Theoreticallyhugo commited on
Commit
f3e4fef
1 Parent(s): a561eb0

Training in progress, epoch 27, checkpoint

Browse files
checkpoint-2187/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:939f403f1c95c73f71e3cefda9eab147defe3d5475a242ac36f60dd19e9ed512
3
  size 592324828
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae71c33deca7b1622caae4857ad74bb29cd3ec641b1993050c3d321e2a214828
3
  size 592324828
checkpoint-2187/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e6e60e467becaff12a7d933681d7a8412cb537f58b3f865ba1c5e8a6ab248dcf
3
  size 1014657786
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42ccb4fb9accc0d409c3dc0fbf0940daa3f11ba0176673af438489e181b10b83
3
  size 1014657786
checkpoint-2187/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:86dee12378f9ab7baf053e194b726ea8cd664a9c8eabe6098746f3b686d76808
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db72830e8d87eee81a1ac8d98a638d23ced5ce02b9e8fce99c7bc0afe79cdab2
3
  size 14244
checkpoint-2187/trainer_state.json CHANGED
@@ -11,1243 +11,1243 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_Claim": {
14
- "f1-score": 0.44734338904742493,
15
- "precision": 0.5477537437603993,
16
- "recall": 0.3780431786862655,
17
- "support": 4354.0
18
  },
19
  "eval_MajorClaim": {
20
- "f1-score": 0.6504965622612681,
21
- "precision": 0.5898856944925528,
22
- "recall": 0.7249893571732652,
23
- "support": 2349.0
24
  },
25
  "eval_O": {
26
- "f1-score": 0.8859784283513097,
27
- "precision": 0.8689081979599547,
28
- "recall": 0.9037328094302554,
29
- "support": 10180.0
30
  },
31
  "eval_Premise": {
32
- "f1-score": 0.8720120805863504,
33
- "precision": 0.859258183929738,
34
- "recall": 0.8851502916105877,
35
- "support": 13374.0
36
  },
37
- "eval_accuracy": 0.8059953068711373,
38
- "eval_loss": 0.47828415036201477,
39
  "eval_macro avg": {
40
- "f1-score": 0.7139576150615883,
41
- "precision": 0.7164514550356611,
42
- "recall": 0.7229789092250934,
43
- "support": 30257.0
44
- },
45
- "eval_runtime": 4.8422,
46
- "eval_samples_per_second": 16.728,
47
- "eval_steps_per_second": 2.272,
48
  "eval_weighted avg": {
49
- "f1-score": 0.7984036588902531,
50
- "precision": 0.7967665566250599,
51
- "recall": 0.8059953068711373,
52
- "support": 30257.0
53
  },
54
  "step": 81
55
  },
56
  {
57
  "epoch": 2.0,
58
  "eval_Claim": {
59
- "f1-score": 0.5787561000934482,
60
- "precision": 0.5281409891984082,
61
- "recall": 0.6401010564997703,
62
- "support": 4354.0
63
  },
64
  "eval_MajorClaim": {
65
- "f1-score": 0.7702650787552824,
66
- "precision": 0.7017850892544627,
67
- "recall": 0.8535547041294168,
68
- "support": 2349.0
69
  },
70
  "eval_O": {
71
- "f1-score": 0.9095337508698677,
72
- "precision": 0.9206077681626081,
73
- "recall": 0.8987229862475442,
74
- "support": 10180.0
75
  },
76
  "eval_Premise": {
77
- "f1-score": 0.8736648538675224,
78
- "precision": 0.9162905211325401,
79
- "recall": 0.8348287722446538,
80
- "support": 13374.0
81
  },
82
- "eval_accuracy": 0.8297584030141786,
83
- "eval_loss": 0.4241589605808258,
84
  "eval_macro avg": {
85
- "f1-score": 0.7830549458965301,
86
- "precision": 0.7667060919370048,
87
- "recall": 0.8068018797803463,
88
- "support": 30257.0
89
- },
90
- "eval_runtime": 4.8636,
91
- "eval_samples_per_second": 16.654,
92
- "eval_steps_per_second": 2.262,
93
  "eval_weighted avg": {
94
- "f1-score": 0.8352680063880269,
95
- "precision": 0.8452350051608073,
96
- "recall": 0.8297584030141786,
97
- "support": 30257.0
98
  },
99
  "step": 162
100
  },
101
  {
102
  "epoch": 3.0,
103
  "eval_Claim": {
104
- "f1-score": 0.5728465083303793,
105
- "precision": 0.5899245558530056,
106
- "recall": 0.5567294441892513,
107
- "support": 4354.0
108
  },
109
  "eval_MajorClaim": {
110
- "f1-score": 0.7862876979822088,
111
- "precision": 0.8017699115044248,
112
- "recall": 0.7713920817369093,
113
- "support": 2349.0
114
  },
115
  "eval_O": {
116
- "f1-score": 0.914532821534425,
117
- "precision": 0.9048168445341794,
118
- "recall": 0.9244597249508841,
119
- "support": 10180.0
120
  },
121
  "eval_Premise": {
122
- "f1-score": 0.8880533114924986,
123
- "precision": 0.8843330614665975,
124
- "recall": 0.8918049947659639,
125
- "support": 13374.0
126
  },
127
- "eval_accuracy": 0.8452258981392735,
128
- "eval_loss": 0.41067537665367126,
129
  "eval_macro avg": {
130
- "f1-score": 0.7904300848348779,
131
- "precision": 0.7952110933395518,
132
- "recall": 0.7860965614107521,
133
- "support": 30257.0
134
- },
135
- "eval_runtime": 4.8684,
136
- "eval_samples_per_second": 16.638,
137
- "eval_steps_per_second": 2.259,
138
  "eval_weighted avg": {
139
- "f1-score": 0.8437033615676307,
140
- "precision": 0.8424495118392472,
141
- "recall": 0.8452258981392735,
142
- "support": 30257.0
143
  },
144
  "step": 243
145
  },
146
  {
147
  "epoch": 4.0,
148
  "eval_Claim": {
149
- "f1-score": 0.5865384615384616,
150
- "precision": 0.5714596949891068,
151
- "recall": 0.6024345429490124,
152
- "support": 4354.0
153
  },
154
  "eval_MajorClaim": {
155
- "f1-score": 0.7783630470016208,
156
- "precision": 0.7425589485890993,
157
- "recall": 0.8177948063005535,
158
- "support": 2349.0
159
  },
160
  "eval_O": {
161
- "f1-score": 0.9108591535529275,
162
- "precision": 0.9063411787589963,
163
- "recall": 0.9154223968565816,
164
- "support": 10180.0
165
  },
166
  "eval_Premise": {
167
- "f1-score": 0.8868256151612411,
168
- "precision": 0.9067823097358962,
169
- "recall": 0.8677284282937042,
170
- "support": 13374.0
171
  },
172
- "eval_accuracy": 0.8417225765938461,
173
- "eval_loss": 0.4545813202857971,
174
  "eval_macro avg": {
175
- "f1-score": 0.7906465693135627,
176
- "precision": 0.7817855330182747,
177
- "recall": 0.8008450435999629,
178
- "support": 30257.0
179
- },
180
- "eval_runtime": 4.9112,
181
- "eval_samples_per_second": 16.493,
182
- "eval_steps_per_second": 2.24,
183
  "eval_weighted avg": {
184
- "f1-score": 0.8432797441676475,
185
- "precision": 0.8456313015960876,
186
- "recall": 0.8417225765938461,
187
- "support": 30257.0
188
  },
189
  "step": 324
190
  },
191
  {
192
  "epoch": 5.0,
193
  "eval_Claim": {
194
- "f1-score": 0.597399608790703,
195
- "precision": 0.5985704403965875,
196
- "recall": 0.5962333486449242,
197
- "support": 4354.0
198
  },
199
  "eval_MajorClaim": {
200
- "f1-score": 0.7933532168726033,
201
- "precision": 0.7940298507462686,
202
- "recall": 0.7926777352064709,
203
- "support": 2349.0
204
  },
205
  "eval_O": {
206
- "f1-score": 0.9101112629318758,
207
- "precision": 0.9042862684251358,
208
- "recall": 0.9160117878192534,
209
- "support": 10180.0
210
  },
211
  "eval_Premise": {
212
- "f1-score": 0.8923677591320345,
213
- "precision": 0.8961019377214808,
214
- "recall": 0.8886645730521908,
215
- "support": 13374.0
216
  },
217
- "eval_accuracy": 0.848332617245596,
218
- "eval_loss": 0.48704993724823,
219
  "eval_macro avg": {
220
- "f1-score": 0.7983079619318041,
221
- "precision": 0.7982471243223681,
222
- "recall": 0.7983968611807098,
223
- "support": 30257.0
224
- },
225
- "eval_runtime": 4.8769,
226
- "eval_samples_per_second": 16.609,
227
- "eval_steps_per_second": 2.256,
228
  "eval_weighted avg": {
229
- "f1-score": 0.8482045037639816,
230
- "precision": 0.8481162489521332,
231
- "recall": 0.848332617245596,
232
- "support": 30257.0
233
  },
234
  "step": 405
235
  },
236
  {
237
  "epoch": 6.0,
238
  "eval_Claim": {
239
- "f1-score": 0.6060809300245921,
240
- "precision": 0.5903745644599303,
241
- "recall": 0.6226458429030777,
242
- "support": 4354.0
243
  },
244
  "eval_MajorClaim": {
245
- "f1-score": 0.7858529221479404,
246
- "precision": 0.7963286713286714,
247
- "recall": 0.7756492124308216,
248
- "support": 2349.0
249
  },
250
  "eval_O": {
251
- "f1-score": 0.9135615470059578,
252
- "precision": 0.9398504051527115,
253
- "recall": 0.8887033398821218,
254
- "support": 10180.0
255
  },
256
  "eval_Premise": {
257
- "f1-score": 0.895557603686636,
258
- "precision": 0.88328121591157,
259
- "recall": 0.908180050844923,
260
- "support": 13374.0
261
  },
262
- "eval_accuracy": 0.8502495290346036,
263
- "eval_loss": 0.5423789024353027,
264
  "eval_macro avg": {
265
- "f1-score": 0.8002632507162816,
266
- "precision": 0.8024587142132208,
267
- "recall": 0.7987946115152361,
268
- "support": 30257.0
269
- },
270
- "eval_runtime": 4.867,
271
- "eval_samples_per_second": 16.643,
272
- "eval_steps_per_second": 2.26,
273
  "eval_weighted avg": {
274
- "f1-score": 0.8514422719925407,
275
- "precision": 0.8534139871324166,
276
- "recall": 0.8502495290346036,
277
- "support": 30257.0
278
  },
279
  "step": 486
280
  },
281
  {
282
  "epoch": 6.17,
283
- "grad_norm": 4.494631767272949,
284
  "learning_rate": 1.7530864197530865e-05,
285
- "loss": 0.35,
286
  "step": 500
287
  },
288
  {
289
  "epoch": 7.0,
290
  "eval_Claim": {
291
- "f1-score": 0.6040515653775324,
292
- "precision": 0.6054453161052146,
293
- "recall": 0.6026642168121268,
294
- "support": 4354.0
295
  },
296
  "eval_MajorClaim": {
297
- "f1-score": 0.8052777186635455,
298
- "precision": 0.8051063829787234,
299
- "recall": 0.8054491272882077,
300
- "support": 2349.0
301
  },
302
  "eval_O": {
303
- "f1-score": 0.9184720354599332,
304
- "precision": 0.9315082331548641,
305
- "recall": 0.9057956777996071,
306
- "support": 10180.0
307
  },
308
  "eval_Premise": {
309
- "f1-score": 0.8908606921029282,
310
- "precision": 0.8810881965774463,
311
- "recall": 0.9008524001794527,
312
- "support": 13374.0
313
  },
314
- "eval_accuracy": 0.8521994910268698,
315
- "eval_loss": 0.5764245986938477,
316
  "eval_macro avg": {
317
- "f1-score": 0.8046655029009848,
318
- "precision": 0.8057870322040621,
319
- "recall": 0.8036903555198486,
320
- "support": 30257.0
321
- },
322
- "eval_runtime": 4.8398,
323
- "eval_samples_per_second": 16.736,
324
- "eval_steps_per_second": 2.273,
325
  "eval_weighted avg": {
326
- "f1-score": 0.8522343290465387,
327
- "precision": 0.8524880574571969,
328
- "recall": 0.8521994910268698,
329
- "support": 30257.0
330
  },
331
  "step": 567
332
  },
333
  {
334
  "epoch": 8.0,
335
  "eval_Claim": {
336
- "f1-score": 0.597346719254213,
337
- "precision": 0.6227261400448543,
338
- "recall": 0.5739549839228296,
339
- "support": 4354.0
340
  },
341
  "eval_MajorClaim": {
342
- "f1-score": 0.8047001620745543,
343
- "precision": 0.7676845767298028,
344
- "recall": 0.8454661558109834,
345
- "support": 2349.0
346
  },
347
  "eval_O": {
348
- "f1-score": 0.9099175618286286,
349
- "precision": 0.9258769700050838,
350
- "recall": 0.8944990176817289,
351
- "support": 10180.0
352
  },
353
  "eval_Premise": {
354
- "f1-score": 0.8942491542873952,
355
- "precision": 0.8797569092750688,
356
- "recall": 0.9092268580828473,
357
- "support": 13374.0
358
  },
359
- "eval_accuracy": 0.8510757841160723,
360
- "eval_loss": 0.5954988598823547,
361
  "eval_macro avg": {
362
- "f1-score": 0.8015533993611977,
363
- "precision": 0.7990111490137024,
364
- "recall": 0.8057867538745973,
365
- "support": 30257.0
366
- },
367
- "eval_runtime": 4.8609,
368
- "eval_samples_per_second": 16.663,
369
- "eval_steps_per_second": 2.263,
370
  "eval_weighted avg": {
371
- "f1-score": 0.8498442431569896,
372
- "precision": 0.8495864475589161,
373
- "recall": 0.8510757841160723,
374
- "support": 30257.0
375
  },
376
  "step": 648
377
  },
378
  {
379
  "epoch": 9.0,
380
  "eval_Claim": {
381
- "f1-score": 0.6049033474776048,
382
- "precision": 0.6213075060532688,
383
- "recall": 0.5893431327514929,
384
- "support": 4354.0
385
  },
386
  "eval_MajorClaim": {
387
- "f1-score": 0.823503227242377,
388
- "precision": 0.8628731343283582,
389
- "recall": 0.787569178373776,
390
- "support": 2349.0
391
  },
392
  "eval_O": {
393
- "f1-score": 0.9189658498938837,
394
- "precision": 0.9027672479150872,
395
- "recall": 0.9357563850687622,
396
- "support": 10180.0
397
  },
398
  "eval_Premise": {
399
- "f1-score": 0.8989367655288193,
400
- "precision": 0.8970292606656243,
401
- "recall": 0.9008524001794527,
402
- "support": 13374.0
403
  },
404
- "eval_accuracy": 0.8589747826949136,
405
- "eval_loss": 0.5689579844474792,
406
  "eval_macro avg": {
407
- "f1-score": 0.8115772975356712,
408
- "precision": 0.8209942872405847,
409
- "recall": 0.803380274093371,
410
- "support": 30257.0
411
- },
412
- "eval_runtime": 4.8593,
413
- "eval_samples_per_second": 16.669,
414
- "eval_steps_per_second": 2.264,
415
  "eval_weighted avg": {
416
- "f1-score": 0.8575077142417292,
417
- "precision": 0.8566315824374819,
418
- "recall": 0.8589747826949136,
419
- "support": 30257.0
420
  },
421
  "step": 729
422
  },
423
  {
424
  "epoch": 10.0,
425
  "eval_Claim": {
426
- "f1-score": 0.60162224797219,
427
- "precision": 0.6071094480823199,
428
- "recall": 0.5962333486449242,
429
- "support": 4354.0
430
  },
431
  "eval_MajorClaim": {
432
- "f1-score": 0.8131676477000218,
433
- "precision": 0.8333333333333334,
434
- "recall": 0.7939548744146445,
435
- "support": 2349.0
436
  },
437
  "eval_O": {
438
- "f1-score": 0.9197247706422018,
439
- "precision": 0.9338801134062373,
440
- "recall": 0.905992141453831,
441
- "support": 10180.0
442
  },
443
  "eval_Premise": {
444
- "f1-score": 0.9008479864909512,
445
- "precision": 0.884834499170693,
446
- "recall": 0.9174517720951099,
447
- "support": 13374.0
448
  },
449
- "eval_accuracy": 0.8577849753775986,
450
- "eval_loss": 0.7088457942008972,
451
  "eval_macro avg": {
452
- "f1-score": 0.8088406632013413,
453
- "precision": 0.8147893484981459,
454
- "recall": 0.8034080341521274,
455
- "support": 30257.0
456
- },
457
- "eval_runtime": 4.9043,
458
- "eval_samples_per_second": 16.516,
459
- "eval_steps_per_second": 2.243,
460
  "eval_weighted avg": {
461
- "f1-score": 0.8573332851434664,
462
- "precision": 0.8573728619273148,
463
- "recall": 0.8577849753775986,
464
- "support": 30257.0
465
  },
466
  "step": 810
467
  },
468
  {
469
  "epoch": 11.0,
470
  "eval_Claim": {
471
- "f1-score": 0.6253176253176252,
472
- "precision": 0.6289498141263941,
473
- "recall": 0.6217271474506201,
474
- "support": 4354.0
475
  },
476
  "eval_MajorClaim": {
477
- "f1-score": 0.7947071414672116,
478
- "precision": 0.7318996415770609,
479
- "recall": 0.8693060876968923,
480
- "support": 2349.0
481
  },
482
  "eval_O": {
483
- "f1-score": 0.9110249527786062,
484
- "precision": 0.9221171261823304,
485
- "recall": 0.900196463654224,
486
- "support": 10180.0
487
  },
488
  "eval_Premise": {
489
- "f1-score": 0.8996578818752584,
490
- "precision": 0.9047258979206049,
491
- "recall": 0.8946463286974727,
492
- "support": 13374.0
493
  },
494
- "eval_accuracy": 0.8552731599299336,
495
- "eval_loss": 0.7312190532684326,
496
  "eval_macro avg": {
497
- "f1-score": 0.8076769003596753,
498
- "precision": 0.7969231199515976,
499
- "recall": 0.8214690068748023,
500
- "support": 30257.0
501
- },
502
- "eval_runtime": 4.9029,
503
- "eval_samples_per_second": 16.521,
504
- "eval_steps_per_second": 2.244,
505
  "eval_weighted avg": {
506
- "f1-score": 0.8558567785115953,
507
- "precision": 0.857475501606145,
508
- "recall": 0.8552731599299336,
509
- "support": 30257.0
510
  },
511
  "step": 891
512
  },
513
  {
514
  "epoch": 12.0,
515
  "eval_Claim": {
516
- "f1-score": 0.633677135241627,
517
- "precision": 0.6395321637426901,
518
- "recall": 0.6279283417547084,
519
- "support": 4354.0
520
  },
521
  "eval_MajorClaim": {
522
- "f1-score": 0.8290997764681975,
523
- "precision": 0.7931570762052877,
524
- "recall": 0.8684546615581098,
525
- "support": 2349.0
526
  },
527
  "eval_O": {
528
- "f1-score": 0.916678984971668,
529
- "precision": 0.9196243203163619,
530
- "recall": 0.9137524557956778,
531
- "support": 10180.0
532
  },
533
  "eval_Premise": {
534
- "f1-score": 0.9029209944129889,
535
- "precision": 0.905603610379842,
536
- "recall": 0.9002542246149244,
537
- "support": 13374.0
538
  },
539
- "eval_accuracy": 0.8631391083055161,
540
- "eval_loss": 0.7092207074165344,
541
  "eval_macro avg": {
542
- "f1-score": 0.8205942227736204,
543
- "precision": 0.8144792926610454,
544
- "recall": 0.8275974209308551,
545
- "support": 30257.0
546
- },
547
- "eval_runtime": 4.8734,
548
- "eval_samples_per_second": 16.621,
549
- "eval_steps_per_second": 2.257,
550
  "eval_weighted avg": {
551
- "f1-score": 0.8630744313070275,
552
- "precision": 0.8633032778855295,
553
- "recall": 0.8631391083055161,
554
- "support": 30257.0
555
  },
556
  "step": 972
557
  },
558
  {
559
  "epoch": 12.35,
560
- "grad_norm": 4.426708698272705,
561
  "learning_rate": 1.506172839506173e-05,
562
- "loss": 0.063,
563
  "step": 1000
564
  },
565
  {
566
  "epoch": 13.0,
567
  "eval_Claim": {
568
- "f1-score": 0.5936837143530521,
569
- "precision": 0.6096321393998064,
570
- "recall": 0.5785484611851172,
571
- "support": 4354.0
572
  },
573
  "eval_MajorClaim": {
574
- "f1-score": 0.8065005417118094,
575
- "precision": 0.821270962047661,
576
- "recall": 0.7922520221370796,
577
- "support": 2349.0
578
  },
579
  "eval_O": {
580
- "f1-score": 0.9133826964170352,
581
- "precision": 0.9242683294780247,
582
- "recall": 0.9027504911591355,
583
- "support": 10180.0
584
  },
585
  "eval_Premise": {
586
- "f1-score": 0.901282521069989,
587
- "precision": 0.883730957171601,
588
- "recall": 0.9195453865709585,
589
- "support": 13374.0
590
  },
591
- "eval_accuracy": 0.854942657897346,
592
- "eval_loss": 0.7807008624076843,
593
  "eval_macro avg": {
594
- "f1-score": 0.8037123683879714,
595
- "precision": 0.8097255970242733,
596
- "recall": 0.7982740902630727,
597
- "support": 30257.0
598
- },
599
- "eval_runtime": 4.8517,
600
- "eval_samples_per_second": 16.695,
601
- "eval_steps_per_second": 2.267,
602
  "eval_weighted avg": {
603
- "f1-score": 0.8537315976828396,
604
- "precision": 0.8530777420132861,
605
- "recall": 0.854942657897346,
606
- "support": 30257.0
607
  },
608
  "step": 1053
609
  },
610
  {
611
  "epoch": 14.0,
612
  "eval_Claim": {
613
- "f1-score": 0.6335702820454024,
614
- "precision": 0.632554945054945,
615
- "recall": 0.6345888837850253,
616
- "support": 4354.0
617
  },
618
  "eval_MajorClaim": {
619
- "f1-score": 0.8183556405353729,
620
- "precision": 0.816793893129771,
621
- "recall": 0.8199233716475096,
622
- "support": 2349.0
623
  },
624
  "eval_O": {
625
- "f1-score": 0.9172396976347231,
626
- "precision": 0.9107990314769976,
627
- "recall": 0.9237721021611002,
628
- "support": 10180.0
629
  },
630
  "eval_Premise": {
631
- "f1-score": 0.8981941309255079,
632
- "precision": 0.9039073148568832,
633
- "recall": 0.8925527142216241,
634
- "support": 13374.0
635
  },
636
- "eval_accuracy": 0.8602967908252636,
637
- "eval_loss": 0.7879041433334351,
638
  "eval_macro avg": {
639
- "f1-score": 0.8168399377852517,
640
- "precision": 0.8160137961296492,
641
- "recall": 0.8177092679538147,
642
- "support": 30257.0
643
- },
644
- "eval_runtime": 4.864,
645
- "eval_samples_per_second": 16.653,
646
- "eval_steps_per_second": 2.262,
647
  "eval_weighted avg": {
648
- "f1-score": 0.8603242501425289,
649
- "precision": 0.8604152313535001,
650
- "recall": 0.8602967908252636,
651
- "support": 30257.0
652
  },
653
  "step": 1134
654
  },
655
  {
656
  "epoch": 15.0,
657
  "eval_Claim": {
658
- "f1-score": 0.6212592513139547,
659
- "precision": 0.5828134433487623,
660
- "recall": 0.6651355075792375,
661
- "support": 4354.0
662
  },
663
  "eval_MajorClaim": {
664
- "f1-score": 0.7988422575976845,
665
- "precision": 0.7765273311897106,
666
- "recall": 0.822477650063857,
667
- "support": 2349.0
668
  },
669
  "eval_O": {
670
- "f1-score": 0.9143625765954267,
671
- "precision": 0.9276255938542404,
672
- "recall": 0.9014734774066797,
673
- "support": 10180.0
674
  },
675
  "eval_Premise": {
676
- "f1-score": 0.8928883984627678,
677
- "precision": 0.9090416053304409,
678
- "recall": 0.8772992373261552,
679
- "support": 13374.0
680
  },
681
- "eval_accuracy": 0.8506461314737086,
682
- "eval_loss": 0.8185116052627563,
683
  "eval_macro avg": {
684
- "f1-score": 0.8068381209924584,
685
- "precision": 0.7990019934307886,
686
- "recall": 0.8165964680939823,
687
- "support": 30257.0
688
- },
689
- "eval_runtime": 4.8575,
690
- "eval_samples_per_second": 16.675,
691
- "eval_steps_per_second": 2.265,
692
  "eval_weighted avg": {
693
- "f1-score": 0.8537245501570023,
694
- "precision": 0.8580620487302317,
695
- "recall": 0.8506461314737086,
696
- "support": 30257.0
697
  },
698
  "step": 1215
699
  },
700
  {
701
  "epoch": 16.0,
702
  "eval_Claim": {
703
- "f1-score": 0.620457750298297,
704
- "precision": 0.5878725590955807,
705
- "recall": 0.6568672485071199,
706
- "support": 4354.0
707
  },
708
  "eval_MajorClaim": {
709
- "f1-score": 0.8217720482913852,
710
- "precision": 0.7911741528762806,
711
- "recall": 0.8548318433375904,
712
- "support": 2349.0
713
  },
714
  "eval_O": {
715
- "f1-score": 0.9171081176529942,
716
- "precision": 0.9427445285758739,
717
- "recall": 0.8928290766208251,
718
- "support": 10180.0
719
  },
720
  "eval_Premise": {
721
- "f1-score": 0.8952495580546884,
722
- "precision": 0.900703852266707,
723
- "recall": 0.8898609241812472,
724
- "support": 13374.0
725
  },
726
- "eval_accuracy": 0.8546121558647586,
727
- "eval_loss": 0.8933361768722534,
728
  "eval_macro avg": {
729
- "f1-score": 0.8136468685743412,
730
- "precision": 0.8056237732036106,
731
- "recall": 0.8235972731616956,
732
- "support": 30257.0
733
- },
734
- "eval_runtime": 4.839,
735
- "eval_samples_per_second": 16.739,
736
- "eval_steps_per_second": 2.273,
737
  "eval_weighted avg": {
738
- "f1-score": 0.8573567707758909,
739
- "precision": 0.8613285464033407,
740
- "recall": 0.8546121558647586,
741
- "support": 30257.0
742
  },
743
  "step": 1296
744
  },
745
  {
746
  "epoch": 17.0,
747
  "eval_Claim": {
748
- "f1-score": 0.577664702731655,
749
- "precision": 0.6926782273603083,
750
- "recall": 0.49540652273771246,
751
- "support": 4354.0
752
  },
753
  "eval_MajorClaim": {
754
- "f1-score": 0.8222534645511148,
755
- "precision": 0.7783269961977186,
756
- "recall": 0.8714346530438485,
757
- "support": 2349.0
758
  },
759
  "eval_O": {
760
- "f1-score": 0.9173265135494024,
761
- "precision": 0.9185462424899045,
762
- "recall": 0.9161100196463654,
763
- "support": 10180.0
764
  },
765
  "eval_Premise": {
766
- "f1-score": 0.901492752578063,
767
- "precision": 0.8705431754874652,
768
- "recall": 0.9347240915208613,
769
- "support": 13374.0
770
  },
771
- "eval_accuracy": 0.8603298410285223,
772
- "eval_loss": 0.8753499984741211,
773
  "eval_macro avg": {
774
- "f1-score": 0.8046843583525589,
775
- "precision": 0.8150236603838491,
776
- "recall": 0.8044188217371968,
777
- "support": 30257.0
778
- },
779
- "eval_runtime": 4.8417,
780
- "eval_samples_per_second": 16.73,
781
- "eval_steps_per_second": 2.272,
782
  "eval_weighted avg": {
783
- "f1-score": 0.8540692561997597,
784
- "precision": 0.8539397922302876,
785
- "recall": 0.8603298410285223,
786
- "support": 30257.0
787
  },
788
  "step": 1377
789
  },
790
  {
791
  "epoch": 18.0,
792
  "eval_Claim": {
793
- "f1-score": 0.6204532891100056,
794
- "precision": 0.5981667021956939,
795
- "recall": 0.6444648598989435,
796
- "support": 4354.0
797
  },
798
  "eval_MajorClaim": {
799
- "f1-score": 0.8362068965517241,
800
- "precision": 0.807372175980975,
801
- "recall": 0.8671775223499362,
802
- "support": 2349.0
803
  },
804
  "eval_O": {
805
- "f1-score": 0.9045747829039926,
806
- "precision": 0.8936068244991853,
807
- "recall": 0.9158153241650294,
808
- "support": 10180.0
809
  },
810
  "eval_Premise": {
811
- "f1-score": 0.8850831280788178,
812
- "precision": 0.9118953211736717,
813
- "recall": 0.8598026020637057,
814
- "support": 13374.0
815
  },
816
- "eval_accuracy": 0.8482334666358198,
817
- "eval_loss": 0.9685499668121338,
818
  "eval_macro avg": {
819
- "f1-score": 0.811579524161135,
820
- "precision": 0.8027602559623814,
821
- "recall": 0.8218150771194037,
822
- "support": 30257.0
823
- },
824
- "eval_runtime": 4.8521,
825
- "eval_samples_per_second": 16.694,
826
- "eval_steps_per_second": 2.267,
827
  "eval_weighted avg": {
828
- "f1-score": 0.8497662248627994,
829
- "precision": 0.8524817583209754,
830
- "recall": 0.8482334666358198,
831
- "support": 30257.0
832
  },
833
  "step": 1458
834
  },
835
  {
836
  "epoch": 18.52,
837
- "grad_norm": 0.44517388939857483,
838
  "learning_rate": 1.2592592592592593e-05,
839
- "loss": 0.0216,
840
  "step": 1500
841
  },
842
  {
843
  "epoch": 19.0,
844
  "eval_Claim": {
845
- "f1-score": 0.6178126034654013,
846
- "precision": 0.5946462715105163,
847
- "recall": 0.6428571428571429,
848
- "support": 4354.0
849
  },
850
  "eval_MajorClaim": {
851
- "f1-score": 0.8151062155782848,
852
- "precision": 0.7575868372943327,
853
- "recall": 0.8820774797786292,
854
- "support": 2349.0
855
  },
856
  "eval_O": {
857
- "f1-score": 0.9146588828971454,
858
- "precision": 0.9213595136051032,
859
- "recall": 0.9080550098231827,
860
- "support": 10180.0
861
  },
862
  "eval_Premise": {
863
- "f1-score": 0.8898914207065302,
864
- "precision": 0.9104991394148021,
865
- "recall": 0.870195902497383,
866
- "support": 13374.0
867
  },
868
- "eval_accuracy": 0.8511418845225898,
869
- "eval_loss": 0.9391252398490906,
870
  "eval_macro avg": {
871
- "f1-score": 0.8093672806618404,
872
- "precision": 0.7960229404561886,
873
- "recall": 0.8257963837390844,
874
- "support": 30257.0
875
- },
876
- "eval_runtime": 4.8661,
877
- "eval_samples_per_second": 16.646,
878
- "eval_steps_per_second": 2.261,
879
  "eval_weighted avg": {
880
- "f1-score": 0.853266214902463,
881
- "precision": 0.8568303759789367,
882
- "recall": 0.8511418845225898,
883
- "support": 30257.0
884
  },
885
  "step": 1539
886
  },
887
  {
888
  "epoch": 20.0,
889
  "eval_Claim": {
890
- "f1-score": 0.6097087378640776,
891
- "precision": 0.6064530788457169,
892
- "recall": 0.6129995406522738,
893
- "support": 4354.0
894
  },
895
  "eval_MajorClaim": {
896
- "f1-score": 0.8268961376826096,
897
- "precision": 0.7802114803625377,
898
- "recall": 0.8795232013622818,
899
- "support": 2349.0
900
  },
901
  "eval_O": {
902
- "f1-score": 0.9215637325877477,
903
- "precision": 0.9370494466443293,
904
- "recall": 0.906581532416503,
905
- "support": 10180.0
906
  },
907
  "eval_Premise": {
908
- "f1-score": 0.896644596566042,
909
- "precision": 0.8971479901190209,
910
- "recall": 0.8961417676087932,
911
- "support": 13374.0
912
  },
913
- "eval_accuracy": 0.8576197243613048,
914
- "eval_loss": 0.8915661573410034,
915
  "eval_macro avg": {
916
- "f1-score": 0.8137033011751192,
917
- "precision": 0.8052154989929012,
918
- "recall": 0.8238115105099629,
919
- "support": 30257.0
920
- },
921
- "eval_runtime": 4.864,
922
- "eval_samples_per_second": 16.653,
923
- "eval_steps_per_second": 2.261,
924
  "eval_weighted avg": {
925
- "f1-score": 0.8583235120565212,
926
- "precision": 0.8596633525913644,
927
- "recall": 0.8576197243613048,
928
- "support": 30257.0
929
  },
930
  "step": 1620
931
  },
932
  {
933
  "epoch": 21.0,
934
  "eval_Claim": {
935
- "f1-score": 0.6211416490486258,
936
- "precision": 0.5754014884449667,
937
- "recall": 0.6747818098300413,
938
- "support": 4354.0
939
  },
940
  "eval_MajorClaim": {
941
- "f1-score": 0.8162756902636495,
942
- "precision": 0.7965964343598055,
943
- "recall": 0.8369518944231588,
944
- "support": 2349.0
945
  },
946
  "eval_O": {
947
- "f1-score": 0.911611088790679,
948
- "precision": 0.9325935059597205,
949
- "recall": 0.8915520628683693,
950
- "support": 10180.0
951
  },
952
  "eval_Premise": {
953
- "f1-score": 0.8885090218423553,
954
- "precision": 0.9030190718863408,
955
- "recall": 0.8744579033946464,
956
- "support": 13374.0
957
  },
958
- "eval_accuracy": 0.8485639686684073,
959
- "eval_loss": 0.9740728735923767,
960
  "eval_macro avg": {
961
- "f1-score": 0.8093843624863275,
962
- "precision": 0.8019026251627084,
963
- "recall": 0.8194359176290539,
964
- "support": 30257.0
965
- },
966
- "eval_runtime": 4.884,
967
- "eval_samples_per_second": 16.585,
968
- "eval_steps_per_second": 2.252,
969
  "eval_weighted avg": {
970
- "f1-score": 0.852199586158436,
971
- "precision": 0.8575629461968616,
972
- "recall": 0.8485639686684073,
973
- "support": 30257.0
974
  },
975
  "step": 1701
976
  },
977
  {
978
  "epoch": 22.0,
979
  "eval_Claim": {
980
- "f1-score": 0.6145704312577412,
981
- "precision": 0.6028274795670422,
982
- "recall": 0.6267799724391364,
983
- "support": 4354.0
984
  },
985
  "eval_MajorClaim": {
986
- "f1-score": 0.8148905706688483,
987
- "precision": 0.784251968503937,
988
- "recall": 0.8480204342273308,
989
- "support": 2349.0
990
  },
991
  "eval_O": {
992
- "f1-score": 0.9175837613108339,
993
- "precision": 0.9137847053093034,
994
- "recall": 0.9214145383104125,
995
- "support": 10180.0
996
  },
997
  "eval_Premise": {
998
- "f1-score": 0.8915167877105593,
999
- "precision": 0.9070019342359767,
1000
- "recall": 0.876551517870495,
1001
- "support": 13374.0
1002
  },
1003
- "eval_accuracy": 0.853488448953961,
1004
- "eval_loss": 0.9675940871238708,
1005
  "eval_macro avg": {
1006
- "f1-score": 0.8096403877369956,
1007
- "precision": 0.8019665219040648,
1008
- "recall": 0.8181916157118437,
1009
- "support": 30257.0
1010
- },
1011
- "eval_runtime": 4.8521,
1012
- "eval_samples_per_second": 16.694,
1013
- "eval_steps_per_second": 2.267,
1014
  "eval_weighted avg": {
1015
- "f1-score": 0.8544854353433136,
1016
- "precision": 0.8559834381654265,
1017
- "recall": 0.853488448953961,
1018
- "support": 30257.0
1019
  },
1020
  "step": 1782
1021
  },
1022
  {
1023
  "epoch": 23.0,
1024
  "eval_Claim": {
1025
- "f1-score": 0.5983347015362965,
1026
- "precision": 0.6113108075724898,
1027
- "recall": 0.5858980248047773,
1028
- "support": 4354.0
1029
  },
1030
  "eval_MajorClaim": {
1031
- "f1-score": 0.8269188475918188,
1032
- "precision": 0.8553230209281165,
1033
- "recall": 0.800340570455513,
1034
- "support": 2349.0
1035
  },
1036
  "eval_O": {
1037
- "f1-score": 0.9219349129879069,
1038
- "precision": 0.9227514268844715,
1039
- "recall": 0.9211198428290767,
1040
- "support": 10180.0
1041
  },
1042
  "eval_Premise": {
1043
- "f1-score": 0.8949737988043397,
1044
- "precision": 0.8835616438356164,
1045
- "recall": 0.9066846119336025,
1046
- "support": 13374.0
1047
  },
1048
- "eval_accuracy": 0.8571239713124236,
1049
- "eval_loss": 0.9897539019584656,
1050
  "eval_macro avg": {
1051
- "f1-score": 0.8105405652300904,
1052
- "precision": 0.8182367248051736,
1053
- "recall": 0.8035107625057423,
1054
- "support": 30257.0
1055
- },
1056
- "eval_runtime": 4.8416,
1057
- "eval_samples_per_second": 16.73,
1058
- "eval_steps_per_second": 2.272,
1059
  "eval_weighted avg": {
1060
- "f1-score": 0.8560749136698399,
1061
- "precision": 0.8553777302003576,
1062
- "recall": 0.8571239713124236,
1063
- "support": 30257.0
1064
  },
1065
  "step": 1863
1066
  },
1067
  {
1068
  "epoch": 24.0,
1069
  "eval_Claim": {
1070
- "f1-score": 0.6090506975161619,
1071
- "precision": 0.6016132646202106,
1072
- "recall": 0.6166743224621039,
1073
- "support": 4354.0
1074
  },
1075
  "eval_MajorClaim": {
1076
- "f1-score": 0.7960317460317461,
1077
- "precision": 0.7454477889260498,
1078
- "recall": 0.853980417198808,
1079
- "support": 2349.0
1080
  },
1081
  "eval_O": {
1082
- "f1-score": 0.9115434760806339,
1083
- "precision": 0.9310592091784471,
1084
- "recall": 0.8928290766208251,
1085
- "support": 10180.0
1086
  },
1087
  "eval_Premise": {
1088
- "f1-score": 0.8995695302264645,
1089
- "precision": 0.9006821077880219,
1090
- "recall": 0.89845969792134,
1091
- "support": 13374.0
1092
  },
1093
- "eval_accuracy": 0.852563043262716,
1094
- "eval_loss": 0.9720127582550049,
1095
  "eval_macro avg": {
1096
- "f1-score": 0.8040488624637516,
1097
- "precision": 0.7947005926281823,
1098
- "recall": 0.8154858785507693,
1099
- "support": 30257.0
1100
- },
1101
- "eval_runtime": 4.8503,
1102
- "eval_samples_per_second": 16.7,
1103
- "eval_steps_per_second": 2.268,
1104
  "eval_weighted avg": {
1105
- "f1-score": 0.8537541987693271,
1106
- "precision": 0.8558147294621835,
1107
- "recall": 0.852563043262716,
1108
- "support": 30257.0
1109
  },
1110
  "step": 1944
1111
  },
1112
  {
1113
  "epoch": 24.69,
1114
- "grad_norm": 3.446370840072632,
1115
  "learning_rate": 1.0123456790123458e-05,
1116
- "loss": 0.0116,
1117
  "step": 2000
1118
  },
1119
  {
1120
  "epoch": 25.0,
1121
  "eval_Claim": {
1122
- "f1-score": 0.6102259215219976,
1123
- "precision": 0.6326429980276134,
1124
- "recall": 0.5893431327514929,
1125
- "support": 4354.0
1126
  },
1127
  "eval_MajorClaim": {
1128
- "f1-score": 0.8064243448858834,
1129
- "precision": 0.8006714225765842,
1130
- "recall": 0.8122605363984674,
1131
- "support": 2349.0
1132
  },
1133
  "eval_O": {
1134
- "f1-score": 0.9169322709163347,
1135
- "precision": 0.9298989898989899,
1136
- "recall": 0.9043222003929273,
1137
- "support": 10180.0
1138
  },
1139
  "eval_Premise": {
1140
- "f1-score": 0.8993844349992672,
1141
- "precision": 0.8818077309958328,
1142
- "recall": 0.917676087931808,
1143
- "support": 13374.0
1144
  },
1145
- "eval_accuracy": 0.8577519251743398,
1146
- "eval_loss": 1.0132465362548828,
1147
  "eval_macro avg": {
1148
- "f1-score": 0.8082417430808707,
1149
- "precision": 0.811255285374755,
1150
- "recall": 0.8059004893686739,
1151
- "support": 30257.0
1152
- },
1153
- "eval_runtime": 4.8662,
1154
- "eval_samples_per_second": 16.645,
1155
- "eval_steps_per_second": 2.26,
1156
  "eval_weighted avg": {
1157
- "f1-score": 0.8564613940592988,
1158
- "precision": 0.8558341242209939,
1159
- "recall": 0.8577519251743398,
1160
- "support": 30257.0
1161
  },
1162
  "step": 2025
1163
  },
1164
  {
1165
  "epoch": 26.0,
1166
  "eval_Claim": {
1167
- "f1-score": 0.6123184326089406,
1168
- "precision": 0.6006185111552905,
1169
- "recall": 0.6244832338079926,
1170
- "support": 4354.0
1171
  },
1172
  "eval_MajorClaim": {
1173
- "f1-score": 0.7939849624060151,
1174
- "precision": 0.8013876843018214,
1175
- "recall": 0.7867177522349936,
1176
- "support": 2349.0
1177
  },
1178
  "eval_O": {
1179
- "f1-score": 0.9205986178093769,
1180
- "precision": 0.93204469948656,
1181
- "recall": 0.9094302554027505,
1182
- "support": 10180.0
1183
  },
1184
  "eval_Premise": {
1185
- "f1-score": 0.8996091568955891,
1186
- "precision": 0.8957082499444073,
1187
- "recall": 0.9035441902198296,
1188
- "support": 13374.0
1189
  },
1190
- "eval_accuracy": 0.8562977162309549,
1191
- "eval_loss": 1.0289610624313354,
1192
  "eval_macro avg": {
1193
- "f1-score": 0.8066277924299805,
1194
- "precision": 0.8074397862220198,
1195
- "recall": 0.8060438579163915,
1196
- "support": 30257.0
1197
- },
1198
- "eval_runtime": 4.8488,
1199
- "eval_samples_per_second": 16.705,
1200
- "eval_steps_per_second": 2.269,
1201
  "eval_weighted avg": {
1202
- "f1-score": 0.8571296534981037,
1203
- "precision": 0.8581475309358099,
1204
- "recall": 0.8562977162309549,
1205
- "support": 30257.0
1206
  },
1207
  "step": 2106
1208
  },
1209
  {
1210
  "epoch": 27.0,
1211
  "eval_Claim": {
1212
- "f1-score": 0.6028743397617001,
1213
- "precision": 0.6480063374702931,
1214
- "recall": 0.5636196600826826,
1215
- "support": 4354.0
1216
  },
1217
  "eval_MajorClaim": {
1218
- "f1-score": 0.8248976954555244,
1219
- "precision": 0.8347863993025283,
1220
- "recall": 0.8152405278842061,
1221
- "support": 2349.0
1222
  },
1223
  "eval_O": {
1224
- "f1-score": 0.9177434297137843,
1225
- "precision": 0.9111240197502178,
1226
- "recall": 0.9244597249508841,
1227
- "support": 10180.0
1228
  },
1229
  "eval_Premise": {
1230
- "f1-score": 0.8988648469931302,
1231
- "precision": 0.883512674225464,
1232
- "recall": 0.9147599820547331,
1233
- "support": 13374.0
1234
  },
1235
- "eval_accuracy": 0.8597679875731236,
1236
- "eval_loss": 1.0470740795135498,
1237
  "eval_macro avg": {
1238
- "f1-score": 0.8110950779810349,
1239
- "precision": 0.8193573576871258,
1240
- "recall": 0.8045199737431264,
1241
- "support": 30257.0
1242
- },
1243
- "eval_runtime": 4.8822,
1244
- "eval_samples_per_second": 16.591,
1245
- "eval_steps_per_second": 2.253,
1246
  "eval_weighted avg": {
1247
- "f1-score": 0.8568809247486504,
1248
- "precision": 0.8551301805022264,
1249
- "recall": 0.8597679875731236,
1250
- "support": 30257.0
1251
  },
1252
  "step": 2187
1253
  }
@@ -1257,7 +1257,7 @@
1257
  "num_input_tokens_seen": 0,
1258
  "num_train_epochs": 50,
1259
  "save_steps": 500,
1260
- "total_flos": 3870238964419800.0,
1261
  "train_batch_size": 4,
1262
  "trial_name": null,
1263
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_Claim": {
14
+ "f1-score": 0.38183790485045715,
15
+ "precision": 0.6497890295358649,
16
+ "recall": 0.27035330261136714,
17
+ "support": 4557.0
18
  },
19
  "eval_MajorClaim": {
20
+ "f1-score": 0.6266258607498087,
21
+ "precision": 0.5535653937140926,
22
+ "recall": 0.7219039224327898,
23
+ "support": 2269.0
24
  },
25
  "eval_O": {
26
+ "f1-score": 0.8545803971812939,
27
+ "precision": 0.8443037974683544,
28
+ "recall": 0.8651102464332037,
29
+ "support": 8481.0
30
  },
31
  "eval_Premise": {
32
+ "f1-score": 0.878949075575738,
33
+ "precision": 0.8314310260186549,
34
+ "recall": 0.9322278794550709,
35
+ "support": 14534.0
36
  },
37
+ "eval_accuracy": 0.7960859220535504,
38
+ "eval_loss": 0.5359914898872375,
39
  "eval_macro avg": {
40
+ "f1-score": 0.6854983095893245,
41
+ "precision": 0.7197723116842417,
42
+ "recall": 0.697398837733108,
43
+ "support": 29841.0
44
+ },
45
+ "eval_runtime": 4.8191,
46
+ "eval_samples_per_second": 16.601,
47
+ "eval_steps_per_second": 2.075,
48
  "eval_weighted avg": {
49
+ "f1-score": 0.7769240850962494,
50
+ "precision": 0.7862232339538372,
51
+ "recall": 0.7960859220535504,
52
+ "support": 29841.0
53
  },
54
  "step": 81
55
  },
56
  {
57
  "epoch": 2.0,
58
  "eval_Claim": {
59
+ "f1-score": 0.6161300742974051,
60
+ "precision": 0.6048625792811839,
61
+ "recall": 0.6278253236778583,
62
+ "support": 4557.0
63
  },
64
  "eval_MajorClaim": {
65
+ "f1-score": 0.7920215150156881,
66
+ "precision": 0.8057455540355677,
67
+ "recall": 0.7787571617452622,
68
+ "support": 2269.0
69
  },
70
  "eval_O": {
71
+ "f1-score": 0.8922796352583586,
72
+ "precision": 0.9209436566695947,
73
+ "recall": 0.865346067680698,
74
+ "support": 8481.0
75
  },
76
  "eval_Premise": {
77
+ "f1-score": 0.8914289590611538,
78
+ "precision": 0.8790554552143957,
79
+ "recall": 0.9041557726709784,
80
+ "support": 14534.0
81
  },
82
+ "eval_accuracy": 0.8413927147213566,
83
+ "eval_loss": 0.443877637386322,
84
  "eval_macro avg": {
85
+ "f1-score": 0.7979650459081514,
86
+ "precision": 0.8026518113001855,
87
+ "recall": 0.7940210814436992,
88
+ "support": 29841.0
89
+ },
90
+ "eval_runtime": 4.8087,
91
+ "eval_samples_per_second": 16.637,
92
+ "eval_steps_per_second": 2.08,
93
  "eval_weighted avg": {
94
+ "f1-score": 0.842071433389123,
95
+ "precision": 0.8435143116581857,
96
+ "recall": 0.8413927147213566,
97
+ "support": 29841.0
98
  },
99
  "step": 162
100
  },
101
  {
102
  "epoch": 3.0,
103
  "eval_Claim": {
104
+ "f1-score": 0.6042799305957202,
105
+ "precision": 0.6389432485322897,
106
+ "recall": 0.5731841123546193,
107
+ "support": 4557.0
108
  },
109
  "eval_MajorClaim": {
110
+ "f1-score": 0.8349557522123894,
111
+ "precision": 0.8382940915148823,
112
+ "recall": 0.8316438959894227,
113
+ "support": 2269.0
114
  },
115
  "eval_O": {
116
+ "f1-score": 0.8941106484235574,
117
+ "precision": 0.9022691799735862,
118
+ "recall": 0.8860983374602052,
119
+ "support": 8481.0
120
  },
121
  "eval_Premise": {
122
+ "f1-score": 0.8917763490086511,
123
+ "precision": 0.8729980887102089,
124
+ "recall": 0.9113802119168846,
125
+ "support": 14534.0
126
  },
127
+ "eval_accuracy": 0.8464863778023525,
128
+ "eval_loss": 0.4555923044681549,
129
  "eval_macro avg": {
130
+ "f1-score": 0.8062806700600795,
131
+ "precision": 0.8131261521827418,
132
+ "recall": 0.8005766394302829,
133
+ "support": 29841.0
134
+ },
135
+ "eval_runtime": 4.7811,
136
+ "eval_samples_per_second": 16.733,
137
+ "eval_steps_per_second": 2.092,
138
  "eval_weighted avg": {
139
+ "f1-score": 0.8442159482345275,
140
+ "precision": 0.8429359878649869,
141
+ "recall": 0.8464863778023525,
142
+ "support": 29841.0
143
  },
144
  "step": 243
145
  },
146
  {
147
  "epoch": 4.0,
148
  "eval_Claim": {
149
+ "f1-score": 0.6278026905829596,
150
+ "precision": 0.5981717011128775,
151
+ "recall": 0.6605222734254992,
152
+ "support": 4557.0
153
  },
154
  "eval_MajorClaim": {
155
+ "f1-score": 0.8182207014542344,
156
+ "precision": 0.7947652679684254,
157
+ "recall": 0.8431026884089907,
158
+ "support": 2269.0
159
  },
160
  "eval_O": {
161
+ "f1-score": 0.9069328626209752,
162
+ "precision": 0.9373427277302466,
163
+ "recall": 0.8784341469166372,
164
+ "support": 8481.0
165
  },
166
  "eval_Premise": {
167
+ "f1-score": 0.888919552918449,
168
+ "precision": 0.8913795489137954,
169
+ "recall": 0.8864730975643319,
170
+ "support": 14534.0
171
  },
172
+ "eval_accuracy": 0.8463858449783854,
173
+ "eval_loss": 0.4377317428588867,
174
  "eval_macro avg": {
175
+ "f1-score": 0.8104689518941546,
176
+ "precision": 0.8054148114313363,
177
+ "recall": 0.8171330515788647,
178
+ "support": 29841.0
179
+ },
180
+ "eval_runtime": 4.8109,
181
+ "eval_samples_per_second": 16.629,
182
+ "eval_steps_per_second": 2.079,
183
  "eval_weighted avg": {
184
+ "f1-score": 0.8487883791626096,
185
+ "precision": 0.8523207959781866,
186
+ "recall": 0.8463858449783854,
187
+ "support": 29841.0
188
  },
189
  "step": 324
190
  },
191
  {
192
  "epoch": 5.0,
193
  "eval_Claim": {
194
+ "f1-score": 0.6247990569070839,
195
+ "precision": 0.6105990783410138,
196
+ "recall": 0.6396752249286811,
197
+ "support": 4557.0
198
  },
199
  "eval_MajorClaim": {
200
+ "f1-score": 0.7998289136013687,
201
+ "precision": 0.7769007062733694,
202
+ "recall": 0.8241516086381666,
203
+ "support": 2269.0
204
  },
205
  "eval_O": {
206
+ "f1-score": 0.9028039585296891,
207
+ "precision": 0.9020600353148911,
208
+ "recall": 0.9035491097747907,
209
+ "support": 8481.0
210
  },
211
  "eval_Premise": {
212
+ "f1-score": 0.8885327014878567,
213
+ "precision": 0.9001058948111542,
214
+ "recall": 0.8772533370028898,
215
+ "support": 14534.0
216
  },
217
+ "eval_accuracy": 0.8444086994403672,
218
+ "eval_loss": 0.5236712694168091,
219
  "eval_macro avg": {
220
+ "f1-score": 0.8039911576314996,
221
+ "precision": 0.7974164286851071,
222
+ "recall": 0.8111573200861321,
223
+ "support": 29841.0
224
+ },
225
+ "eval_runtime": 4.8153,
226
+ "eval_samples_per_second": 16.614,
227
+ "eval_steps_per_second": 2.077,
228
  "eval_weighted avg": {
229
+ "f1-score": 0.8455693764619782,
230
+ "precision": 0.8470828034323642,
231
+ "recall": 0.8444086994403672,
232
+ "support": 29841.0
233
  },
234
  "step": 405
235
  },
236
  {
237
  "epoch": 6.0,
238
  "eval_Claim": {
239
+ "f1-score": 0.6166529266281945,
240
+ "precision": 0.5813095006800078,
241
+ "recall": 0.6565723063418916,
242
+ "support": 4557.0
243
  },
244
  "eval_MajorClaim": {
245
+ "f1-score": 0.8074107959743824,
246
+ "precision": 0.8392772230147408,
247
+ "recall": 0.7778757161745262,
248
+ "support": 2269.0
249
  },
250
  "eval_O": {
251
+ "f1-score": 0.8913107165875667,
252
+ "precision": 0.8858607034707664,
253
+ "recall": 0.8968282042212004,
254
+ "support": 8481.0
255
  },
256
  "eval_Premise": {
257
+ "f1-score": 0.8815305371596762,
258
+ "precision": 0.8981792217065334,
259
+ "recall": 0.865487821659557,
260
+ "support": 14534.0
261
  },
262
+ "eval_accuracy": 0.8358298984618477,
263
+ "eval_loss": 0.5537309050559998,
264
  "eval_macro avg": {
265
+ "f1-score": 0.7992262440874549,
266
+ "precision": 0.8011566622180121,
267
+ "recall": 0.7991910120992938,
268
+ "support": 29841.0
269
+ },
270
+ "eval_runtime": 4.7996,
271
+ "eval_samples_per_second": 16.668,
272
+ "eval_steps_per_second": 2.084,
273
  "eval_weighted avg": {
274
+ "f1-score": 0.838225042631562,
275
+ "precision": 0.8418105575563007,
276
+ "recall": 0.8358298984618477,
277
+ "support": 29841.0
278
  },
279
  "step": 486
280
  },
281
  {
282
  "epoch": 6.17,
283
+ "grad_norm": 6.804128646850586,
284
  "learning_rate": 1.7530864197530865e-05,
285
+ "loss": 0.3426,
286
  "step": 500
287
  },
288
  {
289
  "epoch": 7.0,
290
  "eval_Claim": {
291
+ "f1-score": 0.6206536493682782,
292
+ "precision": 0.5920318725099601,
293
+ "recall": 0.652183454026772,
294
+ "support": 4557.0
295
  },
296
  "eval_MajorClaim": {
297
+ "f1-score": 0.8364370652905541,
298
+ "precision": 0.8519195612431444,
299
+ "recall": 0.8215072719259586,
300
+ "support": 2269.0
301
  },
302
  "eval_O": {
303
+ "f1-score": 0.9008658773634918,
304
+ "precision": 0.9000706214689266,
305
+ "recall": 0.9016625397948355,
306
+ "support": 8481.0
307
  },
308
  "eval_Premise": {
309
+ "f1-score": 0.8854242963272995,
310
+ "precision": 0.8978566881233642,
311
+ "recall": 0.8733314985551122,
312
+ "support": 14534.0
313
  },
314
+ "eval_accuracy": 0.8436714587312758,
315
+ "eval_loss": 0.5738794207572937,
316
  "eval_macro avg": {
317
+ "f1-score": 0.8108452220874058,
318
+ "precision": 0.8104696858363488,
319
+ "recall": 0.8121711910756696,
320
+ "support": 29841.0
321
+ },
322
+ "eval_runtime": 4.8167,
323
+ "eval_samples_per_second": 16.609,
324
+ "eval_steps_per_second": 2.076,
325
  "eval_weighted avg": {
326
+ "f1-score": 0.8456551258354699,
327
+ "precision": 0.8482906998207675,
328
+ "recall": 0.8436714587312758,
329
+ "support": 29841.0
330
  },
331
  "step": 567
332
  },
333
  {
334
  "epoch": 8.0,
335
  "eval_Claim": {
336
+ "f1-score": 0.6410105757931845,
337
+ "precision": 0.5787798408488064,
338
+ "recall": 0.7182356813693219,
339
+ "support": 4557.0
340
  },
341
  "eval_MajorClaim": {
342
+ "f1-score": 0.8115942028985507,
343
+ "precision": 0.8804123711340206,
344
+ "recall": 0.75275451740855,
345
+ "support": 2269.0
346
  },
347
  "eval_O": {
348
+ "f1-score": 0.8960937967338637,
349
+ "precision": 0.9094220495386109,
350
+ "recall": 0.8831505718665251,
351
+ "support": 8481.0
352
  },
353
  "eval_Premise": {
354
+ "f1-score": 0.891816143497758,
355
+ "precision": 0.9084939329050679,
356
+ "recall": 0.8757396449704142,
357
+ "support": 14534.0
358
  },
359
+ "eval_accuracy": 0.8444422103816897,
360
+ "eval_loss": 0.5891872644424438,
361
  "eval_macro avg": {
362
+ "f1-score": 0.8101286797308392,
363
+ "precision": 0.8192770486066265,
364
+ "recall": 0.8074701039037029,
365
+ "support": 29841.0
366
+ },
367
+ "eval_runtime": 4.799,
368
+ "eval_samples_per_second": 16.67,
369
+ "eval_steps_per_second": 2.084,
370
  "eval_weighted avg": {
371
+ "f1-score": 0.8486317402219317,
372
+ "precision": 0.8562720628608399,
373
+ "recall": 0.8444422103816897,
374
+ "support": 29841.0
375
  },
376
  "step": 648
377
  },
378
  {
379
  "epoch": 9.0,
380
  "eval_Claim": {
381
+ "f1-score": 0.6284997910572503,
382
+ "precision": 0.5998005982053839,
383
+ "recall": 0.6600833881939873,
384
+ "support": 4557.0
385
  },
386
  "eval_MajorClaim": {
387
+ "f1-score": 0.816712834718375,
388
+ "precision": 0.857489093553078,
389
+ "recall": 0.7796386073159982,
390
+ "support": 2269.0
391
  },
392
  "eval_O": {
393
+ "f1-score": 0.8971996170416466,
394
+ "precision": 0.9108249301421455,
395
+ "recall": 0.8839759462327555,
396
+ "support": 8481.0
397
  },
398
  "eval_Premise": {
399
+ "f1-score": 0.8942406935938898,
400
+ "precision": 0.8943022295623452,
401
+ "recall": 0.8941791660932985,
402
+ "support": 14534.0
403
  },
404
+ "eval_accuracy": 0.8468214872155759,
405
+ "eval_loss": 0.7009620070457458,
406
  "eval_macro avg": {
407
+ "f1-score": 0.8091632341027905,
408
+ "precision": 0.8156042128657381,
409
+ "recall": 0.8044692769590099,
410
+ "support": 29841.0
411
+ },
412
+ "eval_runtime": 4.8005,
413
+ "eval_samples_per_second": 16.665,
414
+ "eval_steps_per_second": 2.083,
415
  "eval_weighted avg": {
416
+ "f1-score": 0.848605581671113,
417
+ "precision": 0.8512257939173797,
418
+ "recall": 0.8468214872155759,
419
+ "support": 29841.0
420
  },
421
  "step": 729
422
  },
423
  {
424
  "epoch": 10.0,
425
  "eval_Claim": {
426
+ "f1-score": 0.6189745836985101,
427
+ "precision": 0.6180266900021877,
428
+ "recall": 0.619925389510643,
429
+ "support": 4557.0
430
  },
431
  "eval_MajorClaim": {
432
+ "f1-score": 0.8208921766420085,
433
+ "precision": 0.8353102189781022,
434
+ "recall": 0.8069634200088145,
435
+ "support": 2269.0
436
  },
437
  "eval_O": {
438
+ "f1-score": 0.8951902368987796,
439
+ "precision": 0.9085610200364299,
440
+ "recall": 0.8822072868765476,
441
+ "support": 8481.0
442
  },
443
  "eval_Premise": {
444
+ "f1-score": 0.8963474827245805,
445
+ "precision": 0.8870174493027017,
446
+ "recall": 0.905875877253337,
447
+ "support": 14534.0
448
  },
449
+ "eval_accuracy": 0.8479608592205355,
450
+ "eval_loss": 0.7416248917579651,
451
  "eval_macro avg": {
452
+ "f1-score": 0.8078511199909697,
453
+ "precision": 0.8122288445798553,
454
+ "recall": 0.8037429934123356,
455
+ "support": 29841.0
456
+ },
457
+ "eval_runtime": 4.7804,
458
+ "eval_samples_per_second": 16.735,
459
+ "eval_steps_per_second": 2.092,
460
  "eval_weighted avg": {
461
+ "f1-score": 0.8479238041544329,
462
+ "precision": 0.8481312332795722,
463
+ "recall": 0.8479608592205355,
464
+ "support": 29841.0
465
  },
466
  "step": 810
467
  },
468
  {
469
  "epoch": 11.0,
470
  "eval_Claim": {
471
+ "f1-score": 0.6350413962855225,
472
+ "precision": 0.6477973065510158,
473
+ "recall": 0.6227781435154707,
474
+ "support": 4557.0
475
  },
476
  "eval_MajorClaim": {
477
+ "f1-score": 0.8403837767117314,
478
+ "precision": 0.8316788951230039,
479
+ "recall": 0.8492728074041428,
480
+ "support": 2269.0
481
  },
482
  "eval_O": {
483
+ "f1-score": 0.9076665470637484,
484
+ "precision": 0.920883387938357,
485
+ "recall": 0.894823723617498,
486
+ "support": 8481.0
487
  },
488
  "eval_Premise": {
489
+ "f1-score": 0.9016170675363501,
490
+ "precision": 0.8904844987250033,
491
+ "recall": 0.9130315123159488,
492
+ "support": 14534.0
493
  },
494
+ "eval_accuracy": 0.8586843604436849,
495
+ "eval_loss": 0.753268837928772,
496
  "eval_macro avg": {
497
+ "f1-score": 0.8211771968993381,
498
+ "precision": 0.822711022084345,
499
+ "recall": 0.8199765467132651,
500
+ "support": 29841.0
501
+ },
502
+ "eval_runtime": 4.7829,
503
+ "eval_samples_per_second": 16.726,
504
+ "eval_steps_per_second": 2.091,
505
  "eval_weighted avg": {
506
+ "f1-score": 0.8579718131916827,
507
+ "precision": 0.8575920866110881,
508
+ "recall": 0.8586843604436849,
509
+ "support": 29841.0
510
  },
511
  "step": 891
512
  },
513
  {
514
  "epoch": 12.0,
515
  "eval_Claim": {
516
+ "f1-score": 0.6230640668523677,
517
+ "precision": 0.6328655500226347,
518
+ "recall": 0.6135615536537196,
519
+ "support": 4557.0
520
  },
521
  "eval_MajorClaim": {
522
+ "f1-score": 0.837270341207349,
523
+ "precision": 0.8310898827616153,
524
+ "recall": 0.8435434111943587,
525
+ "support": 2269.0
526
  },
527
  "eval_O": {
528
+ "f1-score": 0.8982022214588343,
529
+ "precision": 0.8730105731775181,
530
+ "recall": 0.9248909326730338,
531
+ "support": 8481.0
532
  },
533
  "eval_Premise": {
534
+ "f1-score": 0.8888346297394397,
535
+ "precision": 0.9013795542978422,
536
+ "recall": 0.8766340993532407,
537
+ "support": 14534.0
538
  },
539
+ "eval_accuracy": 0.8476592607486344,
540
+ "eval_loss": 0.8125314712524414,
541
  "eval_macro avg": {
542
+ "f1-score": 0.8118428148144977,
543
+ "precision": 0.8095863900649026,
544
+ "recall": 0.8146574992185882,
545
+ "support": 29841.0
546
+ },
547
+ "eval_runtime": 4.7824,
548
+ "eval_samples_per_second": 16.728,
549
+ "eval_steps_per_second": 2.091,
550
  "eval_weighted avg": {
551
+ "f1-score": 0.846990546753497,
552
+ "precision": 0.8469677413197487,
553
+ "recall": 0.8476592607486344,
554
+ "support": 29841.0
555
  },
556
  "step": 972
557
  },
558
  {
559
  "epoch": 12.35,
560
+ "grad_norm": 15.335172653198242,
561
  "learning_rate": 1.506172839506173e-05,
562
+ "loss": 0.061,
563
  "step": 1000
564
  },
565
  {
566
  "epoch": 13.0,
567
  "eval_Claim": {
568
+ "f1-score": 0.6158673240699238,
569
+ "precision": 0.6292649416075109,
570
+ "recall": 0.6030283080974326,
571
+ "support": 4557.0
572
  },
573
  "eval_MajorClaim": {
574
+ "f1-score": 0.8147180492053124,
575
+ "precision": 0.8050774526678142,
576
+ "recall": 0.8245923314235346,
577
+ "support": 2269.0
578
  },
579
  "eval_O": {
580
+ "f1-score": 0.9060346875742457,
581
+ "precision": 0.9128665469778575,
582
+ "recall": 0.8993043273198915,
583
+ "support": 8481.0
584
  },
585
  "eval_Premise": {
586
+ "f1-score": 0.9008830850011933,
587
+ "precision": 0.8929368029739777,
588
+ "recall": 0.9089720655015825,
589
+ "support": 14534.0
590
  },
591
+ "eval_accuracy": 0.8530880332428538,
592
+ "eval_loss": 0.8537381887435913,
593
  "eval_macro avg": {
594
+ "f1-score": 0.8093757864626687,
595
+ "precision": 0.8100364360567901,
596
+ "recall": 0.8089742580856103,
597
+ "support": 29841.0
598
+ },
599
+ "eval_runtime": 4.7933,
600
+ "eval_samples_per_second": 16.69,
601
+ "eval_steps_per_second": 2.086,
602
  "eval_weighted avg": {
603
+ "f1-score": 0.8522709558043637,
604
+ "precision": 0.8516552983596964,
605
+ "recall": 0.8530880332428538,
606
+ "support": 29841.0
607
  },
608
  "step": 1053
609
  },
610
  {
611
  "epoch": 14.0,
612
  "eval_Claim": {
613
+ "f1-score": 0.6346033728919426,
614
+ "precision": 0.6036838978015449,
615
+ "recall": 0.6688610928242265,
616
+ "support": 4557.0
617
  },
618
  "eval_MajorClaim": {
619
+ "f1-score": 0.8067737733391228,
620
+ "precision": 0.7950363714163458,
621
+ "recall": 0.8188629352137505,
622
+ "support": 2269.0
623
  },
624
  "eval_O": {
625
+ "f1-score": 0.9004597838418822,
626
+ "precision": 0.9121703363174449,
627
+ "recall": 0.8890461030538852,
628
+ "support": 8481.0
629
  },
630
  "eval_Premise": {
631
+ "f1-score": 0.8921770010096438,
632
+ "precision": 0.903023468884347,
633
+ "recall": 0.8815880005504335,
634
+ "support": 14534.0
635
  },
636
+ "eval_accuracy": 0.8464528668610302,
637
+ "eval_loss": 0.9449928998947144,
638
  "eval_macro avg": {
639
+ "f1-score": 0.8085034827706478,
640
+ "precision": 0.8034785186049206,
641
+ "recall": 0.8145895329105739,
642
+ "support": 29841.0
643
+ },
644
+ "eval_runtime": 4.8154,
645
+ "eval_samples_per_second": 16.613,
646
+ "eval_steps_per_second": 2.077,
647
  "eval_weighted avg": {
648
+ "f1-score": 0.8487033685671466,
649
+ "precision": 0.8517001698367573,
650
+ "recall": 0.8464528668610302,
651
+ "support": 29841.0
652
  },
653
  "step": 1134
654
  },
655
  {
656
  "epoch": 15.0,
657
  "eval_Claim": {
658
+ "f1-score": 0.6062717770034843,
659
+ "precision": 0.5687367813881946,
660
+ "recall": 0.6491112574061882,
661
+ "support": 4557.0
662
  },
663
  "eval_MajorClaim": {
664
+ "f1-score": 0.8176843948347561,
665
+ "precision": 0.8121739130434783,
666
+ "recall": 0.8232701630674306,
667
+ "support": 2269.0
668
  },
669
  "eval_O": {
670
+ "f1-score": 0.8942823803967327,
671
+ "precision": 0.8850906571197598,
672
+ "recall": 0.9036670203985379,
673
+ "support": 8481.0
674
  },
675
  "eval_Premise": {
676
+ "f1-score": 0.877618288144604,
677
+ "precision": 0.9049777063080184,
678
+ "recall": 0.8518645933672767,
679
+ "support": 14534.0
680
  },
681
+ "eval_accuracy": 0.8334506216279616,
682
+ "eval_loss": 1.0531803369522095,
683
  "eval_macro avg": {
684
+ "f1-score": 0.7989642100948944,
685
+ "precision": 0.7927447644648628,
686
+ "recall": 0.8069782585598584,
687
+ "support": 29841.0
688
+ },
689
+ "eval_runtime": 4.774,
690
+ "eval_samples_per_second": 16.757,
691
+ "eval_steps_per_second": 2.095,
692
  "eval_weighted avg": {
693
+ "f1-score": 0.8363600230462553,
694
+ "precision": 0.8409220859889104,
695
+ "recall": 0.8334506216279616,
696
+ "support": 29841.0
697
  },
698
  "step": 1215
699
  },
700
  {
701
  "epoch": 16.0,
702
  "eval_Claim": {
703
+ "f1-score": 0.6349479110729246,
704
+ "precision": 0.6217921750105174,
705
+ "recall": 0.6486723721746763,
706
+ "support": 4557.0
707
  },
708
  "eval_MajorClaim": {
709
+ "f1-score": 0.8260292164674634,
710
+ "precision": 0.8297020898176968,
711
+ "recall": 0.8223887174966946,
712
+ "support": 2269.0
713
  },
714
  "eval_O": {
715
+ "f1-score": 0.9045516104339634,
716
+ "precision": 0.9034344860032933,
717
+ "recall": 0.9056715010022403,
718
+ "support": 8481.0
719
  },
720
  "eval_Premise": {
721
+ "f1-score": 0.8956702459300312,
722
+ "precision": 0.90185546875,
723
+ "recall": 0.8895692858125774,
724
+ "support": 14534.0
725
  },
726
+ "eval_accuracy": 0.8522502597097953,
727
+ "eval_loss": 0.9694080352783203,
728
  "eval_macro avg": {
729
+ "f1-score": 0.8152997459760956,
730
+ "precision": 0.8141960548953769,
731
+ "recall": 0.8165754691215471,
732
+ "support": 29841.0
733
+ },
734
+ "eval_runtime": 4.7829,
735
+ "eval_samples_per_second": 16.726,
736
+ "eval_steps_per_second": 2.091,
737
  "eval_weighted avg": {
738
+ "f1-score": 0.8530843968151707,
739
+ "precision": 0.854049671322198,
740
+ "recall": 0.8522502597097953,
741
+ "support": 29841.0
742
  },
743
  "step": 1296
744
  },
745
  {
746
  "epoch": 17.0,
747
  "eval_Claim": {
748
+ "f1-score": 0.6241452751933639,
749
+ "precision": 0.6379468377635197,
750
+ "recall": 0.6109282422646478,
751
+ "support": 4557.0
752
  },
753
  "eval_MajorClaim": {
754
+ "f1-score": 0.8151592652871426,
755
+ "precision": 0.8626968503937008,
756
+ "recall": 0.7725870427501101,
757
+ "support": 2269.0
758
  },
759
  "eval_O": {
760
+ "f1-score": 0.9064921025764218,
761
+ "precision": 0.9237454100367197,
762
+ "recall": 0.8898714774201155,
763
+ "support": 8481.0
764
  },
765
  "eval_Premise": {
766
+ "f1-score": 0.8993928008319636,
767
+ "precision": 0.8775777414075286,
768
+ "recall": 0.9223200770606853,
769
+ "support": 14534.0
770
  },
771
+ "eval_accuracy": 0.8541603833651688,
772
+ "eval_loss": 0.985948920249939,
773
  "eval_macro avg": {
774
+ "f1-score": 0.8112973609722229,
775
+ "precision": 0.8254917099003672,
776
+ "recall": 0.7989267098738898,
777
+ "support": 29841.0
778
+ },
779
+ "eval_runtime": 4.7769,
780
+ "eval_samples_per_second": 16.747,
781
+ "eval_steps_per_second": 2.093,
782
  "eval_weighted avg": {
783
+ "f1-score": 0.8529727851357218,
784
+ "precision": 0.85297351326598,
785
+ "recall": 0.8541603833651688,
786
+ "support": 29841.0
787
  },
788
  "step": 1377
789
  },
790
  {
791
  "epoch": 18.0,
792
  "eval_Claim": {
793
+ "f1-score": 0.638718890248878,
794
+ "precision": 0.5967219363445778,
795
+ "recall": 0.6870748299319728,
796
+ "support": 4557.0
797
  },
798
  "eval_MajorClaim": {
799
+ "f1-score": 0.778357854769815,
800
+ "precision": 0.8431876606683805,
801
+ "recall": 0.7227853680035258,
802
+ "support": 2269.0
803
  },
804
  "eval_O": {
805
+ "f1-score": 0.9013165698019214,
806
+ "precision": 0.9066937119675457,
807
+ "recall": 0.8960028298549699,
808
+ "support": 8481.0
809
  },
810
  "eval_Premise": {
811
+ "f1-score": 0.8934796194708702,
812
+ "precision": 0.9018082422203533,
813
+ "recall": 0.885303426448328,
814
+ "support": 14534.0
815
  },
816
+ "eval_accuracy": 0.8457156261519386,
817
+ "eval_loss": 0.9986578822135925,
818
  "eval_macro avg": {
819
+ "f1-score": 0.8029682335728712,
820
+ "precision": 0.8121028878002143,
821
+ "recall": 0.7977916135596992,
822
+ "support": 29841.0
823
+ },
824
+ "eval_runtime": 4.7848,
825
+ "eval_samples_per_second": 16.719,
826
+ "eval_steps_per_second": 2.09,
827
  "eval_weighted avg": {
828
+ "f1-score": 0.8480491462490054,
829
+ "precision": 0.8521498954326654,
830
+ "recall": 0.8457156261519386,
831
+ "support": 29841.0
832
  },
833
  "step": 1458
834
  },
835
  {
836
  "epoch": 18.52,
837
+ "grad_norm": 11.437110900878906,
838
  "learning_rate": 1.2592592592592593e-05,
839
+ "loss": 0.0189,
840
  "step": 1500
841
  },
842
  {
843
  "epoch": 19.0,
844
  "eval_Claim": {
845
+ "f1-score": 0.6127094530509011,
846
+ "precision": 0.5894160583941606,
847
+ "recall": 0.6379196840026333,
848
+ "support": 4557.0
849
  },
850
  "eval_MajorClaim": {
851
+ "f1-score": 0.800753453327752,
852
+ "precision": 0.762455161418892,
853
+ "recall": 0.8431026884089907,
854
+ "support": 2269.0
855
  },
856
  "eval_O": {
857
+ "f1-score": 0.8959336706699994,
858
+ "precision": 0.9275435496086847,
859
+ "recall": 0.8664072632944229,
860
+ "support": 8481.0
861
  },
862
  "eval_Premise": {
863
+ "f1-score": 0.8912863642630635,
864
+ "precision": 0.8930100842657825,
865
+ "recall": 0.8895692858125774,
866
+ "support": 14534.0
867
  },
868
+ "eval_accuracy": 0.8410240943668108,
869
+ "eval_loss": 1.0641764402389526,
870
  "eval_macro avg": {
871
+ "f1-score": 0.800170735327929,
872
+ "precision": 0.79310621342188,
873
+ "recall": 0.8092497303796561,
874
+ "support": 29841.0
875
+ },
876
+ "eval_runtime": 4.8048,
877
+ "eval_samples_per_second": 16.65,
878
+ "eval_steps_per_second": 2.081,
879
  "eval_weighted avg": {
880
+ "f1-score": 0.8431820663618932,
881
+ "precision": 0.8465361465202841,
882
+ "recall": 0.8410240943668108,
883
+ "support": 29841.0
884
  },
885
  "step": 1539
886
  },
887
  {
888
  "epoch": 20.0,
889
  "eval_Claim": {
890
+ "f1-score": 0.6346276257161044,
891
+ "precision": 0.6142945163277881,
892
+ "recall": 0.6563528637261357,
893
+ "support": 4557.0
894
  },
895
  "eval_MajorClaim": {
896
+ "f1-score": 0.8171902268760908,
897
+ "precision": 0.8090712742980561,
898
+ "recall": 0.8254737769942706,
899
+ "support": 2269.0
900
  },
901
  "eval_O": {
902
+ "f1-score": 0.9044396961063628,
903
+ "precision": 0.9106011712680769,
904
+ "recall": 0.8983610423299139,
905
+ "support": 8481.0
906
  },
907
  "eval_Premise": {
908
+ "f1-score": 0.894879267277269,
909
+ "precision": 0.902519244226732,
910
+ "recall": 0.8873675519471584,
911
+ "support": 14534.0
912
  },
913
+ "eval_accuracy": 0.8505076907610335,
914
+ "eval_loss": 1.087249994277954,
915
  "eval_macro avg": {
916
+ "f1-score": 0.8127842039939568,
917
+ "precision": 0.8091215515301633,
918
+ "recall": 0.8168888087493696,
919
+ "support": 29841.0
920
+ },
921
+ "eval_runtime": 4.7799,
922
+ "eval_samples_per_second": 16.737,
923
+ "eval_steps_per_second": 2.092,
924
  "eval_weighted avg": {
925
+ "f1-score": 0.85194635060675,
926
+ "precision": 0.8536961248417916,
927
+ "recall": 0.8505076907610335,
928
+ "support": 29841.0
929
  },
930
  "step": 1620
931
  },
932
  {
933
  "epoch": 21.0,
934
  "eval_Claim": {
935
+ "f1-score": 0.6397356640375187,
936
+ "precision": 0.6219689119170985,
937
+ "recall": 0.6585472898836954,
938
+ "support": 4557.0
939
  },
940
  "eval_MajorClaim": {
941
+ "f1-score": 0.8133363574506692,
942
+ "precision": 0.8378504672897197,
943
+ "recall": 0.7902159541648304,
944
+ "support": 2269.0
945
  },
946
  "eval_O": {
947
+ "f1-score": 0.9003324267150197,
948
+ "precision": 0.9236111111111112,
949
+ "recall": 0.8781983256691428,
950
+ "support": 8481.0
951
  },
952
  "eval_Premise": {
953
+ "f1-score": 0.8964765215020788,
954
+ "precision": 0.8880637321091007,
955
+ "recall": 0.9050502270538049,
956
+ "support": 14534.0
957
  },
958
+ "eval_accuracy": 0.8510438658221909,
959
+ "eval_loss": 1.093902826309204,
960
  "eval_macro avg": {
961
+ "f1-score": 0.8124702424263215,
962
+ "precision": 0.8178735556067575,
963
+ "recall": 0.8080029491928683,
964
+ "support": 29841.0
965
+ },
966
+ "eval_runtime": 4.813,
967
+ "eval_samples_per_second": 16.622,
968
+ "eval_steps_per_second": 2.078,
969
  "eval_weighted avg": {
970
+ "f1-score": 0.8520439894961911,
971
+ "precision": 0.8537133191814482,
972
+ "recall": 0.8510438658221909,
973
+ "support": 29841.0
974
  },
975
  "step": 1701
976
  },
977
  {
978
  "epoch": 22.0,
979
  "eval_Claim": {
980
+ "f1-score": 0.6279694722132645,
981
+ "precision": 0.6154656552886641,
982
+ "recall": 0.640991880623217,
983
+ "support": 4557.0
984
  },
985
  "eval_MajorClaim": {
986
+ "f1-score": 0.8173439463762043,
987
+ "precision": 0.7788423153692615,
988
+ "recall": 0.8598501542529748,
989
+ "support": 2269.0
990
  },
991
  "eval_O": {
992
+ "f1-score": 0.9038883941822499,
993
+ "precision": 0.910210425633668,
994
+ "recall": 0.8976535785874308,
995
+ "support": 8481.0
996
  },
997
  "eval_Premise": {
998
+ "f1-score": 0.8954102920723226,
999
+ "precision": 0.9051033319274567,
1000
+ "recall": 0.8859226640979772,
1001
+ "support": 14534.0
1002
  },
1003
+ "eval_accuracy": 0.849870982875909,
1004
+ "eval_loss": 1.0930607318878174,
1005
  "eval_macro avg": {
1006
+ "f1-score": 0.8111530262110103,
1007
+ "precision": 0.8024054320547626,
1008
+ "recall": 0.8211045693904,
1009
+ "support": 29841.0
1010
+ },
1011
+ "eval_runtime": 4.8038,
1012
+ "eval_samples_per_second": 16.654,
1013
+ "eval_steps_per_second": 2.082,
1014
  "eval_weighted avg": {
1015
+ "f1-score": 0.8510432276144315,
1016
+ "precision": 0.8527239921837771,
1017
+ "recall": 0.849870982875909,
1018
+ "support": 29841.0
1019
  },
1020
  "step": 1782
1021
  },
1022
  {
1023
  "epoch": 23.0,
1024
  "eval_Claim": {
1025
+ "f1-score": 0.6225736879942487,
1026
+ "precision": 0.5851351351351352,
1027
+ "recall": 0.6651305683563749,
1028
+ "support": 4557.0
1029
  },
1030
  "eval_MajorClaim": {
1031
+ "f1-score": 0.8058810995099083,
1032
+ "precision": 0.7801155115511551,
1033
+ "recall": 0.8334067871308947,
1034
+ "support": 2269.0
1035
  },
1036
  "eval_O": {
1037
+ "f1-score": 0.8976647945610404,
1038
+ "precision": 0.9001659947830211,
1039
+ "recall": 0.8951774554887395,
1040
+ "support": 8481.0
1041
  },
1042
  "eval_Premise": {
1043
+ "f1-score": 0.8881674136288245,
1044
+ "precision": 0.9116858653915816,
1045
+ "recall": 0.8658318425760286,
1046
+ "support": 14534.0
1047
  },
1048
+ "eval_accuracy": 0.8410576053081331,
1049
+ "eval_loss": 1.1820006370544434,
1050
  "eval_macro avg": {
1051
+ "f1-score": 0.8035717489235055,
1052
+ "precision": 0.7942756267152233,
1053
+ "recall": 0.8148866633880094,
1054
+ "support": 29841.0
1055
+ },
1056
+ "eval_runtime": 4.7901,
1057
+ "eval_samples_per_second": 16.701,
1058
+ "eval_steps_per_second": 2.088,
1059
  "eval_weighted avg": {
1060
+ "f1-score": 0.8440512323089471,
1061
+ "precision": 0.8485403664715132,
1062
+ "recall": 0.8410576053081331,
1063
+ "support": 29841.0
1064
  },
1065
  "step": 1863
1066
  },
1067
  {
1068
  "epoch": 24.0,
1069
  "eval_Claim": {
1070
+ "f1-score": 0.6244661044792466,
1071
+ "precision": 0.6233056405771753,
1072
+ "recall": 0.6256308975202984,
1073
+ "support": 4557.0
1074
  },
1075
  "eval_MajorClaim": {
1076
+ "f1-score": 0.8177083333333335,
1077
+ "precision": 0.8054724241128688,
1078
+ "recall": 0.8303217276333187,
1079
+ "support": 2269.0
1080
  },
1081
  "eval_O": {
1082
+ "f1-score": 0.9007499414108273,
1083
+ "precision": 0.8951904040992198,
1084
+ "recall": 0.9063789647447235,
1085
+ "support": 8481.0
1086
  },
1087
  "eval_Premise": {
1088
+ "f1-score": 0.8948225108225109,
1089
+ "precision": 0.9008437347465309,
1090
+ "recall": 0.888881243979634,
1091
+ "support": 14534.0
1092
  },
1093
+ "eval_accuracy": 0.8492007640494621,
1094
+ "eval_loss": 1.1699731349945068,
1095
  "eval_macro avg": {
1096
+ "f1-score": 0.8094367225114796,
1097
+ "precision": 0.8062030508839487,
1098
+ "recall": 0.8128032084694936,
1099
+ "support": 29841.0
1100
+ },
1101
+ "eval_runtime": 4.802,
1102
+ "eval_samples_per_second": 16.66,
1103
+ "eval_steps_per_second": 2.082,
1104
  "eval_weighted avg": {
1105
+ "f1-score": 0.8493576914930753,
1106
+ "precision": 0.8496026739182283,
1107
+ "recall": 0.8492007640494621,
1108
+ "support": 29841.0
1109
  },
1110
  "step": 1944
1111
  },
1112
  {
1113
  "epoch": 24.69,
1114
+ "grad_norm": 9.07358169555664,
1115
  "learning_rate": 1.0123456790123458e-05,
1116
+ "loss": 0.006,
1117
  "step": 2000
1118
  },
1119
  {
1120
  "epoch": 25.0,
1121
  "eval_Claim": {
1122
+ "f1-score": 0.622420524261015,
1123
+ "precision": 0.632940108892922,
1124
+ "recall": 0.6122448979591837,
1125
+ "support": 4557.0
1126
  },
1127
  "eval_MajorClaim": {
1128
+ "f1-score": 0.8149307107733571,
1129
+ "precision": 0.8267573696145125,
1130
+ "recall": 0.8034376377258704,
1131
+ "support": 2269.0
1132
  },
1133
  "eval_O": {
1134
+ "f1-score": 0.9018776855612456,
1135
+ "precision": 0.9004466384579219,
1136
+ "recall": 0.9033132885272963,
1137
+ "support": 8481.0
1138
  },
1139
  "eval_Premise": {
1140
+ "f1-score": 0.8975866548164354,
1141
+ "precision": 0.8919157608695653,
1142
+ "recall": 0.9033301224714463,
1143
+ "support": 14534.0
1144
  },
1145
+ "eval_accuracy": 0.8512784424114473,
1146
+ "eval_loss": 1.1391637325286865,
1147
  "eval_macro avg": {
1148
+ "f1-score": 0.8092038938530133,
1149
+ "precision": 0.8130149694587304,
1150
+ "recall": 0.8055814866709491,
1151
+ "support": 29841.0
1152
+ },
1153
+ "eval_runtime": 4.83,
1154
+ "eval_samples_per_second": 16.563,
1155
+ "eval_steps_per_second": 2.07,
1156
  "eval_weighted avg": {
1157
+ "f1-score": 0.8505008948811765,
1158
+ "precision": 0.8498378793311306,
1159
+ "recall": 0.8512784424114473,
1160
+ "support": 29841.0
1161
  },
1162
  "step": 2025
1163
  },
1164
  {
1165
  "epoch": 26.0,
1166
  "eval_Claim": {
1167
+ "f1-score": 0.6216186854970125,
1168
+ "precision": 0.6155335628227194,
1169
+ "recall": 0.6278253236778583,
1170
+ "support": 4557.0
1171
  },
1172
  "eval_MajorClaim": {
1173
+ "f1-score": 0.8021786492374727,
1174
+ "precision": 0.7931925894011203,
1175
+ "recall": 0.8113706478624945,
1176
+ "support": 2269.0
1177
  },
1178
  "eval_O": {
1179
+ "f1-score": 0.9000058854687776,
1180
+ "precision": 0.8984723854289072,
1181
+ "recall": 0.9015446291710884,
1182
+ "support": 8481.0
1183
  },
1184
  "eval_Premise": {
1185
+ "f1-score": 0.8959025470653377,
1186
+ "precision": 0.9012672329759086,
1187
+ "recall": 0.8906013485619926,
1188
+ "support": 14534.0
1189
  },
1190
+ "eval_accuracy": 0.8475587279246674,
1191
+ "eval_loss": 1.18631911277771,
1192
  "eval_macro avg": {
1193
+ "f1-score": 0.80492644181715,
1194
+ "precision": 0.8021164426571639,
1195
+ "recall": 0.8078354873183584,
1196
+ "support": 29841.0
1197
+ },
1198
+ "eval_runtime": 4.7928,
1199
+ "eval_samples_per_second": 16.692,
1200
+ "eval_steps_per_second": 2.086,
1201
  "eval_weighted avg": {
1202
+ "f1-score": 0.8480566079768785,
1203
+ "precision": 0.848621115111045,
1204
+ "recall": 0.8475587279246674,
1205
+ "support": 29841.0
1206
  },
1207
  "step": 2106
1208
  },
1209
  {
1210
  "epoch": 27.0,
1211
  "eval_Claim": {
1212
+ "f1-score": 0.6274287012839449,
1213
+ "precision": 0.6505655042412818,
1214
+ "recall": 0.6058810621022602,
1215
+ "support": 4557.0
1216
  },
1217
  "eval_MajorClaim": {
1218
+ "f1-score": 0.8276779773785763,
1219
+ "precision": 0.8330357142857143,
1220
+ "recall": 0.8223887174966946,
1221
+ "support": 2269.0
1222
  },
1223
  "eval_O": {
1224
+ "f1-score": 0.9055736912102527,
1225
+ "precision": 0.9291651159905719,
1226
+ "recall": 0.8831505718665251,
1227
+ "support": 8481.0
1228
  },
1229
  "eval_Premise": {
1230
+ "f1-score": 0.8996312437143815,
1231
+ "precision": 0.8772228033472803,
1232
+ "recall": 0.9232145314435117,
1233
+ "support": 14534.0
1234
  },
1235
+ "eval_accuracy": 0.8557018866659964,
1236
+ "eval_loss": 1.173492431640625,
1237
  "eval_macro avg": {
1238
+ "f1-score": 0.8150779033967888,
1239
+ "precision": 0.8224972844662122,
1240
+ "recall": 0.8086587207272479,
1241
+ "support": 29841.0
1242
+ },
1243
+ "eval_runtime": 4.7958,
1244
+ "eval_samples_per_second": 16.681,
1245
+ "eval_steps_per_second": 2.085,
1246
  "eval_weighted avg": {
1247
+ "f1-score": 0.8542811867471567,
1248
+ "precision": 0.8540126205927153,
1249
+ "recall": 0.8557018866659964,
1250
+ "support": 29841.0
1251
  },
1252
  "step": 2187
1253
  }
 
1257
  "num_input_tokens_seen": 0,
1258
  "num_train_epochs": 50,
1259
  "save_steps": 500,
1260
+ "total_flos": 3882295783623600.0,
1261
  "train_batch_size": 4,
1262
  "trial_name": null,
1263
  "trial_params": null
checkpoint-2187/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0ce83aa27c1a93c128c1dae6a1646206ab5d85d935efad925ac4e4beb0be29a0
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19f05e33d3889e7c1ce1bf18bf9e16659f52e74109a304d618b5d09955736add
3
  size 4856