Theoreticallyhugo commited on
Commit
28ef94f
1 Parent(s): ee8099e

Training in progress, epoch 13, checkpoint

Browse files
checkpoint-1053/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:df26ad71e6b9c578551fbade9cd231a514e617764710b9474de3c64dd997ae2f
3
  size 592330980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61ac4bd406da0e6701da1c3e5a63687ac74e6c6452c4399caa4763fe57957f1b
3
  size 592330980
checkpoint-1053/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0757064f821b5ebebf263ce2fc7cfa9511bd0f50b5dde8b3ee62cc9260c22c12
3
  size 1014670074
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e769f065859c4de2bde0aeeaaca7731b556384df8690576047dbe44e2395c177
3
  size 1014670074
checkpoint-1053/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:785282e90233e50783d511149f52a487e725aef0302182724c574e3af2083098
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a92327c1c7423b14915d094fccec26dc0b7df7db995e9e1e5592edda9350390
3
  size 1064
checkpoint-1053/trainer_state.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
  "epoch": 13.0,
5
  "eval_steps": 500,
6
  "global_step": 1053,
@@ -11,841 +11,841 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
- "f1-score": 0.37160751565762,
15
- "precision": 0.42788461538461536,
16
- "recall": 0.3284132841328413,
17
- "support": 271.0
18
  },
19
  "eval_B-MajorClaim": {
20
- "f1-score": 0.3578947368421052,
21
- "precision": 0.6666666666666666,
22
- "recall": 0.2446043165467626,
23
- "support": 139.0
24
  },
25
  "eval_B-Premise": {
26
- "f1-score": 0.8640915593705293,
27
- "precision": 0.7895424836601307,
28
- "recall": 0.9541864139020537,
29
- "support": 633.0
30
  },
31
  "eval_I-Claim": {
32
- "f1-score": 0.5003402749421533,
33
- "precision": 0.5493126120741183,
34
- "recall": 0.4593851537115721,
35
- "support": 4001.0
36
  },
37
  "eval_I-MajorClaim": {
38
- "f1-score": 0.7718093699515347,
39
- "precision": 0.6502211636611093,
40
- "recall": 0.9493293591654247,
41
- "support": 2013.0
42
  },
43
  "eval_I-Premise": {
44
- "f1-score": 0.875016720916752,
45
- "precision": 0.8846812731043188,
46
- "recall": 0.865561044460127,
47
- "support": 11336.0
48
  },
49
  "eval_O": {
50
- "f1-score": 0.9992483530087988,
51
- "precision": 0.9995577178239717,
52
- "recall": 0.998939179632249,
53
- "support": 11312.0
54
  },
55
- "eval_accuracy": 0.8614038040733883,
56
- "eval_loss": 0.31713685393333435,
57
  "eval_macro avg": {
58
- "f1-score": 0.6771440758127848,
59
- "precision": 0.7096952189107044,
60
- "recall": 0.685774107364433,
61
- "support": 29705.0
62
- },
63
- "eval_runtime": 4.8338,
64
- "eval_samples_per_second": 16.55,
65
- "eval_steps_per_second": 2.069,
66
  "eval_weighted avg": {
67
- "f1-score": 0.8576207231627551,
68
- "precision": 0.8601529227027923,
69
- "recall": 0.8614038040733883,
70
- "support": 29705.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
- "f1-score": 0.4708624708624709,
78
- "precision": 0.6392405063291139,
79
- "recall": 0.3726937269372694,
80
- "support": 271.0
81
  },
82
  "eval_B-MajorClaim": {
83
- "f1-score": 0.796875,
84
- "precision": 0.8717948717948718,
85
- "recall": 0.7338129496402878,
86
- "support": 139.0
87
  },
88
  "eval_B-Premise": {
89
- "f1-score": 0.8736616702355461,
90
- "precision": 0.796875,
91
- "recall": 0.966824644549763,
92
- "support": 633.0
93
  },
94
  "eval_I-Claim": {
95
- "f1-score": 0.5100589925881107,
96
- "precision": 0.6459770114942529,
97
- "recall": 0.4213946513371657,
98
- "support": 4001.0
99
  },
100
  "eval_I-MajorClaim": {
101
- "f1-score": 0.8401387776888176,
102
- "precision": 0.9077277970011534,
103
- "recall": 0.7819175360158966,
104
- "support": 2013.0
105
  },
106
  "eval_I-Premise": {
107
- "f1-score": 0.8912891699864469,
108
- "precision": 0.8338584492430646,
109
- "recall": 0.9572159491884262,
110
- "support": 11336.0
111
  },
112
  "eval_O": {
113
- "f1-score": 0.9996904982977407,
114
- "precision": 1.0,
115
- "recall": 0.9993811881188119,
116
- "support": 11312.0
117
  },
118
- "eval_accuracy": 0.8830499915839084,
119
- "eval_loss": 0.2966194748878479,
120
  "eval_macro avg": {
121
- "f1-score": 0.7689395113798762,
122
- "precision": 0.8136390908374939,
123
- "recall": 0.7476058065410885,
124
- "support": 29705.0
125
- },
126
- "eval_runtime": 4.8625,
127
- "eval_samples_per_second": 16.452,
128
- "eval_steps_per_second": 2.057,
129
  "eval_weighted avg": {
130
- "f1-score": 0.8731020208182413,
131
- "precision": 0.874440834821272,
132
- "recall": 0.8830499915839084,
133
- "support": 29705.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
- "f1-score": 0.6085192697768763,
141
- "precision": 0.6756756756756757,
142
- "recall": 0.5535055350553506,
143
- "support": 271.0
144
  },
145
  "eval_B-MajorClaim": {
146
- "f1-score": 0.8571428571428571,
147
- "precision": 0.851063829787234,
148
- "recall": 0.8633093525179856,
149
- "support": 139.0
150
  },
151
  "eval_B-Premise": {
152
- "f1-score": 0.8834729626808834,
153
- "precision": 0.8529411764705882,
154
- "recall": 0.9162717219589257,
155
- "support": 633.0
156
  },
157
  "eval_I-Claim": {
158
- "f1-score": 0.5764474423833614,
159
- "precision": 0.6584269662921348,
160
- "recall": 0.5126218445388653,
161
- "support": 4001.0
162
  },
163
  "eval_I-MajorClaim": {
164
- "f1-score": 0.8581151832460733,
165
- "precision": 0.9070282235749861,
166
- "recall": 0.8142076502732241,
167
- "support": 2013.0
168
  },
169
  "eval_I-Premise": {
170
- "f1-score": 0.8959744247675935,
171
- "precision": 0.8563158317922328,
172
- "recall": 0.939484827099506,
173
- "support": 11336.0
174
  },
175
  "eval_O": {
176
- "f1-score": 0.9996020340481981,
177
- "precision": 1.0,
178
- "recall": 0.9992043847241867,
179
- "support": 11312.0
180
  },
181
- "eval_accuracy": 0.8918700555462044,
182
- "eval_loss": 0.2552729547023773,
183
  "eval_macro avg": {
184
- "f1-score": 0.811324882006549,
185
- "precision": 0.8287788147989789,
186
- "recall": 0.7998007594525776,
187
- "support": 29705.0
188
- },
189
- "eval_runtime": 4.8422,
190
- "eval_samples_per_second": 16.522,
191
- "eval_steps_per_second": 2.065,
192
  "eval_weighted avg": {
193
- "f1-score": 0.8867633844066056,
194
- "precision": 0.886070631898416,
195
- "recall": 0.8918700555462044,
196
- "support": 29705.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
- "f1-score": 0.6722408026755852,
204
- "precision": 0.6146788990825688,
205
- "recall": 0.7416974169741697,
206
- "support": 271.0
207
  },
208
  "eval_B-MajorClaim": {
209
- "f1-score": 0.8664259927797834,
210
- "precision": 0.8695652173913043,
211
- "recall": 0.8633093525179856,
212
- "support": 139.0
213
  },
214
  "eval_B-Premise": {
215
- "f1-score": 0.8687035507844755,
216
- "precision": 0.9100346020761245,
217
- "recall": 0.8309636650868878,
218
- "support": 633.0
219
  },
220
  "eval_I-Claim": {
221
- "f1-score": 0.6483151400094921,
222
- "precision": 0.6171222046532641,
223
- "recall": 0.6828292926768308,
224
- "support": 4001.0
225
  },
226
  "eval_I-MajorClaim": {
227
- "f1-score": 0.8696993060909791,
228
- "precision": 0.9009584664536742,
229
- "recall": 0.8405365126676602,
230
- "support": 2013.0
231
  },
232
  "eval_I-Premise": {
233
- "f1-score": 0.8906159274643798,
234
- "precision": 0.9020175517958925,
235
- "recall": 0.879498941425547,
236
- "support": 11336.0
237
  },
238
  "eval_O": {
239
- "f1-score": 0.999557835160948,
240
- "precision": 0.9999115357395613,
241
- "recall": 0.9992043847241867,
242
- "support": 11312.0
243
  },
244
- "eval_accuracy": 0.8935869382258879,
245
- "eval_loss": 0.29465603828430176,
246
  "eval_macro avg": {
247
- "f1-score": 0.8307940792808061,
248
- "precision": 0.8306126395989127,
249
- "recall": 0.8340056522961811,
250
- "support": 29705.0
251
- },
252
- "eval_runtime": 4.8494,
253
- "eval_samples_per_second": 16.497,
254
- "eval_steps_per_second": 2.062,
255
  "eval_weighted avg": {
256
- "f1-score": 0.8954766464091573,
257
- "precision": 0.8982496227307208,
258
- "recall": 0.8935869382258879,
259
- "support": 29705.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
- "f1-score": 0.6438095238095239,
267
- "precision": 0.6653543307086615,
268
- "recall": 0.6236162361623616,
269
- "support": 271.0
270
  },
271
  "eval_B-MajorClaim": {
272
- "f1-score": 0.8785714285714286,
273
- "precision": 0.8723404255319149,
274
- "recall": 0.8848920863309353,
275
- "support": 139.0
276
  },
277
  "eval_B-Premise": {
278
- "f1-score": 0.8783151326053042,
279
- "precision": 0.8674884437596302,
280
- "recall": 0.8894154818325435,
281
- "support": 633.0
282
  },
283
  "eval_I-Claim": {
284
- "f1-score": 0.6283729628640129,
285
- "precision": 0.6748923959827834,
286
- "recall": 0.5878530367408148,
287
- "support": 4001.0
288
  },
289
  "eval_I-MajorClaim": {
290
- "f1-score": 0.885450461692039,
291
- "precision": 0.8896690070210632,
292
- "recall": 0.8812717337307501,
293
- "support": 2013.0
294
  },
295
  "eval_I-Premise": {
296
- "f1-score": 0.8989315871101154,
297
- "precision": 0.878494442573257,
298
- "recall": 0.9203422724064926,
299
- "support": 11336.0
300
  },
301
  "eval_O": {
302
- "f1-score": 0.9997347245556637,
303
- "precision": 1.0,
304
- "recall": 0.9994695898161244,
305
- "support": 11312.0
306
  },
307
- "eval_accuracy": 0.8995118666891095,
308
- "eval_loss": 0.3176642060279846,
309
  "eval_macro avg": {
310
- "f1-score": 0.830455117315441,
311
- "precision": 0.8354627207967587,
312
- "recall": 0.8266943481457174,
313
- "support": 29705.0
314
- },
315
- "eval_runtime": 4.8682,
316
- "eval_samples_per_second": 16.433,
317
- "eval_steps_per_second": 2.054,
318
  "eval_weighted avg": {
319
- "f1-score": 0.8971010593476484,
320
- "precision": 0.8958911872123141,
321
- "recall": 0.8995118666891095,
322
- "support": 29705.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
- "f1-score": 0.6150712830957231,
330
- "precision": 0.6863636363636364,
331
- "recall": 0.5571955719557196,
332
- "support": 271.0
333
  },
334
  "eval_B-MajorClaim": {
335
- "f1-score": 0.8686131386861314,
336
- "precision": 0.8814814814814815,
337
- "recall": 0.8561151079136691,
338
- "support": 139.0
339
  },
340
  "eval_B-Premise": {
341
- "f1-score": 0.8804841149773072,
342
- "precision": 0.8447024673439768,
343
- "recall": 0.919431279620853,
344
- "support": 633.0
345
  },
346
  "eval_I-Claim": {
347
- "f1-score": 0.5758052970651396,
348
- "precision": 0.6739276139410187,
349
- "recall": 0.5026243439140214,
350
- "support": 4001.0
351
  },
352
  "eval_I-MajorClaim": {
353
- "f1-score": 0.8627552339105711,
354
- "precision": 0.8992456896551724,
355
- "recall": 0.829110779930452,
356
- "support": 2013.0
357
  },
358
  "eval_I-Premise": {
359
- "f1-score": 0.894107779408681,
360
- "precision": 0.8521864257734432,
361
- "recall": 0.9403669724770642,
362
- "support": 11336.0
363
  },
364
  "eval_O": {
365
- "f1-score": 1.0,
366
- "precision": 1.0,
367
- "recall": 1.0,
368
- "support": 11312.0
369
  },
370
- "eval_accuracy": 0.8922403635751557,
371
- "eval_loss": 0.41071853041648865,
372
  "eval_macro avg": {
373
- "f1-score": 0.8138338353062219,
374
- "precision": 0.8339867592226755,
375
- "recall": 0.8006920079731114,
376
- "support": 29705.0
377
- },
378
- "eval_runtime": 4.8571,
379
- "eval_samples_per_second": 16.471,
380
- "eval_steps_per_second": 2.059,
381
  "eval_weighted avg": {
382
- "f1-score": 0.8864802913843919,
383
- "precision": 0.8861194550557427,
384
- "recall": 0.8922403635751557,
385
- "support": 29705.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
- "grad_norm": 5.239528656005859,
392
- "learning_rate": 1.7530864197530865e-05,
393
- "loss": 0.2363,
394
  "step": 500
395
  },
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
- "f1-score": 0.6954954954954955,
400
- "precision": 0.6795774647887324,
401
- "recall": 0.7121771217712177,
402
- "support": 271.0
403
  },
404
  "eval_B-MajorClaim": {
405
- "f1-score": 0.8920863309352518,
406
- "precision": 0.8920863309352518,
407
- "recall": 0.8920863309352518,
408
- "support": 139.0
409
  },
410
  "eval_B-Premise": {
411
- "f1-score": 0.8876494023904382,
412
- "precision": 0.8954983922829582,
413
- "recall": 0.8799368088467614,
414
- "support": 633.0
415
  },
416
  "eval_I-Claim": {
417
- "f1-score": 0.6831545741324923,
418
- "precision": 0.6898572884811417,
419
- "recall": 0.6765808547863035,
420
- "support": 4001.0
421
  },
422
  "eval_I-MajorClaim": {
423
- "f1-score": 0.8984707946853848,
424
- "precision": 0.9068825910931174,
425
- "recall": 0.8902136115250869,
426
- "support": 2013.0
427
  },
428
  "eval_I-Premise": {
429
- "f1-score": 0.9068164859763859,
430
- "precision": 0.9024198479951079,
431
- "recall": 0.9112561750176429,
432
- "support": 11336.0
433
  },
434
  "eval_O": {
435
- "f1-score": 0.9999558011049724,
436
- "precision": 0.9999116061168567,
437
- "recall": 1.0,
438
- "support": 11312.0
439
  },
440
- "eval_accuracy": 0.9094428547382596,
441
- "eval_loss": 0.3520519435405731,
442
  "eval_macro avg": {
443
- "f1-score": 0.8519469835314888,
444
- "precision": 0.8523190745275951,
445
- "recall": 0.8517501289831806,
446
- "support": 29705.0
447
- },
448
- "eval_runtime": 4.8905,
449
- "eval_samples_per_second": 16.358,
450
- "eval_steps_per_second": 2.045,
451
  "eval_weighted avg": {
452
- "f1-score": 0.9091888981291354,
453
- "precision": 0.908989097041669,
454
- "recall": 0.9094428547382596,
455
- "support": 29705.0
456
  },
457
  "step": 567
458
  },
459
  {
460
  "epoch": 8.0,
461
  "eval_B-Claim": {
462
- "f1-score": 0.6666666666666667,
463
- "precision": 0.658273381294964,
464
- "recall": 0.6752767527675276,
465
- "support": 271.0
466
  },
467
  "eval_B-MajorClaim": {
468
- "f1-score": 0.8819444444444444,
469
- "precision": 0.8523489932885906,
470
- "recall": 0.9136690647482014,
471
- "support": 139.0
472
  },
473
  "eval_B-Premise": {
474
- "f1-score": 0.8759007205764612,
475
- "precision": 0.887987012987013,
476
- "recall": 0.8641390205371248,
477
- "support": 633.0
478
  },
479
  "eval_I-Claim": {
480
- "f1-score": 0.642274412855377,
481
- "precision": 0.6353631694790902,
482
- "recall": 0.6493376655836041,
483
- "support": 4001.0
484
  },
485
  "eval_I-MajorClaim": {
486
- "f1-score": 0.87683284457478,
487
- "precision": 0.862914862914863,
488
- "recall": 0.8912071535022354,
489
- "support": 2013.0
490
  },
491
  "eval_I-Premise": {
492
- "f1-score": 0.8911585772014743,
493
- "precision": 0.8972547616918537,
494
- "recall": 0.8851446718419196,
495
- "support": 11336.0
496
  },
497
  "eval_O": {
498
- "f1-score": 0.9998673915926268,
499
- "precision": 0.9999115904871364,
500
- "recall": 0.9998231966053748,
501
- "support": 11312.0
502
  },
503
- "eval_accuracy": 0.8952364921730348,
504
- "eval_loss": 0.4631531238555908,
505
  "eval_macro avg": {
506
- "f1-score": 0.8335207225588329,
507
- "precision": 0.827721967449073,
508
- "recall": 0.8397996465122839,
509
- "support": 29705.0
510
- },
511
- "eval_runtime": 4.8517,
512
- "eval_samples_per_second": 16.489,
513
- "eval_steps_per_second": 2.061,
514
  "eval_weighted avg": {
515
- "f1-score": 0.8956465277692122,
516
- "precision": 0.8961582200263131,
517
- "recall": 0.8952364921730348,
518
- "support": 29705.0
519
  },
520
  "step": 648
521
  },
522
  {
523
  "epoch": 9.0,
524
  "eval_B-Claim": {
525
- "f1-score": 0.6604127579737337,
526
- "precision": 0.6717557251908397,
527
- "recall": 0.6494464944649446,
528
- "support": 271.0
529
  },
530
  "eval_B-MajorClaim": {
531
- "f1-score": 0.8661417322834645,
532
- "precision": 0.9565217391304348,
533
- "recall": 0.7913669064748201,
534
- "support": 139.0
535
  },
536
  "eval_B-Premise": {
537
- "f1-score": 0.8852963818321786,
538
- "precision": 0.8633633633633634,
539
- "recall": 0.9083728278041074,
540
- "support": 633.0
541
  },
542
  "eval_I-Claim": {
543
- "f1-score": 0.6148568412719356,
544
- "precision": 0.6512017887087759,
545
- "recall": 0.5823544113971507,
546
- "support": 4001.0
547
  },
548
  "eval_I-MajorClaim": {
549
- "f1-score": 0.8500414250207126,
550
- "precision": 0.957089552238806,
551
- "recall": 0.7645305514157973,
552
- "support": 2013.0
553
  },
554
  "eval_I-Premise": {
555
- "f1-score": 0.8985408601693112,
556
- "precision": 0.8677183468901487,
557
- "recall": 0.9316337332392378,
558
- "support": 11336.0
559
  },
560
  "eval_O": {
561
- "f1-score": 0.9995136401821638,
562
- "precision": 0.9998230871295887,
563
- "recall": 0.9992043847241867,
564
- "support": 11312.0
565
  },
566
- "eval_accuracy": 0.8952701565393032,
567
- "eval_loss": 0.5578464865684509,
568
  "eval_macro avg": {
569
- "f1-score": 0.8249719483905,
570
- "precision": 0.8524962289502797,
571
- "recall": 0.8038441870743206,
572
- "support": 29705.0
573
- },
574
- "eval_runtime": 4.8325,
575
- "eval_samples_per_second": 16.555,
576
- "eval_steps_per_second": 2.069,
577
  "eval_weighted avg": {
578
- "f1-score": 0.8928897918536849,
579
- "precision": 0.8934537922744324,
580
- "recall": 0.8952701565393032,
581
- "support": 29705.0
582
  },
583
  "step": 729
584
  },
585
  {
586
  "epoch": 10.0,
587
  "eval_B-Claim": {
588
- "f1-score": 0.6775956284153005,
589
- "precision": 0.6690647482014388,
590
- "recall": 0.6863468634686347,
591
- "support": 271.0
592
  },
593
  "eval_B-MajorClaim": {
594
- "f1-score": 0.8832116788321168,
595
- "precision": 0.8962962962962963,
596
- "recall": 0.8705035971223022,
597
- "support": 139.0
598
  },
599
  "eval_B-Premise": {
600
- "f1-score": 0.8820269200316706,
601
- "precision": 0.8841269841269841,
602
- "recall": 0.8799368088467614,
603
- "support": 633.0
604
  },
605
  "eval_I-Claim": {
606
- "f1-score": 0.6348430262480701,
607
- "precision": 0.6542031291434632,
608
- "recall": 0.6165958510372407,
609
- "support": 4001.0
610
  },
611
  "eval_I-MajorClaim": {
612
- "f1-score": 0.8864321608040201,
613
- "precision": 0.896797153024911,
614
- "recall": 0.8763040238450075,
615
- "support": 2013.0
616
  },
617
  "eval_I-Premise": {
618
- "f1-score": 0.8934983440822729,
619
- "precision": 0.8828797795384086,
620
- "recall": 0.9043754410726887,
621
- "support": 11336.0
622
  },
623
  "eval_O": {
624
- "f1-score": 1.0,
625
- "precision": 1.0,
626
- "recall": 1.0,
627
- "support": 11312.0
628
  },
629
- "eval_accuracy": 0.897458340346743,
630
- "eval_loss": 0.6301500201225281,
631
  "eval_macro avg": {
632
- "f1-score": 0.8368011083447787,
633
- "precision": 0.8404811557616432,
634
- "recall": 0.8334375121989479,
635
- "support": 29705.0
636
- },
637
- "eval_runtime": 4.8426,
638
- "eval_samples_per_second": 16.52,
639
- "eval_steps_per_second": 2.065,
640
  "eval_weighted avg": {
641
- "f1-score": 0.8964756773362035,
642
- "precision": 0.895761611933671,
643
- "recall": 0.897458340346743,
644
- "support": 29705.0
645
  },
646
  "step": 810
647
  },
648
  {
649
  "epoch": 11.0,
650
  "eval_B-Claim": {
651
- "f1-score": 0.6483300589390962,
652
- "precision": 0.6932773109243697,
653
- "recall": 0.6088560885608856,
654
- "support": 271.0
655
  },
656
  "eval_B-MajorClaim": {
657
- "f1-score": 0.8805970149253732,
658
- "precision": 0.9147286821705426,
659
- "recall": 0.8489208633093526,
660
- "support": 139.0
661
  },
662
  "eval_B-Premise": {
663
- "f1-score": 0.8848207475209764,
664
- "precision": 0.855457227138643,
665
- "recall": 0.9162717219589257,
666
- "support": 633.0
667
  },
668
  "eval_I-Claim": {
669
- "f1-score": 0.6216666666666667,
670
- "precision": 0.6995936230071897,
671
- "recall": 0.55936015996001,
672
- "support": 4001.0
673
  },
674
  "eval_I-MajorClaim": {
675
- "f1-score": 0.8927083333333333,
676
- "precision": 0.9381499726327313,
677
- "recall": 0.851465474416294,
678
- "support": 2013.0
679
  },
680
  "eval_I-Premise": {
681
- "f1-score": 0.9015977681968045,
682
- "precision": 0.8655250770978737,
683
- "recall": 0.9408080451658434,
684
- "support": 11336.0
685
  },
686
  "eval_O": {
687
- "f1-score": 1.0,
688
- "precision": 1.0,
689
- "recall": 1.0,
690
- "support": 11312.0
691
  },
692
- "eval_accuracy": 0.9019357010604275,
693
- "eval_loss": 0.5967662334442139,
694
  "eval_macro avg": {
695
- "f1-score": 0.8328172270831785,
696
- "precision": 0.8523902704244785,
697
- "recall": 0.8179546219101873,
698
- "support": 29705.0
699
- },
700
- "eval_runtime": 4.8502,
701
- "eval_samples_per_second": 16.494,
702
- "eval_steps_per_second": 2.062,
703
  "eval_weighted avg": {
704
- "f1-score": 0.8979974574260295,
705
- "precision": 0.8977509549706758,
706
- "recall": 0.9019357010604275,
707
- "support": 29705.0
708
  },
709
  "step": 891
710
  },
711
  {
712
  "epoch": 12.0,
713
  "eval_B-Claim": {
714
- "f1-score": 0.6921898928024502,
715
- "precision": 0.5916230366492147,
716
- "recall": 0.8339483394833949,
717
- "support": 271.0
718
  },
719
  "eval_B-MajorClaim": {
720
- "f1-score": 0.8978102189781022,
721
- "precision": 0.9111111111111111,
722
- "recall": 0.8848920863309353,
723
- "support": 139.0
724
  },
725
  "eval_B-Premise": {
726
- "f1-score": 0.8482758620689655,
727
- "precision": 0.9335863377609108,
728
- "recall": 0.7772511848341233,
729
- "support": 633.0
730
  },
731
  "eval_I-Claim": {
732
- "f1-score": 0.6750635055038103,
733
- "precision": 0.5854598861758766,
734
- "recall": 0.7970507373156711,
735
- "support": 4001.0
736
  },
737
  "eval_I-MajorClaim": {
738
- "f1-score": 0.8948170731707318,
739
- "precision": 0.9157566302652106,
740
- "recall": 0.8748137108792846,
741
- "support": 2013.0
742
  },
743
  "eval_I-Premise": {
744
- "f1-score": 0.8742200328407225,
745
- "precision": 0.9336606874436316,
746
- "recall": 0.821894848270995,
747
- "support": 11336.0
748
  },
749
  "eval_O": {
750
- "f1-score": 1.0,
751
- "precision": 1.0,
752
- "recall": 1.0,
753
- "support": 11312.0
754
  },
755
- "eval_accuracy": 0.8894125568086181,
756
- "eval_loss": 0.6401852965354919,
757
  "eval_macro avg": {
758
- "f1-score": 0.8403395121949689,
759
- "precision": 0.8387425270579937,
760
- "recall": 0.8556929867306292,
761
- "support": 29705.0
762
- },
763
- "eval_runtime": 4.8727,
764
- "eval_samples_per_second": 16.418,
765
- "eval_steps_per_second": 2.052,
766
  "eval_weighted avg": {
767
- "f1-score": 0.8945864954779644,
768
- "precision": 0.9075830699664568,
769
- "recall": 0.8894125568086181,
770
- "support": 29705.0
771
  },
772
  "step": 972
773
  },
774
  {
775
  "epoch": 12.35,
776
- "grad_norm": 23.30899429321289,
777
- "learning_rate": 1.506172839506173e-05,
778
- "loss": 0.0272,
779
  "step": 1000
780
  },
781
  {
782
  "epoch": 13.0,
783
  "eval_B-Claim": {
784
- "f1-score": 0.706766917293233,
785
- "precision": 0.7203065134099617,
786
- "recall": 0.6937269372693727,
787
- "support": 271.0
788
  },
789
  "eval_B-MajorClaim": {
790
- "f1-score": 0.8958333333333333,
791
- "precision": 0.8657718120805369,
792
- "recall": 0.9280575539568345,
793
- "support": 139.0
794
  },
795
  "eval_B-Premise": {
796
- "f1-score": 0.8957345971563981,
797
- "precision": 0.8957345971563981,
798
- "recall": 0.8957345971563981,
799
- "support": 633.0
800
  },
801
  "eval_I-Claim": {
802
- "f1-score": 0.6850271528316524,
803
- "precision": 0.7096169300830432,
804
- "recall": 0.6620844788802799,
805
- "support": 4001.0
806
  },
807
  "eval_I-MajorClaim": {
808
- "f1-score": 0.8991678903573177,
809
- "precision": 0.8861553304389773,
810
- "recall": 0.912568306010929,
811
- "support": 2013.0
812
  },
813
  "eval_I-Premise": {
814
- "f1-score": 0.9089637690660374,
815
- "precision": 0.9007362494586401,
816
- "recall": 0.9173429781227946,
817
- "support": 11336.0
818
  },
819
  "eval_O": {
820
- "f1-score": 0.9999557971975424,
821
- "precision": 1.0,
822
- "recall": 0.9999115983026874,
823
- "support": 11312.0
824
  },
825
- "eval_accuracy": 0.9116310385456994,
826
- "eval_loss": 0.6060317754745483,
827
  "eval_macro avg": {
828
- "f1-score": 0.855921351033645,
829
- "precision": 0.8540459189467938,
830
- "recall": 0.8584894928141853,
831
- "support": 29705.0
832
- },
833
- "eval_runtime": 4.8344,
834
- "eval_samples_per_second": 16.548,
835
- "eval_steps_per_second": 2.068,
836
  "eval_weighted avg": {
837
- "f1-score": 0.9106004556040976,
838
- "precision": 0.9098905735839876,
839
- "recall": 0.9116310385456994,
840
- "support": 29705.0
841
  },
842
  "step": 1053
843
  }
844
  ],
845
  "logging_steps": 500,
846
- "max_steps": 4050,
847
  "num_input_tokens_seen": 0,
848
- "num_train_epochs": 50,
849
  "save_steps": 500,
850
  "total_flos": 1869280565334000.0,
851
  "train_batch_size": 4,
 
1
  {
2
+ "best_metric": 0.276384562253952,
3
+ "best_model_checkpoint": "longformer-sep_tok_full_labels/checkpoint-324",
4
  "epoch": 13.0,
5
  "eval_steps": 500,
6
  "global_step": 1053,
 
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
+ "f1-score": 0.21800947867298578,
15
+ "precision": 0.3333333333333333,
16
+ "recall": 0.1619718309859155,
17
+ "support": 284.0
18
  },
19
  "eval_B-MajorClaim": {
20
+ "f1-score": 0.0,
21
+ "precision": 0.0,
22
+ "recall": 0.0,
23
+ "support": 141.0
24
  },
25
  "eval_B-Premise": {
26
+ "f1-score": 0.8305489260143198,
27
+ "precision": 0.71900826446281,
28
+ "recall": 0.9830508474576272,
29
+ "support": 708.0
30
  },
31
  "eval_I-Claim": {
32
+ "f1-score": 0.5590492696211935,
33
+ "precision": 0.5643589102724319,
34
+ "recall": 0.5538386068187393,
35
+ "support": 4077.0
36
  },
37
  "eval_I-MajorClaim": {
38
+ "f1-score": 0.7399038461538461,
39
+ "precision": 0.7205056179775281,
40
+ "recall": 0.7603754940711462,
41
+ "support": 2024.0
42
  },
43
  "eval_I-Premise": {
44
+ "f1-score": 0.8867229093291834,
45
+ "precision": 0.8883965205974068,
46
+ "recall": 0.8850555918901243,
47
+ "support": 12232.0
48
  },
49
  "eval_O": {
50
+ "f1-score": 0.9976548035383666,
51
+ "precision": 0.9961383616794018,
52
+ "recall": 0.9991758694577221,
53
+ "support": 12134.0
54
  },
55
+ "eval_accuracy": 0.8699050632911393,
56
+ "eval_loss": 0.32671716809272766,
57
  "eval_macro avg": {
58
+ "f1-score": 0.6045556047614136,
59
+ "precision": 0.6031058583318446,
60
+ "recall": 0.6204954629544678,
61
+ "support": 31600.0
62
+ },
63
+ "eval_runtime": 6.234,
64
+ "eval_samples_per_second": 12.833,
65
+ "eval_steps_per_second": 1.604,
66
  "eval_weighted avg": {
67
+ "f1-score": 0.8664142595402331,
68
+ "precision": 0.8644597558999653,
69
+ "recall": 0.8699050632911393,
70
+ "support": 31600.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
+ "f1-score": 0.5352697095435685,
78
+ "precision": 0.6515151515151515,
79
+ "recall": 0.45422535211267606,
80
+ "support": 284.0
81
  },
82
  "eval_B-MajorClaim": {
83
+ "f1-score": 0.7766990291262136,
84
+ "precision": 0.7142857142857143,
85
+ "recall": 0.851063829787234,
86
+ "support": 141.0
87
  },
88
  "eval_B-Premise": {
89
+ "f1-score": 0.8802721088435375,
90
+ "precision": 0.8490813648293963,
91
+ "recall": 0.9138418079096046,
92
+ "support": 708.0
93
  },
94
  "eval_I-Claim": {
95
+ "f1-score": 0.5747557221255521,
96
+ "precision": 0.6325869180907484,
97
+ "recall": 0.5266127054206524,
98
+ "support": 4077.0
99
  },
100
  "eval_I-MajorClaim": {
101
+ "f1-score": 0.7833260963960051,
102
+ "precision": 0.6986831913245546,
103
+ "recall": 0.8913043478260869,
104
+ "support": 2024.0
105
  },
106
  "eval_I-Premise": {
107
+ "f1-score": 0.8995515695067264,
108
+ "precision": 0.8971377459749553,
109
+ "recall": 0.9019784172661871,
110
+ "support": 12232.0
111
  },
112
  "eval_O": {
113
+ "f1-score": 0.9971231300345225,
114
+ "precision": 0.9945072962780783,
115
+ "recall": 0.9997527608373167,
116
+ "support": 12134.0
117
  },
118
+ "eval_accuracy": 0.8864240506329114,
119
+ "eval_loss": 0.28716403245925903,
120
  "eval_macro avg": {
121
+ "f1-score": 0.7781424807965893,
122
+ "precision": 0.7768281974712284,
123
+ "recall": 0.791254174451394,
124
+ "support": 31600.0
125
+ },
126
+ "eval_runtime": 6.2816,
127
+ "eval_samples_per_second": 12.736,
128
+ "eval_steps_per_second": 1.592,
129
  "eval_weighted avg": {
130
+ "f1-score": 0.8834146129726252,
131
+ "precision": 0.8835831101628018,
132
+ "recall": 0.8864240506329114,
133
+ "support": 31600.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
+ "f1-score": 0.5580448065173116,
141
+ "precision": 0.6618357487922706,
142
+ "recall": 0.4823943661971831,
143
+ "support": 284.0
144
  },
145
  "eval_B-MajorClaim": {
146
+ "f1-score": 0.8046875,
147
+ "precision": 0.8956521739130435,
148
+ "recall": 0.7304964539007093,
149
+ "support": 141.0
150
  },
151
  "eval_B-Premise": {
152
+ "f1-score": 0.8868421052631579,
153
+ "precision": 0.8300492610837439,
154
+ "recall": 0.9519774011299436,
155
+ "support": 708.0
156
  },
157
  "eval_I-Claim": {
158
+ "f1-score": 0.5488426262920367,
159
+ "precision": 0.6751432664756447,
160
+ "recall": 0.4623497669855286,
161
+ "support": 4077.0
162
  },
163
  "eval_I-MajorClaim": {
164
+ "f1-score": 0.8110936682365254,
165
+ "precision": 0.8620689655172413,
166
+ "recall": 0.7658102766798419,
167
+ "support": 2024.0
168
  },
169
  "eval_I-Premise": {
170
+ "f1-score": 0.9039491427470623,
171
+ "precision": 0.854842235662756,
172
+ "recall": 0.9590418574231524,
173
+ "support": 12232.0
174
  },
175
  "eval_O": {
176
+ "f1-score": 0.9979824597521307,
177
+ "precision": 0.9972023368715544,
178
+ "recall": 0.9987638041865832,
179
+ "support": 12134.0
180
  },
181
+ "eval_accuracy": 0.892373417721519,
182
+ "eval_loss": 0.28271517157554626,
183
  "eval_macro avg": {
184
+ "f1-score": 0.7873489012583177,
185
+ "precision": 0.825256284045179,
186
+ "recall": 0.7644048466432775,
187
+ "support": 31600.0
188
+ },
189
+ "eval_runtime": 6.2071,
190
+ "eval_samples_per_second": 12.888,
191
+ "eval_steps_per_second": 1.611,
192
  "eval_weighted avg": {
193
+ "f1-score": 0.88435885840807,
194
+ "precision": 0.8846770016417851,
195
+ "recall": 0.892373417721519,
196
+ "support": 31600.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
+ "f1-score": 0.659016393442623,
204
+ "precision": 0.6165644171779141,
205
+ "recall": 0.7077464788732394,
206
+ "support": 284.0
207
  },
208
  "eval_B-MajorClaim": {
209
+ "f1-score": 0.8675496688741722,
210
+ "precision": 0.8136645962732919,
211
+ "recall": 0.9290780141843972,
212
+ "support": 141.0
213
  },
214
  "eval_B-Premise": {
215
+ "f1-score": 0.8698224852071005,
216
+ "precision": 0.9130434782608695,
217
+ "recall": 0.8305084745762712,
218
+ "support": 708.0
219
  },
220
  "eval_I-Claim": {
221
+ "f1-score": 0.6472172351885099,
222
+ "precision": 0.5964839710444674,
223
+ "recall": 0.70738287956831,
224
+ "support": 4077.0
225
  },
226
  "eval_I-MajorClaim": {
227
+ "f1-score": 0.8549332083430982,
228
+ "precision": 0.8131966116807846,
229
+ "recall": 0.9011857707509882,
230
+ "support": 2024.0
231
  },
232
  "eval_I-Premise": {
233
+ "f1-score": 0.8876102800153433,
234
+ "precision": 0.9271658801531475,
235
+ "recall": 0.8512916939175932,
236
+ "support": 12232.0
237
  },
238
  "eval_O": {
239
+ "f1-score": 0.9981065283609122,
240
+ "precision": 0.9970394736842105,
241
+ "recall": 0.9991758694577221,
242
+ "support": 12134.0
243
  },
244
+ "eval_accuracy": 0.8912974683544304,
245
+ "eval_loss": 0.276384562253952,
246
  "eval_macro avg": {
247
+ "f1-score": 0.8263222570616798,
248
+ "precision": 0.8110226326106693,
249
+ "recall": 0.8466241687612174,
250
+ "support": 31600.0
251
+ },
252
+ "eval_runtime": 6.2529,
253
+ "eval_samples_per_second": 12.794,
254
+ "eval_steps_per_second": 1.599,
255
  "eval_weighted avg": {
256
+ "f1-score": 0.8943886873545748,
257
+ "precision": 0.9004180663566287,
258
+ "recall": 0.8912974683544304,
259
+ "support": 31600.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
+ "f1-score": 0.665474060822898,
267
+ "precision": 0.6763636363636364,
268
+ "recall": 0.6549295774647887,
269
+ "support": 284.0
270
  },
271
  "eval_B-MajorClaim": {
272
+ "f1-score": 0.8896551724137931,
273
+ "precision": 0.8657718120805369,
274
+ "recall": 0.9148936170212766,
275
+ "support": 141.0
276
  },
277
  "eval_B-Premise": {
278
+ "f1-score": 0.8843441466854725,
279
+ "precision": 0.8830985915492958,
280
+ "recall": 0.885593220338983,
281
+ "support": 708.0
282
  },
283
  "eval_I-Claim": {
284
+ "f1-score": 0.6698067632850242,
285
+ "precision": 0.6597668332143707,
286
+ "recall": 0.6801569781702232,
287
+ "support": 4077.0
288
  },
289
  "eval_I-MajorClaim": {
290
+ "f1-score": 0.8809815950920246,
291
+ "precision": 0.8751828376401756,
292
+ "recall": 0.8868577075098815,
293
+ "support": 2024.0
294
  },
295
  "eval_I-Premise": {
296
+ "f1-score": 0.9024631353488756,
297
+ "precision": 0.9094305163539764,
298
+ "recall": 0.8956017004578156,
299
+ "support": 12232.0
300
  },
301
  "eval_O": {
302
+ "f1-score": 0.9980246913580246,
303
+ "precision": 0.9967121486108828,
304
+ "recall": 0.9993406955661777,
305
+ "support": 12134.0
306
  },
307
+ "eval_accuracy": 0.9047784810126582,
308
+ "eval_loss": 0.2893199920654297,
309
  "eval_macro avg": {
310
+ "f1-score": 0.8415356521437304,
311
+ "precision": 0.838046625116125,
312
+ "recall": 0.8453390709327352,
313
+ "support": 31600.0
314
+ },
315
+ "eval_runtime": 6.2466,
316
+ "eval_samples_per_second": 12.807,
317
+ "eval_steps_per_second": 1.601,
318
  "eval_weighted avg": {
319
+ "f1-score": 0.9051715590931133,
320
+ "precision": 0.9056611908459661,
321
+ "recall": 0.9047784810126582,
322
+ "support": 31600.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
+ "f1-score": 0.6263345195729537,
330
+ "precision": 0.6330935251798561,
331
+ "recall": 0.6197183098591549,
332
+ "support": 284.0
333
  },
334
  "eval_B-MajorClaim": {
335
+ "f1-score": 0.8372093023255814,
336
+ "precision": 0.9230769230769231,
337
+ "recall": 0.7659574468085106,
338
+ "support": 141.0
339
  },
340
  "eval_B-Premise": {
341
+ "f1-score": 0.8784530386740332,
342
+ "precision": 0.8594594594594595,
343
+ "recall": 0.8983050847457628,
344
+ "support": 708.0
345
  },
346
  "eval_I-Claim": {
347
+ "f1-score": 0.6109127995920448,
348
+ "precision": 0.6360499070878683,
349
+ "recall": 0.5876870247731175,
350
+ "support": 4077.0
351
  },
352
  "eval_I-MajorClaim": {
353
+ "f1-score": 0.8409272581934453,
354
+ "precision": 0.9126662810873337,
355
+ "recall": 0.7796442687747036,
356
+ "support": 2024.0
357
  },
358
  "eval_I-Premise": {
359
+ "f1-score": 0.8964938902643559,
360
+ "precision": 0.8762685402029664,
361
+ "recall": 0.9176749509483323,
362
+ "support": 12232.0
363
  },
364
  "eval_O": {
365
+ "f1-score": 0.9987239122380934,
366
+ "precision": 0.9976971790443293,
367
+ "recall": 0.9997527608373167,
368
+ "support": 12134.0
369
  },
370
+ "eval_accuracy": 0.8939873417721519,
371
+ "eval_loss": 0.33532533049583435,
372
  "eval_macro avg": {
373
+ "f1-score": 0.8127221029800725,
374
+ "precision": 0.8340445450198194,
375
+ "recall": 0.7955342638209855,
376
+ "support": 31600.0
377
+ },
378
+ "eval_runtime": 6.2278,
379
+ "eval_samples_per_second": 12.846,
380
+ "eval_steps_per_second": 1.606,
381
  "eval_weighted avg": {
382
+ "f1-score": 0.8922477132246446,
383
+ "precision": 0.891880888702747,
384
+ "recall": 0.8939873417721519,
385
+ "support": 31600.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
+ "grad_norm": 4.557807445526123,
392
+ "learning_rate": 1.3827160493827162e-05,
393
+ "loss": 0.2608,
394
  "step": 500
395
  },
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
+ "f1-score": 0.5553235908141962,
400
+ "precision": 0.6820512820512821,
401
+ "recall": 0.46830985915492956,
402
+ "support": 284.0
403
  },
404
  "eval_B-MajorClaim": {
405
+ "f1-score": 0.8412698412698413,
406
+ "precision": 0.954954954954955,
407
+ "recall": 0.75177304964539,
408
+ "support": 141.0
409
  },
410
  "eval_B-Premise": {
411
+ "f1-score": 0.8846905537459283,
412
+ "precision": 0.8210399032648126,
413
+ "recall": 0.9590395480225988,
414
+ "support": 708.0
415
  },
416
  "eval_I-Claim": {
417
+ "f1-score": 0.5484584747012834,
418
+ "precision": 0.6880088823094005,
419
+ "recall": 0.45597252882021094,
420
+ "support": 4077.0
421
  },
422
  "eval_I-MajorClaim": {
423
+ "f1-score": 0.8530857454942655,
424
+ "precision": 0.9536019536019537,
425
+ "recall": 0.7717391304347826,
426
+ "support": 2024.0
427
  },
428
  "eval_I-Premise": {
429
+ "f1-score": 0.900171657448026,
430
+ "precision": 0.8438103411285132,
431
+ "recall": 0.9646010464355788,
432
+ "support": 12232.0
433
  },
434
  "eval_O": {
435
+ "f1-score": 0.9986819342614713,
436
+ "precision": 0.9982707509881423,
437
+ "recall": 0.9990934564034943,
438
+ "support": 12134.0
439
  },
440
+ "eval_accuracy": 0.8943354430379746,
441
+ "eval_loss": 0.45137375593185425,
442
  "eval_macro avg": {
443
+ "f1-score": 0.7973831139621446,
444
+ "precision": 0.8488197240427228,
445
+ "recall": 0.7672183741309979,
446
+ "support": 31600.0
447
+ },
448
+ "eval_runtime": 6.1951,
449
+ "eval_samples_per_second": 12.913,
450
+ "eval_steps_per_second": 1.614,
451
  "eval_weighted avg": {
452
+ "f1-score": 0.8858958517067363,
453
+ "precision": 0.8885840321741321,
454
+ "recall": 0.8943354430379746,
455
+ "support": 31600.0
456
  },
457
  "step": 567
458
  },
459
  {
460
  "epoch": 8.0,
461
  "eval_B-Claim": {
462
+ "f1-score": 0.6642599277978338,
463
+ "precision": 0.6814814814814815,
464
+ "recall": 0.647887323943662,
465
+ "support": 284.0
466
  },
467
  "eval_B-MajorClaim": {
468
+ "f1-score": 0.9015151515151516,
469
+ "precision": 0.967479674796748,
470
+ "recall": 0.8439716312056738,
471
+ "support": 141.0
472
  },
473
  "eval_B-Premise": {
474
+ "f1-score": 0.8873531444367657,
475
+ "precision": 0.8687415426251691,
476
+ "recall": 0.9067796610169492,
477
+ "support": 708.0
478
  },
479
  "eval_I-Claim": {
480
+ "f1-score": 0.6521516393442622,
481
+ "precision": 0.6823907799517556,
482
+ "recall": 0.6244787834191807,
483
+ "support": 4077.0
484
  },
485
  "eval_I-MajorClaim": {
486
+ "f1-score": 0.8962213943587014,
487
+ "precision": 0.9711649365628604,
488
+ "recall": 0.8320158102766798,
489
+ "support": 2024.0
490
  },
491
  "eval_I-Premise": {
492
+ "f1-score": 0.9046460618145563,
493
+ "precision": 0.8831269952503309,
494
+ "recall": 0.9272400261608895,
495
+ "support": 12232.0
496
  },
497
  "eval_O": {
498
+ "f1-score": 0.9979418786531653,
499
+ "precision": 0.996875,
500
+ "recall": 0.9990110433492665,
501
+ "support": 12134.0
502
  },
503
+ "eval_accuracy": 0.9062974683544304,
504
+ "eval_loss": 0.40965867042541504,
505
  "eval_macro avg": {
506
+ "f1-score": 0.8434413139886338,
507
+ "precision": 0.8644657729526208,
508
+ "recall": 0.8259120399103289,
509
+ "support": 31600.0
510
+ },
511
+ "eval_runtime": 6.2033,
512
+ "eval_samples_per_second": 12.896,
513
+ "eval_steps_per_second": 1.612,
514
  "eval_weighted avg": {
515
+ "f1-score": 0.9047924430886448,
516
+ "precision": 0.9047867115327305,
517
+ "recall": 0.9062974683544304,
518
+ "support": 31600.0
519
  },
520
  "step": 648
521
  },
522
  {
523
  "epoch": 9.0,
524
  "eval_B-Claim": {
525
+ "f1-score": 0.6900175131348512,
526
+ "precision": 0.686411149825784,
527
+ "recall": 0.6936619718309859,
528
+ "support": 284.0
529
  },
530
  "eval_B-MajorClaim": {
531
+ "f1-score": 0.9064748201438848,
532
+ "precision": 0.9197080291970803,
533
+ "recall": 0.8936170212765957,
534
+ "support": 141.0
535
  },
536
  "eval_B-Premise": {
537
+ "f1-score": 0.8892025405786873,
538
+ "precision": 0.8885754583921015,
539
+ "recall": 0.8898305084745762,
540
+ "support": 708.0
541
  },
542
  "eval_I-Claim": {
543
+ "f1-score": 0.6670815183571872,
544
+ "precision": 0.6771096513390601,
545
+ "recall": 0.657346087809664,
546
+ "support": 4077.0
547
  },
548
  "eval_I-MajorClaim": {
549
+ "f1-score": 0.8905813658288906,
550
+ "precision": 0.9159268929503916,
551
+ "recall": 0.866600790513834,
552
+ "support": 2024.0
553
  },
554
  "eval_I-Premise": {
555
+ "f1-score": 0.9051692681937971,
556
+ "precision": 0.8978524893428779,
557
+ "recall": 0.9126062786134729,
558
+ "support": 12232.0
559
  },
560
  "eval_O": {
561
+ "f1-score": 0.9987240172875077,
562
+ "precision": 0.9976153276868679,
563
+ "recall": 0.9998351738915444,
564
+ "support": 12134.0
565
  },
566
+ "eval_accuracy": 0.9076582278481012,
567
+ "eval_loss": 0.44169458746910095,
568
  "eval_macro avg": {
569
+ "f1-score": 0.849607291932115,
570
+ "precision": 0.8547427141048803,
571
+ "recall": 0.8447854046300962,
572
+ "support": 31600.0
573
+ },
574
+ "eval_runtime": 6.2166,
575
+ "eval_samples_per_second": 12.869,
576
+ "eval_steps_per_second": 1.609,
577
  "eval_weighted avg": {
578
+ "f1-score": 0.9071553503542205,
579
+ "precision": 0.9068271879381139,
580
+ "recall": 0.9076582278481012,
581
+ "support": 31600.0
582
  },
583
  "step": 729
584
  },
585
  {
586
  "epoch": 10.0,
587
  "eval_B-Claim": {
588
+ "f1-score": 0.6761565836298933,
589
+ "precision": 0.6834532374100719,
590
+ "recall": 0.6690140845070423,
591
+ "support": 284.0
592
  },
593
  "eval_B-MajorClaim": {
594
+ "f1-score": 0.8872727272727273,
595
+ "precision": 0.9104477611940298,
596
+ "recall": 0.8652482269503546,
597
+ "support": 141.0
598
  },
599
  "eval_B-Premise": {
600
+ "f1-score": 0.8901329601119664,
601
+ "precision": 0.8821081830790569,
602
+ "recall": 0.8983050847457628,
603
+ "support": 708.0
604
  },
605
  "eval_I-Claim": {
606
+ "f1-score": 0.6512742871561948,
607
+ "precision": 0.6705637828007275,
608
+ "recall": 0.6330635271032622,
609
+ "support": 4077.0
610
  },
611
  "eval_I-MajorClaim": {
612
+ "f1-score": 0.8682926829268293,
613
+ "precision": 0.9037947621592731,
614
+ "recall": 0.8354743083003953,
615
+ "support": 2024.0
616
  },
617
  "eval_I-Premise": {
618
+ "f1-score": 0.905623866156017,
619
+ "precision": 0.8933428775948461,
620
+ "recall": 0.9182472204054938,
621
+ "support": 12232.0
622
  },
623
  "eval_O": {
624
+ "f1-score": 0.9981898963304262,
625
+ "precision": 0.9965500246426812,
626
+ "recall": 0.9998351738915444,
627
+ "support": 12134.0
628
  },
629
+ "eval_accuracy": 0.9045569620253164,
630
+ "eval_loss": 0.45926380157470703,
631
  "eval_macro avg": {
632
+ "f1-score": 0.8395632862262934,
633
+ "precision": 0.8486086612686695,
634
+ "recall": 0.8313125179862652,
635
+ "support": 31600.0
636
+ },
637
+ "eval_runtime": 6.2031,
638
+ "eval_samples_per_second": 12.897,
639
+ "eval_steps_per_second": 1.612,
640
  "eval_weighted avg": {
641
+ "f1-score": 0.9034697801243392,
642
+ "precision": 0.9028380907030437,
643
+ "recall": 0.9045569620253164,
644
+ "support": 31600.0
645
  },
646
  "step": 810
647
  },
648
  {
649
  "epoch": 11.0,
650
  "eval_B-Claim": {
651
+ "f1-score": 0.6407766990291263,
652
+ "precision": 0.7142857142857143,
653
+ "recall": 0.5809859154929577,
654
+ "support": 284.0
655
  },
656
  "eval_B-MajorClaim": {
657
+ "f1-score": 0.9032258064516129,
658
+ "precision": 0.9130434782608695,
659
+ "recall": 0.8936170212765957,
660
+ "support": 141.0
661
  },
662
  "eval_B-Premise": {
663
+ "f1-score": 0.8872282608695653,
664
+ "precision": 0.8547120418848168,
665
+ "recall": 0.922316384180791,
666
+ "support": 708.0
667
  },
668
  "eval_I-Claim": {
669
+ "f1-score": 0.5974395448079659,
670
+ "precision": 0.7111412123264477,
671
+ "recall": 0.515084621044886,
672
+ "support": 4077.0
673
  },
674
  "eval_I-MajorClaim": {
675
+ "f1-score": 0.8862901219208758,
676
+ "precision": 0.8927318295739348,
677
+ "recall": 0.8799407114624506,
678
+ "support": 2024.0
679
  },
680
  "eval_I-Premise": {
681
+ "f1-score": 0.9007579901539423,
682
+ "precision": 0.862670258943272,
683
+ "recall": 0.9423642903858731,
684
+ "support": 12232.0
685
  },
686
  "eval_O": {
687
+ "f1-score": 0.9989708122349842,
688
+ "precision": 0.9980258287406433,
689
+ "recall": 0.9999175869457723,
690
+ "support": 12134.0
691
  },
692
+ "eval_accuracy": 0.9014240506329114,
693
+ "eval_loss": 0.5345993638038635,
694
  "eval_macro avg": {
695
+ "f1-score": 0.8306698907811532,
696
+ "precision": 0.8495157662879569,
697
+ "recall": 0.8191752186841895,
698
+ "support": 31600.0
699
+ },
700
+ "eval_runtime": 6.2101,
701
+ "eval_samples_per_second": 12.882,
702
+ "eval_steps_per_second": 1.61,
703
  "eval_weighted avg": {
704
+ "f1-score": 0.8957812921551218,
705
+ "precision": 0.8957333024681017,
706
+ "recall": 0.9014240506329114,
707
+ "support": 31600.0
708
  },
709
  "step": 891
710
  },
711
  {
712
  "epoch": 12.0,
713
  "eval_B-Claim": {
714
+ "f1-score": 0.6859083191850596,
715
+ "precision": 0.6622950819672131,
716
+ "recall": 0.7112676056338029,
717
+ "support": 284.0
718
  },
719
  "eval_B-MajorClaim": {
720
+ "f1-score": 0.8921933085501859,
721
+ "precision": 0.9375,
722
+ "recall": 0.851063829787234,
723
+ "support": 141.0
724
  },
725
  "eval_B-Premise": {
726
+ "f1-score": 0.8863636363636365,
727
+ "precision": 0.8914285714285715,
728
+ "recall": 0.8813559322033898,
729
+ "support": 708.0
730
  },
731
  "eval_I-Claim": {
732
+ "f1-score": 0.6563715953307394,
733
+ "precision": 0.6508319266939957,
734
+ "recall": 0.6620063772381654,
735
+ "support": 4077.0
736
  },
737
  "eval_I-MajorClaim": {
738
+ "f1-score": 0.8721605916534602,
739
+ "precision": 0.9370034052213394,
740
+ "recall": 0.8157114624505929,
741
+ "support": 2024.0
742
  },
743
  "eval_I-Premise": {
744
+ "f1-score": 0.9015022330491272,
745
+ "precision": 0.8954670108081949,
746
+ "recall": 0.907619359058208,
747
+ "support": 12232.0
748
  },
749
  "eval_O": {
750
+ "f1-score": 0.9988474520457726,
751
+ "precision": 0.9977796052631579,
752
+ "recall": 0.9999175869457723,
753
+ "support": 12134.0
754
  },
755
+ "eval_accuracy": 0.902879746835443,
756
+ "eval_loss": 0.6067038774490356,
757
  "eval_macro avg": {
758
+ "f1-score": 0.8419067337397116,
759
+ "precision": 0.8531865144832105,
760
+ "recall": 0.8327060219024521,
761
+ "support": 31600.0
762
+ },
763
+ "eval_runtime": 6.2295,
764
+ "eval_samples_per_second": 12.842,
765
+ "eval_steps_per_second": 1.605,
766
  "eval_weighted avg": {
767
+ "f1-score": 0.9030573735174033,
768
+ "precision": 0.9038530884689406,
769
+ "recall": 0.902879746835443,
770
+ "support": 31600.0
771
  },
772
  "step": 972
773
  },
774
  {
775
  "epoch": 12.35,
776
+ "grad_norm": 0.581019937992096,
777
+ "learning_rate": 7.654320987654322e-06,
778
+ "loss": 0.0322,
779
  "step": 1000
780
  },
781
  {
782
  "epoch": 13.0,
783
  "eval_B-Claim": {
784
+ "f1-score": 0.6728624535315986,
785
+ "precision": 0.7125984251968503,
786
+ "recall": 0.6373239436619719,
787
+ "support": 284.0
788
  },
789
  "eval_B-MajorClaim": {
790
+ "f1-score": 0.9154929577464789,
791
+ "precision": 0.9090909090909091,
792
+ "recall": 0.9219858156028369,
793
+ "support": 141.0
794
  },
795
  "eval_B-Premise": {
796
+ "f1-score": 0.8905817174515235,
797
+ "precision": 0.873641304347826,
798
+ "recall": 0.9081920903954802,
799
+ "support": 708.0
800
  },
801
  "eval_I-Claim": {
802
+ "f1-score": 0.6460571129159025,
803
+ "precision": 0.6932808546527973,
804
+ "recall": 0.6048565121412803,
805
+ "support": 4077.0
806
  },
807
  "eval_I-MajorClaim": {
808
+ "f1-score": 0.9011916583912611,
809
+ "precision": 0.905688622754491,
810
+ "recall": 0.8967391304347826,
811
+ "support": 2024.0
812
  },
813
  "eval_I-Premise": {
814
+ "f1-score": 0.9046914370275567,
815
+ "precision": 0.8856001879257693,
816
+ "recall": 0.9246239372138653,
817
+ "support": 12232.0
818
  },
819
  "eval_O": {
820
+ "f1-score": 0.9991346985866744,
821
+ "precision": 0.9990935311083642,
822
+ "recall": 0.9991758694577221,
823
+ "support": 12134.0
824
  },
825
+ "eval_accuracy": 0.907246835443038,
826
+ "eval_loss": 0.5913504362106323,
827
  "eval_macro avg": {
828
+ "f1-score": 0.8471445765215709,
829
+ "precision": 0.8541419764395725,
830
+ "recall": 0.8418424712725627,
831
+ "support": 31600.0
832
+ },
833
+ "eval_runtime": 6.2496,
834
+ "eval_samples_per_second": 12.801,
835
+ "eval_steps_per_second": 1.6,
836
  "eval_weighted avg": {
837
+ "f1-score": 0.9050120935479347,
838
+ "precision": 0.9039360771033997,
839
+ "recall": 0.907246835443038,
840
+ "support": 31600.0
841
  },
842
  "step": 1053
843
  }
844
  ],
845
  "logging_steps": 500,
846
+ "max_steps": 1620,
847
  "num_input_tokens_seen": 0,
848
+ "num_train_epochs": 20,
849
  "save_steps": 500,
850
  "total_flos": 1869280565334000.0,
851
  "train_batch_size": 4,
checkpoint-1053/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3aa7d1dcb1ecb96052506e9849f1d3fa19e688784f8ea37f784dfaca9546785
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a6eed06a4e3317966121cd80a8275e0b82aa59ebaa0185c3b4676ff92bfed33
3
  size 4920