },
708
+ "eval_accuracy": 0.938049893189471,
709
+ "eval_loss": 0.4151785373687744,
710
  "eval_macro avg": {
711
+ "f1-score": 0.9144483238936143,
712
+ "precision": 0.9183828702157042,
713
+ "recall": 0.9121182020544989,
714
+ "support": 27619.0
715
+ },
716
+ "eval_runtime": 4.6304,
717
+ "eval_samples_per_second": 17.277,
718
+ "eval_steps_per_second": 2.16,
719
  "eval_weighted avg": {
720
+ "f1-score": 0.9374292386470396,
721
+ "precision": 0.9385993125948691,
722
+ "recall": 0.938049893189471,
723
+ "support": 27619.0
724
  },
725
  "step": 1458
726
  },
727
  {
728
  "epoch": 18.52,
729
+ "grad_norm": 10.955327033996582,
730
  "learning_rate": 1.2592592592592593e-05,
731
+ "loss": 0.0058,
732
  "step": 1500
733
  },
734
  {
735
  "epoch": 19.0,
736
  "eval_B": {
737
+ "f1-score": 0.8850241545893719,
738
+ "precision": 0.8919182083739046,
739
+ "recall": 0.87823585810163,
740
+ "support": 1043.0
741
  },
742
  "eval_I": {
743
+ "f1-score": 0.9626489459211732,
744
+ "precision": 0.956838628857761,
745
+ "recall": 0.9685302593659942,
746
+ "support": 17350.0
747
  },
748
  "eval_O": {
749
+ "f1-score": 0.927475898334794,
750
+ "precision": 0.9375415282392027,
751
+ "recall": 0.9176241057879905,
752
+ "support": 9226.0
753
  },
754
+ "eval_accuracy": 0.9481154277852203,
755
+ "eval_loss": 0.3719603419303894,
756
  "eval_macro avg": {
757
+ "f1-score": 0.9250496662817796,
758
+ "precision": 0.9287661218236227,
759
+ "recall": 0.9214634077518715,
760
+ "support": 27619.0
761
+ },
762
+ "eval_runtime": 4.6041,
763
+ "eval_samples_per_second": 17.376,
764
+ "eval_steps_per_second": 2.172,
765
  "eval_weighted avg": {
766
+ "f1-score": 0.9479681394332119,
767
+ "precision": 0.9479408755404257,
768
+ "recall": 0.9481154277852203,
769
+ "support": 27619.0
770
  },
771
  "step": 1539
772
  },
773
  {
774
  "epoch": 20.0,
775
  "eval_B": {
776
+ "f1-score": 0.8742857142857142,
777
+ "precision": 0.8684957426679281,
778
+ "recall": 0.8801534036433365,
779
+ "support": 1043.0
780
  },
781
  "eval_I": {
782
+ "f1-score": 0.9602204921293401,
783
+ "precision": 0.9469289396996189,
784
+ "recall": 0.9738904899135447,
785
+ "support": 17350.0
786
  },
787
  "eval_O": {
788
+ "f1-score": 0.9205305394560855,
789
+ "precision": 0.9473503097040605,
790
+ "recall": 0.8951875135486668,
791
+ "support": 9226.0
792
  },
793
+ "eval_accuracy": 0.9440602483797386,
794
+ "eval_loss": 0.36765894293785095,
795
  "eval_macro avg": {
796
+ "f1-score": 0.9183455819570466,
797
+ "precision": 0.9209249973572025,
798
+ "recall": 0.9164104690351826,
799
+ "support": 27619.0
800
+ },
801
+ "eval_runtime": 4.6404,
802
+ "eval_samples_per_second": 17.24,
803
+ "eval_steps_per_second": 2.155,
804
  "eval_weighted avg": {
805
+ "f1-score": 0.9437170171065533,
806
+ "precision": 0.9441077562808465,
807
+ "recall": 0.9440602483797386,
808
+ "support": 27619.0
809
  },
810
  "step": 1620
811
  },
812
  {
813
  "epoch": 21.0,
814
  "eval_B": {
815
+ "f1-score": 0.8885767790262172,
816
+ "precision": 0.868252516010979,
817
+ "recall": 0.909875359539789,
818
+ "support": 1043.0
819
  },
820
  "eval_I": {
821
+ "f1-score": 0.9616143138880155,
822
+ "precision": 0.9537903271531439,
823
+ "recall": 0.9695677233429395,
824
+ "support": 17350.0
825
  },
826
  "eval_O": {
827
+ "f1-score": 0.9245376759591498,
828
+ "precision": 0.9420632242096973,
829
+ "recall": 0.9076522870149577,
830
+ "support": 9226.0
831
  },
832
+ "eval_accuracy": 0.9466309424671422,
833
+ "eval_loss": 0.374899685382843,
834
  "eval_macro avg": {
835
+ "f1-score": 0.9249095896244609,
836
+ "precision": 0.9213686891246068,
837
+ "recall": 0.9290317899658954,
838
+ "support": 27619.0
839
+ },
840
+ "eval_runtime": 4.5979,
841
+ "eval_samples_per_second": 17.399,
842
+ "eval_steps_per_second": 2.175,
843
  "eval_weighted avg": {
844
+ "f1-score": 0.9464708542988716,
845
+ "precision": 0.9466427045463329,
846
+ "recall": 0.9466309424671422,
847
+ "support": 27619.0
848
  },
849
  "step": 1701
850
  },
851
  {
852
  "epoch": 22.0,
853
  "eval_B": {
854
+ "f1-score": 0.875898418782942,
855
+ "precision": 0.8754789272030651,
856
+ "recall": 0.8763183125599233,
857
+ "support": 1043.0
858
  },
859
  "eval_I": {
860
+ "f1-score": 0.9606979453806588,
861
+ "precision": 0.9475306911822412,
862
+ "recall": 0.9742363112391931,
863
+ "support": 17350.0
864
  },
865
  "eval_O": {
866
+ "f1-score": 0.9217236387930074,
867
+ "precision": 0.94757326007326,
868
+ "recall": 0.897246910903967,
869
+ "support": 9226.0
870
  },
871
+ "eval_accuracy": 0.9448205945182664,
872
+ "eval_loss": 0.3899074196815491,
873
  "eval_macro avg": {
874
+ "f1-score": 0.919440000985536,
875
+ "precision": 0.9235276261528554,
876
+ "recall": 0.9159338449010278,
877
+ "support": 27619.0
878
+ },
879
+ "eval_runtime": 4.5958,
880
+ "eval_samples_per_second": 17.407,
881
+ "eval_steps_per_second": 2.176,
882
  "eval_weighted avg": {
883
+ "f1-score": 0.9444764001104068,
884
+ "precision": 0.9448239585256736,
885
+ "recall": 0.9448205945182664,
886
+ "support": 27619.0
887
  },
888
  "step": 1782
889
  },
890
  {
891
  "epoch": 23.0,
892
  "eval_B": {
893
+ "f1-score": 0.8766079085278704,
894
+ "precision": 0.8712121212121212,
895
+ "recall": 0.8820709491850431,
896
+ "support": 1043.0
897
  },
898
  "eval_I": {
899
+ "f1-score": 0.9571775605897973,
900
+ "precision": 0.9385663638378019,
901
+ "recall": 0.9765417867435159,
902
+ "support": 17350.0
903
  },
904
  "eval_O": {
905
+ "f1-score": 0.9132322264193494,
906
+ "precision": 0.9515920573375631,
907
+ "recall": 0.877845220030349,
908
+ "support": 9226.0
909
  },
910
+ "eval_accuracy": 0.9400050689742568,
911
+ "eval_loss": 0.4506397247314453,
912
  "eval_macro avg": {
913
+ "f1-score": 0.9156725651790056,
914
+ "precision": 0.9204568474624955,
915
+ "recall": 0.9121526519863027,
916
+ "support": 27619.0
917
+ },
918
+ "eval_runtime": 4.6193,
919
+ "eval_samples_per_second": 17.319,
920
+ "eval_steps_per_second": 2.165,
921
  "eval_weighted avg": {
922
+ "f1-score": 0.939455202786939,
923
+ "precision": 0.9403739808105457,
924
+ "recall": 0.9400050689742568,
925
+ "support": 27619.0
926
  },
927
  "step": 1863
928
  },
929
  {
930
  "epoch": 24.0,
931
  "eval_B": {
932
+ "f1-score": 0.8786853552440792,
933
+ "precision": 0.8859649122807017,
934
+ "recall": 0.8715244487056567,
935
+ "support": 1043.0
936
  },
937
  "eval_I": {
938
+ "f1-score": 0.9558590028580501,
939
+ "precision": 0.9575452599919023,
940
+ "recall": 0.954178674351585,
941
+ "support": 17350.0
942
  },
943
  "eval_O": {
944
+ "f1-score": 0.9160280626011873,
945
+ "precision": 0.9121883061049011,
946
+ "recall": 0.9199002818122697,
947
+ "support": 9226.0
948
  },
949
+ "eval_accuracy": 0.9396067924255042,
950
+ "eval_loss": 0.42392534017562866,
951
  "eval_macro avg": {
952
+ "f1-score": 0.9168574735677723,
953
+ "precision": 0.9185661594591683,
954
+ "recall": 0.9152011349565038,
955
+ "support": 27619.0
956
+ },
957
+ "eval_runtime": 4.6242,
958
+ "eval_samples_per_second": 17.3,
959
+ "eval_steps_per_second": 2.163,
960
  "eval_weighted avg": {
961
+ "f1-score": 0.9396392856607877,
962
+ "precision": 0.9396908279261412,
963
+ "recall": 0.9396067924255042,
964
+ "support": 27619.0
965
  },
966
  "step": 1944
967
  },
968
  {
969
  "epoch": 24.69,
970
+ "grad_norm": 0.030169092118740082,
971
  "learning_rate": 1.0123456790123458e-05,
972
+ "loss": 0.0026,
973
  "step": 2000
974
  },
975
  {
976
  "epoch": 25.0,
977
  "eval_B": {
978
+ "f1-score": 0.8887832699619772,
979
+ "precision": 0.88124410933082,
980
+ "recall": 0.8964525407478428,
981
+ "support": 1043.0
982
  },
983
  "eval_I": {
984
+ "f1-score": 0.9607269808508199,
985
+ "precision": 0.9556860955857192,
986
+ "recall": 0.9658213256484149,
987
+ "support": 17350.0
988
  },
989
  "eval_O": {
990
+ "f1-score": 0.9237260273972602,
991
+ "precision": 0.9340647163120568,
992
+ "recall": 0.9136137004118795,
993
+ "support": 9226.0
994
  },
995
+ "eval_accuracy": 0.9457619754516818,
996
+ "eval_loss": 0.398232638835907,
997
  "eval_macro avg": {
998
+ "f1-score": 0.9244120927366858,
999
+ "precision": 0.9236649737428652,
1000
+ "recall": 0.9252958556027124,
1001
+ "support": 27619.0
1002
+ },
1003
+ "eval_runtime": 4.5953,
1004
+ "eval_samples_per_second": 17.409,
1005
+ "eval_steps_per_second": 2.176,
1006
  "eval_weighted avg": {
1007
+ "f1-score": 0.9456501103261954,
1008
+ "precision": 0.9456523566073828,
1009
+ "recall": 0.9457619754516818,
1010
+ "support": 27619.0
1011
  },
1012
  "step": 2025
1013
  },
1014
  {
1015
  "epoch": 26.0,
1016
  "eval_B": {
1017
+ "f1-score": 0.8827717133364974,
1018
+ "precision": 0.8740601503759399,
1019
+ "recall": 0.8916586768935763,
1020
+ "support": 1043.0
1021
  },
1022
  "eval_I": {
1023
+ "f1-score": 0.9590283123770422,
1024
+ "precision": 0.9489364103142809,
1025
+ "recall": 0.9693371757925072,
1026
+ "support": 17350.0
1027
  },
1028
  "eval_O": {
1029
+ "f1-score": 0.9188171447557869,
1030
+ "precision": 0.9393115942028986,
1031
+ "recall": 0.8991979189247779,
1032
+ "support": 9226.0
1033
  },
1034
+ "eval_accuracy": 0.9429740396104132,
1035
+ "eval_loss": 0.4066696763038635,
1036
  "eval_macro avg": {
1037
+ "f1-score": 0.9202057234897755,
1038
+ "precision": 0.9207693849643731,
1039
+ "recall": 0.9200645905369539,
1040
+ "support": 27619.0
1041
+ },
1042
+ "eval_runtime": 4.6361,
1043
+ "eval_samples_per_second": 17.256,
1044
+ "eval_steps_per_second": 2.157,
1045
  "eval_weighted avg": {
1046
+ "f1-score": 0.9427162132687112,
1047
+ "precision": 0.942893668268613,
1048
+ "recall": 0.9429740396104132,
1049
+ "support": 27619.0
1050
  },
1051
  "step": 2106
1052
  },
1053
  {
1054
  "epoch": 27.0,
1055
  "eval_B": {
1056
+ "f1-score": 0.8731307284129282,
1057
+ "precision": 0.8786407766990292,
1058
+ "recall": 0.8676893576222435,
1059
+ "support": 1043.0
1060
  },
1061
  "eval_I": {
1062
+ "f1-score": 0.9568967963451661,
1063
+ "precision": 0.9424291543234028,
1064
+ "recall": 0.9718155619596541,
1065
+ "support": 17350.0
1066
  },
1067
  "eval_O": {
1068
+ "f1-score": 0.9140816781968311,
1069
+ "precision": 0.9418257070590941,
1070
+ "recall": 0.8879254281378712,
1071
+ "support": 9226.0
1072
  },
1073
+ "eval_accuracy": 0.9398602411383468,
1074
+ "eval_loss": 0.44146138429641724,
1075
  "eval_macro avg": {
1076
+ "f1-score": 0.9147030676516418,
1077
+ "precision": 0.920965212693842,
1078
+ "recall": 0.9091434492399229,
1079
+ "support": 27619.0
1080
+ },
1081
+ "eval_runtime": 4.6584,
1082
+ "eval_samples_per_second": 17.173,
1083
+ "eval_steps_per_second": 2.147,
1084
  "eval_weighted avg": {
1085
+ "f1-score": 0.9394312730137688,
1086
+ "precision": 0.9398186802902107,
1087
+ "recall": 0.9398602411383468,
1088
+ "support": 27619.0
1089
  },
1090
  "step": 2187
1091
  },
1092
  {
1093
  "epoch": 28.0,
1094
  "eval_B": {
1095
+ "f1-score": 0.8846153846153846,
1096
+ "precision": 0.8871745419479267,
1097
+ "recall": 0.8820709491850431,
1098
+ "support": 1043.0
1099
  },
1100
  "eval_I": {
1101
+ "f1-score": 0.9619680394066098,
1102
+ "precision": 0.9559995446265938,
1103
+ "recall": 0.9680115273775216,
1104
+ "support": 17350.0
1105
  },
1106
  "eval_O": {
1107
+ "f1-score": 0.9258771929824562,
1108
+ "precision": 0.9367650321721767,
1109
+ "recall": 0.9152395404292217,
1110
+ "support": 9226.0
1111
  },
1112
+ "eval_accuracy": 0.9471378398928274,
1113
+ "eval_loss": 0.4067588746547699,
1114
  "eval_macro avg": {
1115
+ "f1-score": 0.9241535390014834,
1116
+ "precision": 0.9266463729155657,
1117
+ "recall": 0.9217740056639289,
1118
+ "support": 27619.0
1119
+ },
1120
+ "eval_runtime": 4.6224,
1121
+ "eval_samples_per_second": 17.307,
1122
+ "eval_steps_per_second": 2.163,
1123
  "eval_weighted avg": {
1124
+ "f1-score": 0.9469909233612611,
1125
+ "precision": 0.9469752465094172,
1126
+ "recall": 0.9471378398928274,
1127
+ "support": 27619.0
1128
  },
1129
  "step": 2268
1130
  },
1131
  {
1132
  "epoch": 29.0,
1133
  "eval_B": {
1134
+ "f1-score": 0.8919301557338367,
1135
+ "precision": 0.8782527881040892,
1136
+ "recall": 0.9060402684563759,
1137
+ "support": 1043.0
1138
  },
1139
  "eval_I": {
1140
+ "f1-score": 0.964135443798838,
1141
+ "precision": 0.9576391652925457,
1142
+ "recall": 0.9707204610951009,
1143
+ "support": 17350.0
1144
  },
1145
  "eval_O": {
1146
+ "f1-score": 0.9300406995930042,
1147
+ "precision": 0.944059848146494,
1148
+ "recall": 0.9164318231086062,
1149
+ "support": 9226.0
1150
  },
1151
+ "eval_accuracy": 0.9501430174879612,
1152
+ "eval_loss": 0.4006032943725586,
1153
  "eval_macro avg": {
1154
+ "f1-score": 0.9287020997085597,
1155
+ "precision": 0.9266506005143763,
1156
+ "recall": 0.9310641842200277,
1157
+ "support": 27619.0
1158
+ },
1159
+ "eval_runtime": 4.5962,
1160
+ "eval_samples_per_second": 17.406,
1161
+ "eval_steps_per_second": 2.176,
1162
  "eval_weighted avg": {
1163
+ "f1-score": 0.9500195009517102,
1164
+ "precision": 0.9501051209246456,
1165
+ "recall": 0.9501430174879612,
1166
+ "support": 27619.0
1167
  },
1168
  "step": 2349
1169
  },
1170
  {
1171
  "epoch": 30.0,
1172
  "eval_B": {
1173
+ "f1-score": 0.8796956728483118,
1174
+ "precision": 0.8726415094339622,
1175
+ "recall": 0.8868648130393096,
1176
+ "support": 1043.0
1177
  },
1178
  "eval_I": {
1179
+ "f1-score": 0.9600954979536152,
1180
+ "precision": 0.947067399349557,
1181
+ "recall": 0.9734870317002882,
1182
+ "support": 17350.0
1183
  },
1184
  "eval_O": {
1185
+ "f1-score": 0.9212857222438862,
1186
+ "precision": 0.9477363896848138,
1187
+ "recall": 0.8962714068935617,
1188
+ "support": 9226.0
1189
  },
1190
+ "eval_accuracy": 0.9444223179695137,
1191
+ "eval_loss": 0.46608424186706543,
1192
  "eval_macro avg": {
1193
+ "f1-score": 0.9203589643486044,
1194
+ "precision": 0.922481766156111,
1195
+ "recall": 0.9188744172110531,
1196
+ "support": 27619.0
1197
+ },
1198
+ "eval_runtime": 4.5968,
1199
+ "eval_samples_per_second": 17.403,
1200
+ "eval_steps_per_second": 2.175,
1201
  "eval_weighted avg": {
1202
+ "f1-score": 0.9440950631702127,
1203
+ "precision": 0.9444802637418637,
1204
+ "recall": 0.9444223179695137,
1205
+ "support": 27619.0
1206
  },
1207
  "step": 2430
1208
  },
1209
  {
1210
  "epoch": 30.86,
1211
+ "grad_norm": 0.0009192481520585716,
1212
  "learning_rate": 7.654320987654322e-06,
1213
+ "loss": 0.0015,
1214
  "step": 2500
1215
  },
1216
  {
1217
  "epoch": 31.0,
1218
  "eval_B": {
1219
+ "f1-score": 0.8760724499523355,
1220
+ "precision": 0.8710900473933649,
1221
+ "recall": 0.8811121764141898,
1222
+ "support": 1043.0
1223
  },
1224
  "eval_I": {
1225
+ "f1-score": 0.9596285601041844,
1226
+ "precision": 0.9430224794124193,
1227
+ "recall": 0.9768299711815562,
1228
+ "support": 17350.0
1229
  },
1230
  "eval_O": {
1231
+ "f1-score": 0.9194073408912335,
1232
+ "precision": 0.953328677839851,
1233
+ "recall": 0.8878170388033817,
1234
+ "support": 9226.0
1235
  },
1236
+ "eval_accuracy": 0.9434809370360984,
1237
+ "eval_loss": 0.48370352387428284,
1238
  "eval_macro avg": {
1239
+ "f1-score": 0.9183694503159178,
1240
+ "precision": 0.9224804015485452,
1241
+ "recall": 0.9152530621330426,
1242
+ "support": 27619.0
1243
+ },
1244
+ "eval_runtime": 4.6119,
1245
+ "eval_samples_per_second": 17.347,
1246
+ "eval_steps_per_second": 2.168,
1247
  "eval_weighted avg": {
1248
+ "f1-score": 0.943037445605214,
1249
+ "precision": 0.9437487714612122,
1250
+ "recall": 0.9434809370360984,
1251
+ "support": 27619.0
1252
  },
1253
  "step": 2511
1254
  },
1255
  {
1256
  "epoch": 32.0,
1257
  "eval_B": {
1258
+ "f1-score": 0.8770764119601329,
1259
+ "precision": 0.868421052631579,
1260
+ "recall": 0.8859060402684564,
1261
+ "support": 1043.0
1262
  },
1263
  "eval_I": {
1264
+ "f1-score": 0.9595270577535243,
1265
+ "precision": 0.9465066726477515,
1266
+ "recall": 0.9729106628242075,
1267
+ "support": 17350.0
1268
  },
1269
  "eval_O": {
1270
+ "f1-score": 0.919819468434836,
1271
+ "precision": 0.9464510950579063,
1272
+ "recall": 0.8946455668762194,
1273
+ "support": 9226.0
1274
  },
1275
+ "eval_accuracy": 0.9434809370360984,
1276
+ "eval_loss": 0.4701555371284485,
1277
  "eval_macro avg": {
1278
+ "f1-score": 0.9188076460494976,
1279
+ "precision": 0.9204596067790788,
1280
+ "recall": 0.9178207566562945,
1281
+ "support": 27619.0
1282
+ },
1283
+ "eval_runtime": 4.6158,
1284
+ "eval_samples_per_second": 17.332,
1285
+ "eval_steps_per_second": 2.166,
1286
  "eval_weighted avg": {
1287
+ "f1-score": 0.943149265559139,
1288
+ "precision": 0.9435392929265168,
1289
+ "recall": 0.9434809370360984,
1290
+ "support": 27619.0
1291
  },
1292
  "step": 2592
1293
  },
1294
  {
1295
  "epoch": 33.0,
1296
  "eval_B": {
1297
+ "f1-score": 0.8849976336961666,
1298
+ "precision": 0.8738317757009346,
1299
+ "recall": 0.8964525407478428,
1300
+ "support": 1043.0
1301
  },
1302
  "eval_I": {
1303
+ "f1-score": 0.9615693472524951,
1304
+ "precision": 0.9569585569128896,
1305
+ "recall": 0.9662247838616714,
1306
+ "support": 17350.0
1307
  },
1308
  "eval_O": {
1309
+ "f1-score": 0.9260009859232076,
1310
+ "precision": 0.9359982283246595,
1311
+ "recall": 0.9162150444396271,
1312
+ "support": 9226.0
1313
  },
1314
+ "eval_accuracy": 0.9468843911799848,
1315
+ "eval_loss": 0.4380284249782562,
1316
  "eval_macro avg": {
1317
+ "f1-score": 0.9241893222906231,
1318
+ "precision": 0.9222628536461612,
1319
+ "recall": 0.9262974563497138,
1320
+ "support": 27619.0
1321
+ },
1322
+ "eval_runtime": 4.5881,
1323
+ "eval_samples_per_second": 17.436,
1324
+ "eval_steps_per_second": 2.18,
1325
  "eval_weighted avg": {
1326
+ "f1-score": 0.9467962563055651,
1327
+ "precision": 0.946817667512148,
1328
+ "recall": 0.9468843911799848,
1329
+ "support": 27619.0
1330
  },
1331
  "step": 2673
1332
  },
1333
  {
1334
  "epoch": 34.0,
1335
  "eval_B": {
1336
+ "f1-score": 0.8833652007648184,
1337
+ "precision": 0.880838894184938,
1338
+ "recall": 0.8859060402684564,
1339
+ "support": 1043.0
1340
  },
1341
  "eval_I": {
1342
+ "f1-score": 0.9618386073235572,
1343
+ "precision": 0.9541201156921681,
1344
+ "recall": 0.9696829971181556,
1345
+ "support": 17350.0
1346
  },
1347
  "eval_O": {
1348
+ "f1-score": 0.9257281286131146,
1349
+ "precision": 0.9406959829920555,
1350
+ "recall": 0.9112291350531108,
1351
+ "support": 9226.0
1352
  },
1353
+ "eval_accuracy": 0.9469930120569173,
1354
+ "eval_loss": 0.4419035315513611,
1355
  "eval_macro avg": {
1356
+ "f1-score": 0.9236439789004968,
1357
+ "precision": 0.9252183309563873,
1358
+ "recall": 0.9222727241465742,
1359
+ "support": 27619.0
1360
+ },
1361
+ "eval_runtime": 4.6449,
1362
+ "eval_samples_per_second": 17.223,
1363
+ "eval_steps_per_second": 2.153,
1364
  "eval_weighted avg": {
1365
+ "f1-score": 0.9468126092923719,
1366
+ "precision": 0.9468684642086502,
1367
+ "recall": 0.9469930120569173,
1368
+ "support": 27619.0
1369
  },
1370
  "step": 2754
1371
  },
1372
  {
1373
  "epoch": 35.0,
1374
  "eval_B": {
1375
+ "f1-score": 0.8785405664906385,
1376
+ "precision": 0.8798076923076923,
1377
+ "recall": 0.8772770853307766,
1378
+ "support": 1043.0
1379
  },
1380
  "eval_I": {
1381
+ "f1-score": 0.9603404255319149,
1382
+ "precision": 0.945586592178771,
1383
+ "recall": 0.9755619596541787,
1384
+ "support": 17350.0
1385
  },
1386
  "eval_O": {
1387
+ "f1-score": 0.9217537000837756,
1388
+ "precision": 0.9508007835004033,
1389
+ "recall": 0.8944287882072404,
1390
+ "support": 9226.0
1391
  },
1392
+ "eval_accuracy": 0.9447481806003114,
1393
+ "eval_loss": 0.4607029855251312,
1394
  "eval_macro avg": {
1395
+ "f1-score": 0.920211564035443,
1396
+ "precision": 0.9253983559956221,
1397
+ "recall": 0.9157559443973985,
1398
+ "support": 27619.0
1399
+ },
1400
+ "eval_runtime": 4.595,
1401
+ "eval_samples_per_second": 17.41,
1402
+ "eval_steps_per_second": 2.176,
1403
  "eval_weighted avg": {
1404
+ "f1-score": 0.9443616289800995,
1405
+ "precision": 0.9448443037746955,
1406
+ "recall": 0.9447481806003114,
1407
+ "support": 27619.0
1408
  },
1409
  "step": 2835
1410
  },
1411
  {
1412
  "epoch": 36.0,
1413
  "eval_B": {
1414
+ "f1-score": 0.8802714493456132,
1415
+ "precision": 0.8901960784313725,
1416
+ "recall": 0.8705656759348035,
1417
+ "support": 1043.0
1418
  },
1419
  "eval_I": {
1420
+ "f1-score": 0.9600229918091681,
1421
+ "precision": 0.9574089997133849,
1422
+ "recall": 0.9626512968299712,
1423
+ "support": 17350.0
1424
  },
1425
  "eval_O": {
1426
+ "f1-score": 0.9233949945593034,
1427
+ "precision": 0.9270264365304784,
1428
+ "recall": 0.9197918924777801,
1429
+ "support": 9226.0
1430
  },
1431
+ "eval_accuracy": 0.9448568014772439,
1432
+ "eval_loss": 0.44131579995155334,
1433
  "eval_macro avg": {
1434
+ "f1-score": 0.9212298119046949,
1435
+ "precision": 0.924877171558412,
1436
+ "recall": 0.9176696217475183,
1437
+ "support": 27619.0
1438
+ },
1439
+ "eval_runtime": 4.6044,
1440
+ "eval_samples_per_second": 17.375,
1441
+ "eval_steps_per_second": 2.172,
1442
  "eval_weighted avg": {
1443
+ "f1-score": 0.944775851745562,
1444
+ "precision": 0.9447216249053674,
1445
+ "recall": 0.9448568014772439,
1446
+ "support": 27619.0
1447
  },
1448
  "step": 2916
1449
  },
1450
  {
1451
  "epoch": 37.0,
1452
  "eval_B": {
1453
+ "f1-score": 0.8851513695338779,
1454
+ "precision": 0.8872832369942196,
1455
+ "recall": 0.8830297219558965,
1456
+ "support": 1043.0
1457
  },
1458
  "eval_I": {
1459
+ "f1-score": 0.9621385956860945,
1460
+ "precision": 0.9576338928856915,
1461
+ "recall": 0.966685878962536,
1462
+ "support": 17350.0
1463
  },
1464
  "eval_O": {
1465
+ "f1-score": 0.9272399278412508,
1466
+ "precision": 0.9353700231609132,
1467
+ "recall": 0.9192499458053327,
1468
+ "support": 9226.0
1469
  },
1470
+ "eval_accuracy": 0.9476809442774902,
1471
+ "eval_loss": 0.4391210675239563,
1472
  "eval_macro avg": {
1473
+ "f1-score": 0.9248432976870743,
1474
+ "precision": 0.9267623843469414,
1475
+ "recall": 0.9229885155745885,
1476
+ "support": 27619.0
1477
+ },
1478
+ "eval_runtime": 4.6195,
1479
+ "eval_samples_per_second": 17.318,
1480
+ "eval_steps_per_second": 2.165,
1481
  "eval_weighted avg": {
1482
+ "f1-score": 0.9475735214106575,
1483
+ "precision": 0.9475400373450994,
1484
+ "recall": 0.9476809442774902,
1485
+ "support": 27619.0
1486
  },
1487
  "step": 2997
1488
  },
1489
  {
1490
  "epoch": 37.04,
1491
+ "grad_norm": 0.36456331610679626,
1492
  "learning_rate": 5.185185185185185e-06,
1493
+ "loss": 0.0012,
1494
  "step": 3000
1495
  },
1496
  {
1497
  "epoch": 38.0,
1498
  "eval_B": {
1499
+ "f1-score": 0.8848715509039011,
1500
+ "precision": 0.8781869688385269,
1501
+ "recall": 0.8916586768935763,
1502
+ "support": 1043.0
1503
  },
1504
  "eval_I": {
1505
+ "f1-score": 0.9604875396984349,
1506
+ "precision": 0.9536389977842168,
1507
+ "recall": 0.9674351585014409,
1508
+ "support": 17350.0
1509
  },
1510
  "eval_O": {
1511
+ "f1-score": 0.9237283475391805,
1512
+ "precision": 0.9374930237749749,
1513
+ "recall": 0.9103620203771949,
1514
+ "support": 9226.0
1515
  },
1516
+ "eval_accuracy": 0.9455085267388392,
1517
+ "eval_loss": 0.43638840317726135,
1518
  "eval_macro avg": {
1519
+ "f1-score": 0.9230291460471722,
1520
+ "precision": 0.9231063301325729,
1521
+ "recall": 0.9231519519240706,
1522
+ "support": 27619.0
1523
+ },
1524
+ "eval_runtime": 4.6165,
1525
+ "eval_samples_per_second": 17.329,
1526
+ "eval_steps_per_second": 2.166,
1527
  "eval_weighted avg": {
1528
+ "f1-score": 0.9453527490407726,
1529
+ "precision": 0.9453961496579408,
1530
+ "recall": 0.9455085267388392,
1531
+ "support": 27619.0
1532
  },
1533
  "step": 3078
1534
  },
1535
  {
1536
  "epoch": 39.0,
1537
  "eval_B": {
1538
+ "f1-score": 0.8896321070234113,
1539
+ "precision": 0.8866666666666667,
1540
+ "recall": 0.8926174496644296,
1541
+ "support": 1043.0
1542
  },
1543
  "eval_I": {
1544
+ "f1-score": 0.9625824964131994,
1545
+ "precision": 0.9584571428571429,
1546
+ "recall": 0.9667435158501441,
1547
+ "support": 17350.0
1548
  },
1549
  "eval_O": {
1550
+ "f1-score": 0.9277944793659471,
1551
+ "precision": 0.9358253390671518,
1552
+ "recall": 0.9199002818122697,
1553
+ "support": 9226.0
1554
  },
1555
+ "eval_accuracy": 0.9482964625801079,
1556
+ "eval_loss": 0.44264644384384155,
1557
  "eval_macro avg": {
1558
+ "f1-score": 0.9266696942675193,
1559
+ "precision": 0.9269830495303205,
1560
+ "recall": 0.9264204157756145,
1561
+ "support": 27619.0
1562
+ },
1563
+ "eval_runtime": 4.6185,
1564
+ "eval_samples_per_second": 17.322,
1565
+ "eval_steps_per_second": 2.165,
1566
  "eval_weighted avg": {
1567
+ "f1-score": 0.9482068310592221,
1568
+ "precision": 0.9481860074636412,
1569
+ "recall": 0.9482964625801079,
1570
+ "support": 27619.0
1571
  },
1572
  "step": 3159
1573
  },
1574
  {
1575
  "epoch": 40.0,
1576
  "eval_B": {
1577
+ "f1-score": 0.8865485878410723,
1578
+ "precision": 0.8852772466539197,
1579
+ "recall": 0.887823585810163,
1580
+ "support": 1043.0
1581
  },
1582
  "eval_I": {
1583
+ "f1-score": 0.9624967964233846,
1584
+ "precision": 0.9512016660100185,
1585
+ "recall": 0.9740634005763689,
1586
+ "support": 17350.0
1587
  },
1588
  "eval_O": {
1589
+ "f1-score": 0.9265749778172139,
1590
+ "precision": 0.9486713604360664,
1591
+ "recall": 0.905484500325168,
1592
+ "support": 9226.0
1593
  },
1594
+ "eval_accuracy": 0.9478981860313552,
1595
+ "eval_loss": 0.4583560824394226,
1596
  "eval_macro avg": {
1597
+ "f1-score": 0.925206787360557,
1598
+ "precision": 0.9283834243666682,
1599
+ "recall": 0.9224571622372332,
1600
+ "support": 27619.0
1601
+ },
1602
+ "eval_runtime": 4.643,
1603
+ "eval_samples_per_second": 17.23,
1604
+ "eval_steps_per_second": 2.154,
1605
  "eval_weighted avg": {
1606
+ "f1-score": 0.9476291806512029,
1607
+ "precision": 0.947866868638148,
1608
+ "recall": 0.9478981860313552,
1609
+ "support": 27619.0
1610
  },
1611
  "step": 3240
1612
  }
 
1616
  "num_input_tokens_seen": 0,
1617
  "num_train_epochs": 50,
1618
  "save_steps": 500,
1619
+ "total_flos": 5751466109424000.0,
1620
  "train_batch_size": 4,
1621
  "trial_name": null,
1622
  "trial_params": null
checkpoint-3240/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b8ca363922308e3892528dd031742ffedc372734b464a042d09444af9700901e
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:821ca433db9ddb761291ab61fc06d651ad248cd9d560cdb25f7f905e5aaff14e
3
  size 4856