Theoreticallyhugo commited on
Commit
bfdeb26
1 Parent(s): c1976e8

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:b9e6f99752a736b2899b6d1acc8bae397e69d1e205886936d19145aac2a2f5a6
3
  size 592324828
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5050d529730987c6455f80796717d099332971c4f05e1fb86a2bc7ec59bca4ff
3
  size 592324828
checkpoint-1053/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c5066f72de26a9725aca1a81554f06c1447dbde60143b71564bc693781a1eb62
3
  size 1014657786
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29f093ae09694d81f6132610f1597138305e8f7188c6f04263cc796fb2b0d904
3
  size 1014657786
checkpoint-1053/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a8fed4ff2e0a1f5b060e9416aef90a5efb47a3301fd9fdfb8a444f0ad6d3aa3
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2d0b6fd3e0e222c69d0c818ba4ff34a17c0844523a335fc5dcaffe1d7766cfb
3
  size 14244
checkpoint-1053/trainer_state.json CHANGED
@@ -11,599 +11,599 @@
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
  }
@@ -613,7 +613,7 @@
613
  "num_input_tokens_seen": 0,
614
  "num_train_epochs": 50,
615
  "save_steps": 500,
616
- "total_flos": 1869253525448400.0,
617
  "train_batch_size": 4,
618
  "trial_name": null,
619
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_Claim": {
14
+ "f1-score": 0.4888068880688806,
15
+ "precision": 0.6228840125391849,
16
+ "recall": 0.40222672064777326,
17
+ "support": 4940.0
18
  },
19
  "eval_MajorClaim": {
20
+ "f1-score": 0.7325278370054489,
21
+ "precision": 0.7604525332021643,
22
+ "recall": 0.706581352833638,
23
+ "support": 2188.0
24
  },
25
  "eval_O": {
26
+ "f1-score": 0.8634848138844484,
27
+ "precision": 0.827613377692173,
28
+ "recall": 0.902606702950444,
29
+ "support": 10473.0
30
  },
31
  "eval_Premise": {
32
+ "f1-score": 0.8792819197655248,
33
+ "precision": 0.8543458914268763,
34
+ "recall": 0.9057173407132524,
35
+ "support": 15899.0
36
  },
37
+ "eval_accuracy": 0.8174925373134329,
38
+ "eval_loss": 0.49806877970695496,
39
  "eval_macro avg": {
40
+ "f1-score": 0.7410253646810757,
41
+ "precision": 0.7663239537150996,
42
+ "recall": 0.7292830292862769,
43
+ "support": 33500.0
44
+ },
45
+ "eval_runtime": 4.9194,
46
+ "eval_samples_per_second": 16.465,
47
+ "eval_steps_per_second": 2.236,
48
  "eval_weighted avg": {
49
+ "f1-score": 0.8071778099281224,
50
+ "precision": 0.80572410140167,
51
+ "recall": 0.8174925373134329,
52
+ "support": 33500.0
53
  },
54
  "step": 81
55
  },
56
  {
57
  "epoch": 2.0,
58
  "eval_Claim": {
59
+ "f1-score": 0.5993816254416962,
60
+ "precision": 0.6593780369290574,
61
+ "recall": 0.5493927125506073,
62
+ "support": 4940.0
63
  },
64
  "eval_MajorClaim": {
65
+ "f1-score": 0.8017817371937639,
66
+ "precision": 0.7197382769901853,
67
+ "recall": 0.9049360146252285,
68
+ "support": 2188.0
69
  },
70
  "eval_O": {
71
+ "f1-score": 0.9058997050147494,
72
+ "precision": 0.9337184554575859,
73
+ "recall": 0.8796906330564308,
74
+ "support": 10473.0
75
  },
76
  "eval_Premise": {
77
+ "f1-score": 0.8854737486606459,
78
+ "precision": 0.862579028987236,
79
+ "recall": 0.9096169570413234,
80
+ "support": 15899.0
81
  },
82
+ "eval_accuracy": 0.8468358208955223,
83
+ "eval_loss": 0.4562840759754181,
84
  "eval_macro avg": {
85
+ "f1-score": 0.7981342040777137,
86
+ "precision": 0.7938534495910161,
87
+ "recall": 0.8109090793183975,
88
+ "support": 33500.0
89
+ },
90
+ "eval_runtime": 4.904,
91
+ "eval_samples_per_second": 16.517,
92
+ "eval_steps_per_second": 2.243,
93
  "eval_weighted avg": {
94
+ "f1-score": 0.8442053257085675,
95
+ "precision": 0.8455251408465503,
96
+ "recall": 0.8468358208955223,
97
+ "support": 33500.0
98
  },
99
  "step": 162
100
  },
101
  {
102
  "epoch": 3.0,
103
  "eval_Claim": {
104
+ "f1-score": 0.6235462150581514,
105
+ "precision": 0.6285479226655697,
106
+ "recall": 0.6186234817813765,
107
+ "support": 4940.0
108
  },
109
  "eval_MajorClaim": {
110
+ "f1-score": 0.8338848444738584,
111
+ "precision": 0.8059701492537313,
112
+ "recall": 0.8638025594149908,
113
+ "support": 2188.0
114
  },
115
  "eval_O": {
116
+ "f1-score": 0.9066250826953974,
117
+ "precision": 0.8974646833193002,
118
+ "recall": 0.9159744103886184,
119
+ "support": 10473.0
120
  },
121
  "eval_Premise": {
122
+ "f1-score": 0.8827730692315017,
123
+ "precision": 0.8911176621379133,
124
+ "recall": 0.8745833071262343,
125
+ "support": 15899.0
126
  },
127
+ "eval_accuracy": 0.8490746268656716,
128
+ "eval_loss": 0.43820467591285706,
129
  "eval_macro avg": {
130
+ "f1-score": 0.8117073028647273,
131
+ "precision": 0.8057751043441286,
132
+ "recall": 0.818245939677805,
133
+ "support": 33500.0
134
+ },
135
+ "eval_runtime": 4.9179,
136
+ "eval_samples_per_second": 16.47,
137
+ "eval_steps_per_second": 2.237,
138
  "eval_weighted avg": {
139
+ "f1-score": 0.8488105033097496,
140
+ "precision": 0.8488213959184715,
141
+ "recall": 0.8490746268656716,
142
+ "support": 33500.0
143
  },
144
  "step": 243
145
  },
146
  {
147
  "epoch": 4.0,
148
  "eval_Claim": {
149
+ "f1-score": 0.6415287144900879,
150
+ "precision": 0.6477507222451506,
151
+ "recall": 0.6354251012145749,
152
+ "support": 4940.0
153
  },
154
  "eval_MajorClaim": {
155
+ "f1-score": 0.8005126014523707,
156
+ "precision": 0.7514033680834001,
157
+ "recall": 0.856489945155393,
158
+ "support": 2188.0
159
  },
160
  "eval_O": {
161
+ "f1-score": 0.9065169636319258,
162
+ "precision": 0.9273871354374751,
163
+ "recall": 0.8865654540246348,
164
+ "support": 10473.0
165
  },
166
  "eval_Premise": {
167
+ "f1-score": 0.8932505382719131,
168
+ "precision": 0.8863636363636364,
169
+ "recall": 0.9002452984464432,
170
+ "support": 15899.0
171
  },
172
+ "eval_accuracy": 0.8540597014925373,
173
+ "eval_loss": 0.4749625325202942,
174
  "eval_macro avg": {
175
+ "f1-score": 0.8104522044615743,
176
+ "precision": 0.8032262155324155,
177
+ "recall": 0.8196814497102615,
178
+ "support": 33500.0
179
+ },
180
+ "eval_runtime": 4.8972,
181
+ "eval_samples_per_second": 16.54,
182
+ "eval_steps_per_second": 2.246,
183
  "eval_weighted avg": {
184
+ "f1-score": 0.8542213698406307,
185
+ "precision": 0.8551874645146165,
186
+ "recall": 0.8540597014925373,
187
+ "support": 33500.0
188
  },
189
  "step": 324
190
  },
191
  {
192
  "epoch": 5.0,
193
  "eval_Claim": {
194
+ "f1-score": 0.6099687080911935,
195
+ "precision": 0.6808882235528942,
196
+ "recall": 0.5524291497975709,
197
+ "support": 4940.0
198
  },
199
  "eval_MajorClaim": {
200
+ "f1-score": 0.8315018315018317,
201
+ "precision": 0.8330275229357799,
202
+ "recall": 0.829981718464351,
203
+ "support": 2188.0
204
  },
205
  "eval_O": {
206
+ "f1-score": 0.9103454839630722,
207
+ "precision": 0.9074091642159188,
208
+ "recall": 0.9133008689009835,
209
+ "support": 10473.0
210
  },
211
  "eval_Premise": {
212
+ "f1-score": 0.8913988368533824,
213
+ "precision": 0.8682249120505635,
214
+ "recall": 0.915843763758727,
215
+ "support": 15899.0
216
  },
217
+ "eval_accuracy": 0.8558507462686568,
218
+ "eval_loss": 0.5162495374679565,
219
  "eval_macro avg": {
220
+ "f1-score": 0.8108037151023699,
221
+ "precision": 0.8223874556887891,
222
+ "recall": 0.8028888752304081,
223
+ "support": 33500.0
224
+ },
225
+ "eval_runtime": 4.9413,
226
+ "eval_samples_per_second": 16.392,
227
+ "eval_steps_per_second": 2.226,
228
  "eval_weighted avg": {
229
+ "f1-score": 0.8519095458499607,
230
+ "precision": 0.8505509283002989,
231
+ "recall": 0.8558507462686568,
232
+ "support": 33500.0
233
  },
234
  "step": 405
235
  },
236
  {
237
  "epoch": 6.0,
238
  "eval_Claim": {
239
+ "f1-score": 0.646279703934554,
240
+ "precision": 0.6227477477477478,
241
+ "recall": 0.6716599190283401,
242
+ "support": 4940.0
243
  },
244
  "eval_MajorClaim": {
245
+ "f1-score": 0.8414067684140676,
246
+ "precision": 0.8152593227603944,
247
+ "recall": 0.8692870201096892,
248
+ "support": 2188.0
249
  },
250
  "eval_O": {
251
+ "f1-score": 0.9112506693929215,
252
+ "precision": 0.9295788637266588,
253
+ "recall": 0.8936312422419556,
254
+ "support": 10473.0
255
  },
256
  "eval_Premise": {
257
+ "f1-score": 0.8861383012314494,
258
+ "precision": 0.8897343224906474,
259
+ "recall": 0.8825712308950249,
260
+ "support": 15899.0
261
  },
262
+ "eval_accuracy": 0.8540597014925373,
263
+ "eval_loss": 0.552895724773407,
264
  "eval_macro avg": {
265
+ "f1-score": 0.8212688607432481,
266
+ "precision": 0.8143300641813621,
267
+ "recall": 0.8292873530687525,
268
+ "support": 33500.0
269
+ },
270
+ "eval_runtime": 4.9336,
271
+ "eval_samples_per_second": 16.418,
272
+ "eval_steps_per_second": 2.23,
273
  "eval_weighted avg": {
274
+ "f1-score": 0.8556973390614198,
275
+ "precision": 0.8579560210496036,
276
+ "recall": 0.8540597014925373,
277
+ "support": 33500.0
278
  },
279
  "step": 486
280
  },
281
  {
282
  "epoch": 6.17,
283
+ "grad_norm": 9.790245056152344,
284
  "learning_rate": 1.7530864197530865e-05,
285
+ "loss": 0.3328,
286
  "step": 500
287
  },
288
  {
289
  "epoch": 7.0,
290
  "eval_Claim": {
291
+ "f1-score": 0.6202600808654791,
292
+ "precision": 0.6739491807171694,
293
+ "recall": 0.5744939271255061,
294
+ "support": 4940.0
295
  },
296
  "eval_MajorClaim": {
297
+ "f1-score": 0.8670041417395307,
298
+ "precision": 0.8730305838739574,
299
+ "recall": 0.8610603290676416,
300
+ "support": 2188.0
301
  },
302
  "eval_O": {
303
+ "f1-score": 0.9109892729439809,
304
+ "precision": 0.9097314797181489,
305
+ "recall": 0.9122505490308412,
306
+ "support": 10473.0
307
  },
308
  "eval_Premise": {
309
+ "f1-score": 0.8872356123954747,
310
+ "precision": 0.8677611401767996,
311
+ "recall": 0.9076042518397384,
312
+ "support": 15899.0
313
  },
314
+ "eval_accuracy": 0.8568955223880597,
315
+ "eval_loss": 0.5894299149513245,
316
  "eval_macro avg": {
317
+ "f1-score": 0.8213722769861164,
318
+ "precision": 0.8311180961215189,
319
+ "recall": 0.8138522642659318,
320
+ "support": 33500.0
321
+ },
322
+ "eval_runtime": 4.9037,
323
+ "eval_samples_per_second": 16.518,
324
+ "eval_steps_per_second": 2.243,
325
  "eval_weighted avg": {
326
+ "f1-score": 0.8539713289140156,
327
+ "precision": 0.8526463291050193,
328
+ "recall": 0.8568955223880597,
329
+ "support": 33500.0
330
  },
331
  "step": 567
332
  },
333
  {
334
  "epoch": 8.0,
335
  "eval_Claim": {
336
+ "f1-score": 0.6441500279902966,
337
+ "precision": 0.5974385600553825,
338
+ "recall": 0.6987854251012146,
339
+ "support": 4940.0
340
  },
341
  "eval_MajorClaim": {
342
+ "f1-score": 0.834510811389424,
343
+ "precision": 0.7849375755134917,
344
+ "recall": 0.8907678244972578,
345
+ "support": 2188.0
346
  },
347
  "eval_O": {
348
+ "f1-score": 0.9059161060546101,
349
+ "precision": 0.9398542543364466,
350
+ "recall": 0.874343550081161,
351
+ "support": 10473.0
352
  },
353
  "eval_Premise": {
354
+ "f1-score": 0.8801401497053671,
355
+ "precision": 0.8915849251419721,
356
+ "recall": 0.8689854707843261,
357
+ "support": 15899.0
358
  },
359
+ "eval_accuracy": 0.8469850746268657,
360
+ "eval_loss": 0.5878348350524902,
361
  "eval_macro avg": {
362
+ "f1-score": 0.8161792737849245,
363
+ "precision": 0.8034538287618231,
364
+ "recall": 0.8332205676159898,
365
+ "support": 33500.0
366
+ },
367
+ "eval_runtime": 4.9086,
368
+ "eval_samples_per_second": 16.502,
369
+ "eval_steps_per_second": 2.241,
370
  "eval_weighted avg": {
371
+ "f1-score": 0.8504184600736624,
372
+ "precision": 0.8563340964894008,
373
+ "recall": 0.8469850746268657,
374
+ "support": 33500.0
375
  },
376
  "step": 648
377
  },
378
  {
379
  "epoch": 9.0,
380
  "eval_Claim": {
381
+ "f1-score": 0.6323094193285029,
382
+ "precision": 0.6375797489195307,
383
+ "recall": 0.6271255060728745,
384
+ "support": 4940.0
385
  },
386
  "eval_MajorClaim": {
387
+ "f1-score": 0.8525914292124747,
388
+ "precision": 0.8373732921992068,
389
+ "recall": 0.8683729433272395,
390
+ "support": 2188.0
391
  },
392
  "eval_O": {
393
+ "f1-score": 0.9054132151801194,
394
+ "precision": 0.9134194927606647,
395
+ "recall": 0.8975460708488494,
396
+ "support": 10473.0
397
  },
398
  "eval_Premise": {
399
+ "f1-score": 0.8888055034396498,
400
+ "precision": 0.883775884584292,
401
+ "recall": 0.8938926976539405,
402
+ "support": 15899.0
403
  },
404
+ "eval_accuracy": 0.8540298507462687,
405
+ "eval_loss": 0.7043101191520691,
406
  "eval_macro avg": {
407
+ "f1-score": 0.8197798917901866,
408
+ "precision": 0.8180371046159236,
409
+ "recall": 0.821734304475726,
410
+ "support": 33500.0
411
+ },
412
+ "eval_runtime": 4.935,
413
+ "eval_samples_per_second": 16.414,
414
+ "eval_steps_per_second": 2.229,
415
  "eval_weighted avg": {
416
+ "f1-score": 0.853808653145316,
417
+ "precision": 0.8537078167069386,
418
+ "recall": 0.8540298507462687,
419
+ "support": 33500.0
420
  },
421
  "step": 729
422
  },
423
  {
424
  "epoch": 10.0,
425
  "eval_Claim": {
426
+ "f1-score": 0.6347110795166041,
427
+ "precision": 0.6368453230079478,
428
+ "recall": 0.6325910931174089,
429
+ "support": 4940.0
430
  },
431
  "eval_MajorClaim": {
432
+ "f1-score": 0.8442090055381651,
433
+ "precision": 0.8921119592875318,
434
+ "recall": 0.8011882998171846,
435
+ "support": 2188.0
436
  },
437
  "eval_O": {
438
+ "f1-score": 0.9106677408653618,
439
+ "precision": 0.9040270982310877,
440
+ "recall": 0.9174066647569942,
441
+ "support": 10473.0
442
  },
443
  "eval_Premise": {
444
+ "f1-score": 0.8841656478259506,
445
+ "precision": 0.881375,
446
+ "recall": 0.8869740235234921,
447
+ "support": 15899.0
448
  },
449
+ "eval_accuracy": 0.8533731343283583,
450
+ "eval_loss": 0.7164433002471924,
451
  "eval_macro avg": {
452
+ "f1-score": 0.8184383684365204,
453
+ "precision": 0.8285898451316418,
454
+ "recall": 0.8095400203037699,
455
+ "support": 33500.0
456
+ },
457
+ "eval_runtime": 4.9093,
458
+ "eval_samples_per_second": 16.499,
459
+ "eval_steps_per_second": 2.241,
460
  "eval_weighted avg": {
461
+ "f1-score": 0.8530559678148433,
462
+ "precision": 0.8530989190255094,
463
+ "recall": 0.8533731343283583,
464
+ "support": 33500.0
465
  },
466
  "step": 810
467
  },
468
  {
469
  "epoch": 11.0,
470
  "eval_Claim": {
471
+ "f1-score": 0.645658668208538,
472
+ "precision": 0.6161486113665624,
473
+ "recall": 0.6781376518218624,
474
+ "support": 4940.0
475
  },
476
  "eval_MajorClaim": {
477
+ "f1-score": 0.8331177231565331,
478
+ "precision": 0.7885714285714286,
479
+ "recall": 0.8829981718464351,
480
+ "support": 2188.0
481
  },
482
  "eval_O": {
483
+ "f1-score": 0.9059711059612445,
484
+ "precision": 0.9366843393148451,
485
+ "recall": 0.8772080588179128,
486
+ "support": 10473.0
487
  },
488
  "eval_Premise": {
489
+ "f1-score": 0.8902977542265961,
490
+ "precision": 0.8929452704840241,
491
+ "recall": 0.8876658909365369,
492
+ "support": 15899.0
493
  },
494
+ "eval_accuracy": 0.8531940298507462,
495
+ "eval_loss": 0.7852362394332886,
496
  "eval_macro avg": {
497
+ "f1-score": 0.818761312888228,
498
+ "precision": 0.8085874124342151,
499
+ "recall": 0.8315024433556868,
500
+ "support": 33500.0
501
+ },
502
+ "eval_runtime": 4.89,
503
+ "eval_samples_per_second": 16.565,
504
+ "eval_steps_per_second": 2.25,
505
  "eval_weighted avg": {
506
+ "f1-score": 0.8553879040715653,
507
+ "precision": 0.8589850855801485,
508
+ "recall": 0.8531940298507462,
509
+ "support": 33500.0
510
  },
511
  "step": 891
512
  },
513
  {
514
  "epoch": 12.0,
515
  "eval_Claim": {
516
+ "f1-score": 0.6329215627196947,
517
+ "precision": 0.6280645804265498,
518
+ "recall": 0.6378542510121458,
519
+ "support": 4940.0
520
  },
521
  "eval_MajorClaim": {
522
+ "f1-score": 0.8528368794326241,
523
+ "precision": 0.8278829604130808,
524
+ "recall": 0.8793418647166362,
525
+ "support": 2188.0
526
  },
527
  "eval_O": {
528
+ "f1-score": 0.9126755145220234,
529
+ "precision": 0.9193064031773709,
530
+ "recall": 0.9061395970591044,
531
+ "support": 10473.0
532
  },
533
  "eval_Premise": {
534
+ "f1-score": 0.8876634630534112,
535
+ "precision": 0.8894291487749432,
536
+ "recall": 0.88590477388515,
537
+ "support": 15899.0
538
  },
539
+ "eval_accuracy": 0.8552238805970149,
540
+ "eval_loss": 0.8324545621871948,
541
  "eval_macro avg": {
542
+ "f1-score": 0.8215243549319383,
543
+ "precision": 0.8161707731979861,
544
+ "recall": 0.827310121668259,
545
+ "support": 33500.0
546
+ },
547
+ "eval_runtime": 4.8939,
548
+ "eval_samples_per_second": 16.551,
549
+ "eval_steps_per_second": 2.248,
550
  "eval_weighted avg": {
551
+ "f1-score": 0.8556433335734092,
552
+ "precision": 0.8562082669116539,
553
+ "recall": 0.8552238805970149,
554
+ "support": 33500.0
555
  },
556
  "step": 972
557
  },
558
  {
559
  "epoch": 12.35,
560
+ "grad_norm": 2.6090729236602783,
561
  "learning_rate": 1.506172839506173e-05,
562
+ "loss": 0.0589,
563
  "step": 1000
564
  },
565
  {
566
  "epoch": 13.0,
567
  "eval_Claim": {
568
+ "f1-score": 0.6431152147731835,
569
+ "precision": 0.6377388535031847,
570
+ "recall": 0.648582995951417,
571
+ "support": 4940.0
572
  },
573
  "eval_MajorClaim": {
574
+ "f1-score": 0.8326953388876379,
575
+ "precision": 0.8206835330670218,
576
+ "recall": 0.8450639853747715,
577
+ "support": 2188.0
578
  },
579
  "eval_O": {
580
+ "f1-score": 0.904438383337466,
581
+ "precision": 0.9408790755262072,
582
+ "recall": 0.8707151723479424,
583
+ "support": 10473.0
584
  },
585
  "eval_Premise": {
586
+ "f1-score": 0.892013567684243,
587
+ "precision": 0.8749621922448733,
588
+ "recall": 0.9097427511164224,
589
+ "support": 15899.0
590
  },
591
+ "eval_accuracy": 0.8548059701492537,
592
+ "eval_loss": 0.8543176651000977,
593
  "eval_macro avg": {
594
+ "f1-score": 0.8180656261706326,
595
+ "precision": 0.8185659135853218,
596
+ "recall": 0.8185262261976383,
597
+ "support": 33500.0
598
+ },
599
+ "eval_runtime": 4.9338,
600
+ "eval_samples_per_second": 16.417,
601
+ "eval_steps_per_second": 2.23,
602
  "eval_weighted avg": {
603
+ "f1-score": 0.8553204019036041,
604
+ "precision": 0.8570428644520474,
605
+ "recall": 0.8548059701492537,
606
+ "support": 33500.0
607
  },
608
  "step": 1053
609
  }
 
613
  "num_input_tokens_seen": 0,
614
  "num_train_epochs": 50,
615
  "save_steps": 500,
616
+ "total_flos": 1863448390276200.0,
617
  "train_batch_size": 4,
618
  "trial_name": null,
619
  "trial_params": null
checkpoint-1053/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:19f05e33d3889e7c1ce1bf18bf9e16659f52e74109a304d618b5d09955736add
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da575726aa6fb9f10d3d3191c16916e8d9dc133cfc573c1d27486b346adcb4eb
3
  size 4856