Theoreticallyhugo commited on
Commit
4dc4655
1 Parent(s): 53dfca8

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 sha256:9bbc50dd4341e396f9b7177f9c5a155237a563d5fc0b172f02368f816a80ac3e
3
  size 592330980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84194d908c9ad309628cd303fe52aefabc438366a26375abb78b68fb24f032e7
3
  size 592330980
checkpoint-1053/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:92ffd46783202746f1e38e691b3392ba5509230d69c730e402239f86513e856e
3
  size 1014670074
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da0ac0e2ee89227b3fc9c98f8b7bd0f8e2e79cf77b6673056f89fe0429f05977
3
  size 1014670074
checkpoint-1053/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a2d0b6fd3e0e222c69d0c818ba4ff34a17c0844523a335fc5dcaffe1d7766cfb
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a8fed4ff2e0a1f5b060e9416aef90a5efb47a3301fd9fdfb8a444f0ad6d3aa3
3
  size 14244
checkpoint-1053/trainer_state.json CHANGED
@@ -11,384 +11,384 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
- "f1-score": 0.32421052631578945,
15
- "precision": 0.42777777777777776,
16
- "recall": 0.26101694915254237,
17
- "support": 295.0
18
  },
19
  "eval_B-MajorClaim": {
20
- "f1-score": 0.192090395480226,
21
- "precision": 0.8095238095238095,
22
- "recall": 0.10897435897435898,
23
- "support": 156.0
24
  },
25
  "eval_B-Premise": {
26
- "f1-score": 0.8566265060240965,
27
- "precision": 0.7620578778135049,
28
- "recall": 0.9779917469050894,
29
- "support": 727.0
30
  },
31
  "eval_I-Claim": {
32
- "f1-score": 0.5001390820584144,
33
- "precision": 0.5900886117492615,
34
- "recall": 0.43398503499879315,
35
- "support": 4143.0
36
  },
37
  "eval_I-MajorClaim": {
38
- "f1-score": 0.776470588235294,
39
- "precision": 0.6649333766655834,
40
- "recall": 0.93296853625171,
41
- "support": 2193.0
42
  },
43
  "eval_I-Premise": {
44
- "f1-score": 0.8861458374413376,
45
- "precision": 0.8781460059402845,
46
- "recall": 0.8942927644670858,
47
- "support": 12563.0
48
  },
49
  "eval_O": {
50
- "f1-score": 0.9987647926046939,
51
- "precision": 0.9977708781147998,
52
- "recall": 0.9997606892150607,
53
- "support": 12536.0
54
  },
55
- "eval_accuracy": 0.8713396498328887,
56
- "eval_loss": 0.3092103600502014,
57
  "eval_macro avg": {
58
- "f1-score": 0.647778246879979,
59
- "precision": 0.732899762512146,
60
- "recall": 0.6584271542806628,
61
- "support": 32613.0
62
- },
63
- "eval_runtime": 4.965,
64
- "eval_samples_per_second": 16.314,
65
- "eval_steps_per_second": 2.215,
66
  "eval_weighted avg": {
67
- "f1-score": 0.8639627599323451,
68
- "precision": 0.8662078389010613,
69
- "recall": 0.8713396498328887,
70
- "support": 32613.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
- "f1-score": 0.6186291739894552,
78
- "precision": 0.6423357664233577,
79
- "recall": 0.5966101694915255,
80
- "support": 295.0
81
  },
82
  "eval_B-MajorClaim": {
83
- "f1-score": 0.8113879003558718,
84
- "precision": 0.912,
85
- "recall": 0.7307692307692307,
86
- "support": 156.0
87
  },
88
  "eval_B-Premise": {
89
- "f1-score": 0.8871181938911022,
90
- "precision": 0.8575096277278562,
91
- "recall": 0.9188445667125172,
92
- "support": 727.0
93
  },
94
  "eval_I-Claim": {
95
- "f1-score": 0.656253149249219,
96
- "precision": 0.5633217993079584,
97
- "recall": 0.7859039343470915,
98
- "support": 4143.0
99
  },
100
  "eval_I-MajorClaim": {
101
- "f1-score": 0.8368159203980099,
102
- "precision": 0.9206349206349206,
103
- "recall": 0.7669858641130871,
104
- "support": 2193.0
105
  },
106
  "eval_I-Premise": {
107
- "f1-score": 0.8831027253668763,
108
- "precision": 0.933020288827855,
109
- "recall": 0.8382551938231314,
110
- "support": 12563.0
111
  },
112
  "eval_O": {
113
- "f1-score": 0.9998006141085457,
114
- "precision": 0.9996013077107089,
115
- "recall": 1.0,
116
- "support": 12536.0
117
  },
118
- "eval_accuracy": 0.8880814399165977,
119
- "eval_loss": 0.2581149935722351,
120
  "eval_macro avg": {
121
- "f1-score": 0.8133010967655829,
122
- "precision": 0.8326319586618081,
123
- "recall": 0.8053384227509405,
124
- "support": 32613.0
125
- },
126
- "eval_runtime": 5.0296,
127
- "eval_samples_per_second": 16.105,
128
- "eval_steps_per_second": 2.187,
129
  "eval_weighted avg": {
130
- "f1-score": 0.8933836566327042,
131
- "precision": 0.9064023864086126,
132
- "recall": 0.8880814399165977,
133
- "support": 32613.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
- "f1-score": 0.6576271186440678,
141
- "precision": 0.6576271186440678,
142
- "recall": 0.6576271186440678,
143
- "support": 295.0
144
  },
145
  "eval_B-MajorClaim": {
146
- "f1-score": 0.8673139158576051,
147
- "precision": 0.8758169934640523,
148
- "recall": 0.8589743589743589,
149
- "support": 156.0
150
  },
151
  "eval_B-Premise": {
152
- "f1-score": 0.8838487972508591,
153
- "precision": 0.8832417582417582,
154
- "recall": 0.8844566712517193,
155
- "support": 727.0
156
  },
157
  "eval_I-Claim": {
158
- "f1-score": 0.6468460321049062,
159
- "precision": 0.6083351052519669,
160
- "recall": 0.6905623944001931,
161
- "support": 4143.0
162
  },
163
  "eval_I-MajorClaim": {
164
- "f1-score": 0.8695855771482088,
165
- "precision": 0.8936477382098171,
166
- "recall": 0.8467852257181943,
167
- "support": 2193.0
168
  },
169
  "eval_I-Premise": {
170
- "f1-score": 0.8926256077795786,
171
- "precision": 0.9090533960551291,
172
- "recall": 0.8767810236408501,
173
- "support": 12563.0
174
  },
175
  "eval_O": {
176
- "f1-score": 0.9998803589232302,
177
- "precision": 0.9997607464710104,
178
- "recall": 1.0,
179
- "support": 12536.0
180
  },
181
- "eval_accuracy": 0.8965749854352559,
182
- "eval_loss": 0.2399493306875229,
183
  "eval_macro avg": {
184
- "f1-score": 0.8311039153869222,
185
- "precision": 0.8324975509054003,
186
- "recall": 0.8307409703756262,
187
- "support": 32613.0
188
- },
189
- "eval_runtime": 5.0024,
190
- "eval_samples_per_second": 16.192,
191
- "eval_steps_per_second": 2.199,
192
  "eval_weighted avg": {
193
- "f1-score": 0.8986385496019461,
194
- "precision": 0.9016737059631544,
195
- "recall": 0.8965749854352559,
196
- "support": 32613.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
- "f1-score": 0.6106346483704975,
204
- "precision": 0.6180555555555556,
205
- "recall": 0.6033898305084746,
206
- "support": 295.0
207
  },
208
  "eval_B-MajorClaim": {
209
- "f1-score": 0.8498583569405098,
210
- "precision": 0.7614213197969543,
211
- "recall": 0.9615384615384616,
212
- "support": 156.0
213
  },
214
  "eval_B-Premise": {
215
- "f1-score": 0.8746478873239436,
216
- "precision": 0.8961038961038961,
217
- "recall": 0.8541953232462174,
218
- "support": 727.0
219
  },
220
  "eval_I-Claim": {
221
- "f1-score": 0.5964912280701754,
222
- "precision": 0.6109845608706657,
223
- "recall": 0.5826695631185131,
224
- "support": 4143.0
225
  },
226
  "eval_I-MajorClaim": {
227
- "f1-score": 0.8631961259079903,
228
- "precision": 0.7741585233441911,
229
- "recall": 0.9753761969904241,
230
- "support": 2193.0
231
  },
232
  "eval_I-Premise": {
233
- "f1-score": 0.8933694290678411,
234
- "precision": 0.9071885770556376,
235
- "recall": 0.8799649765183475,
236
- "support": 12563.0
237
  },
238
  "eval_O": {
239
- "f1-score": 0.9999601132782897,
240
- "precision": 1.0,
241
- "recall": 0.9999202297383536,
242
- "support": 12536.0
243
  },
244
- "eval_accuracy": 0.8920369177935179,
245
- "eval_loss": 0.2815629839897156,
246
  "eval_macro avg": {
247
- "f1-score": 0.8125939698513209,
248
- "precision": 0.7954160618181286,
249
- "recall": 0.8367220830941131,
250
- "support": 32613.0
251
- },
252
- "eval_runtime": 4.9474,
253
- "eval_samples_per_second": 16.372,
254
- "eval_steps_per_second": 2.223,
255
  "eval_weighted avg": {
256
- "f1-score": 0.8914155863760895,
257
- "precision": 0.892730641712682,
258
- "recall": 0.8920369177935179,
259
- "support": 32613.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
- "f1-score": 0.6796116504854369,
267
- "precision": 0.6501547987616099,
268
- "recall": 0.711864406779661,
269
- "support": 295.0
270
  },
271
  "eval_B-MajorClaim": {
272
- "f1-score": 0.8746177370030581,
273
- "precision": 0.8362573099415205,
274
- "recall": 0.9166666666666666,
275
- "support": 156.0
276
  },
277
  "eval_B-Premise": {
278
- "f1-score": 0.8836879432624114,
279
- "precision": 0.9121522693997072,
280
- "recall": 0.8569463548830811,
281
- "support": 727.0
282
  },
283
  "eval_I-Claim": {
284
- "f1-score": 0.6557152635181382,
285
- "precision": 0.6216742375081116,
286
- "recall": 0.6937002172338885,
287
- "support": 4143.0
288
  },
289
  "eval_I-MajorClaim": {
290
- "f1-score": 0.8859746610357857,
291
- "precision": 0.8642671292281006,
292
- "recall": 0.9088007295941632,
293
- "support": 2193.0
294
  },
295
  "eval_I-Premise": {
296
- "f1-score": 0.8956745077255495,
297
- "precision": 0.9180177168644492,
298
- "recall": 0.874393058982727,
299
- "support": 12563.0
300
  },
301
  "eval_O": {
302
- "f1-score": 0.9998006141085457,
303
- "precision": 0.9996013077107089,
304
- "recall": 1.0,
305
- "support": 12536.0
306
  },
307
- "eval_accuracy": 0.9003771502161715,
308
- "eval_loss": 0.2959797978401184,
309
  "eval_macro avg": {
310
- "f1-score": 0.839297482448418,
311
- "precision": 0.8288749670591724,
312
- "recall": 0.8517673477343124,
313
- "support": 32613.0
314
- },
315
- "eval_runtime": 4.9768,
316
- "eval_samples_per_second": 16.275,
317
- "eval_steps_per_second": 2.21,
318
  "eval_weighted avg": {
319
- "f1-score": 0.9022413468876027,
320
- "precision": 0.9051721478141229,
321
- "recall": 0.9003771502161715,
322
- "support": 32613.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
- "f1-score": 0.6655518394648828,
330
- "precision": 0.6567656765676567,
331
- "recall": 0.6745762711864407,
332
- "support": 295.0
333
  },
334
  "eval_B-MajorClaim": {
335
- "f1-score": 0.8807339449541285,
336
- "precision": 0.8421052631578947,
337
- "recall": 0.9230769230769231,
338
- "support": 156.0
339
  },
340
  "eval_B-Premise": {
341
- "f1-score": 0.8812849162011174,
342
- "precision": 0.8950354609929078,
343
- "recall": 0.8679504814305364,
344
- "support": 727.0
345
  },
346
  "eval_I-Claim": {
347
- "f1-score": 0.634203994799669,
348
- "precision": 0.6213524779990737,
349
- "recall": 0.647598358677287,
350
- "support": 4143.0
351
  },
352
  "eval_I-MajorClaim": {
353
- "f1-score": 0.8926454445664106,
354
- "precision": 0.8607112616426758,
355
- "recall": 0.9270405836753306,
356
- "support": 2193.0
357
  },
358
  "eval_I-Premise": {
359
- "f1-score": 0.8911586596689544,
360
- "precision": 0.9041533546325878,
361
- "recall": 0.8785321977234737,
362
- "support": 12563.0
363
  },
364
  "eval_O": {
365
- "f1-score": 0.9995614559661922,
366
- "precision": 0.9991232964055152,
367
- "recall": 1.0,
368
- "support": 12536.0
369
  },
370
- "eval_accuracy": 0.8972802256768774,
371
- "eval_loss": 0.33718833327293396,
372
  "eval_macro avg": {
373
- "f1-score": 0.8350200365173363,
374
- "precision": 0.8256066844854731,
375
- "recall": 0.8455392593957132,
376
- "support": 32613.0
377
- },
378
- "eval_runtime": 4.9878,
379
- "eval_samples_per_second": 16.24,
380
- "eval_steps_per_second": 2.205,
381
  "eval_weighted avg": {
382
- "f1-score": 0.8979742337698349,
383
- "precision": 0.8990738793396261,
384
- "recall": 0.8972802256768774,
385
- "support": 32613.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
- "grad_norm": 10.722956657409668,
392
  "learning_rate": 1.7530864197530865e-05,
393
  "loss": 0.2434,
394
  "step": 500
@@ -396,448 +396,448 @@
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
- "f1-score": 0.6608695652173914,
400
- "precision": 0.6785714285714286,
401
- "recall": 0.6440677966101694,
402
- "support": 295.0
403
  },
404
  "eval_B-MajorClaim": {
405
- "f1-score": 0.875,
406
- "precision": 0.8536585365853658,
407
- "recall": 0.8974358974358975,
408
- "support": 156.0
409
  },
410
  "eval_B-Premise": {
411
- "f1-score": 0.8905608755129958,
412
- "precision": 0.8857142857142857,
413
- "recall": 0.8954607977991746,
414
- "support": 727.0
415
  },
416
  "eval_I-Claim": {
417
- "f1-score": 0.6306948332909138,
418
- "precision": 0.6670255720053836,
419
- "recall": 0.5981173062997828,
420
- "support": 4143.0
421
  },
422
  "eval_I-MajorClaim": {
423
- "f1-score": 0.890302764370338,
424
- "precision": 0.8579281183932347,
425
- "recall": 0.9252165982672138,
426
- "support": 2193.0
427
  },
428
  "eval_I-Premise": {
429
- "f1-score": 0.9037468972853708,
430
- "precision": 0.8947573724449992,
431
- "recall": 0.9129188888004458,
432
- "support": 12563.0
433
  },
434
  "eval_O": {
435
- "f1-score": 1.0,
436
  "precision": 1.0,
437
- "recall": 1.0,
438
- "support": 12536.0
439
  },
440
- "eval_accuracy": 0.9043326280930917,
441
- "eval_loss": 0.43479764461517334,
442
  "eval_macro avg": {
443
- "f1-score": 0.8358821336681442,
444
- "precision": 0.8339507591020997,
445
- "recall": 0.8390310407446692,
446
- "support": 32613.0
447
- },
448
- "eval_runtime": 4.9479,
449
- "eval_samples_per_second": 16.371,
450
- "eval_steps_per_second": 2.223,
451
  "eval_weighted avg": {
452
- "f1-score": 0.9025256555797259,
453
- "precision": 0.9014510706568353,
454
- "recall": 0.9043326280930917,
455
- "support": 32613.0
456
  },
457
  "step": 567
458
  },
459
  {
460
  "epoch": 8.0,
461
  "eval_B-Claim": {
462
- "f1-score": 0.6611295681063122,
463
- "precision": 0.6482084690553745,
464
- "recall": 0.6745762711864407,
465
- "support": 295.0
466
  },
467
  "eval_B-MajorClaim": {
468
- "f1-score": 0.8925081433224756,
469
- "precision": 0.9072847682119205,
470
- "recall": 0.8782051282051282,
471
- "support": 156.0
472
  },
473
  "eval_B-Premise": {
474
- "f1-score": 0.8810511756569848,
475
- "precision": 0.885952712100139,
476
- "recall": 0.8762035763411279,
477
- "support": 727.0
478
  },
479
  "eval_I-Claim": {
480
- "f1-score": 0.6376404494382022,
481
- "precision": 0.6189502385821404,
482
- "recall": 0.6574945691527878,
483
- "support": 4143.0
484
  },
485
  "eval_I-MajorClaim": {
486
- "f1-score": 0.8858131487889275,
487
- "precision": 0.896358543417367,
488
- "recall": 0.8755129958960328,
489
- "support": 2193.0
490
  },
491
  "eval_I-Premise": {
492
- "f1-score": 0.8946945982823662,
493
- "precision": 0.9022258195062728,
494
- "recall": 0.8872880681365916,
495
- "support": 12563.0
496
  },
497
  "eval_O": {
498
- "f1-score": 0.9999202361011406,
499
- "precision": 0.9998404849258254,
500
- "recall": 1.0,
501
- "support": 12536.0
502
  },
503
- "eval_accuracy": 0.8984147425873118,
504
- "eval_loss": 0.3937377631664276,
505
  "eval_macro avg": {
506
- "f1-score": 0.8361081885280585,
507
- "precision": 0.836974433685577,
508
- "recall": 0.8356115155597299,
509
- "support": 32613.0
510
- },
511
- "eval_runtime": 5.0071,
512
- "eval_samples_per_second": 16.177,
513
- "eval_steps_per_second": 2.197,
514
  "eval_weighted avg": {
515
- "f1-score": 0.8994624730308206,
516
- "precision": 0.9007308115646135,
517
- "recall": 0.8984147425873118,
518
- "support": 32613.0
519
  },
520
  "step": 648
521
  },
522
  {
523
  "epoch": 9.0,
524
  "eval_B-Claim": {
525
- "f1-score": 0.6821192052980132,
526
- "precision": 0.6666666666666666,
527
- "recall": 0.6983050847457627,
528
- "support": 295.0
529
  },
530
  "eval_B-MajorClaim": {
531
- "f1-score": 0.8510638297872339,
532
- "precision": 0.9523809523809523,
533
- "recall": 0.7692307692307693,
534
- "support": 156.0
535
  },
536
  "eval_B-Premise": {
537
- "f1-score": 0.8959891230455472,
538
- "precision": 0.885752688172043,
539
- "recall": 0.90646492434663,
540
- "support": 727.0
541
  },
542
  "eval_I-Claim": {
543
- "f1-score": 0.6547967670141736,
544
- "precision": 0.636094674556213,
545
- "recall": 0.6746319092445088,
546
- "support": 4143.0
547
  },
548
  "eval_I-MajorClaim": {
549
- "f1-score": 0.8507386653082017,
550
- "precision": 0.9636468551644547,
551
- "recall": 0.7615139078887369,
552
- "support": 2193.0
553
  },
554
  "eval_I-Premise": {
555
- "f1-score": 0.9060476867203537,
556
- "precision": 0.8987391338397682,
557
- "recall": 0.9134760805540078,
558
- "support": 12563.0
559
  },
560
  "eval_O": {
561
- "f1-score": 0.9999202361011406,
562
- "precision": 0.9998404849258254,
563
- "recall": 1.0,
564
- "support": 12536.0
565
  },
566
- "eval_accuracy": 0.9033820868978628,
567
- "eval_loss": 0.4608707129955292,
568
  "eval_macro avg": {
569
- "f1-score": 0.8343822161820948,
570
- "precision": 0.8575887793865605,
571
- "recall": 0.8176603822872022,
572
- "support": 32613.0
573
- },
574
- "eval_runtime": 4.9809,
575
- "eval_samples_per_second": 16.262,
576
- "eval_steps_per_second": 2.208,
577
  "eval_weighted avg": {
578
- "f1-score": 0.9039813964448579,
579
- "precision": 0.9064685293306464,
580
- "recall": 0.9033820868978628,
581
- "support": 32613.0
582
  },
583
  "step": 729
584
  },
585
  {
586
  "epoch": 10.0,
587
  "eval_B-Claim": {
588
- "f1-score": 0.6863354037267081,
589
- "precision": 0.6332378223495702,
590
- "recall": 0.7491525423728813,
591
- "support": 295.0
592
  },
593
  "eval_B-MajorClaim": {
594
- "f1-score": 0.8976897689768977,
595
- "precision": 0.9251700680272109,
596
- "recall": 0.8717948717948718,
597
- "support": 156.0
598
  },
599
  "eval_B-Premise": {
600
- "f1-score": 0.8780141843971632,
601
- "precision": 0.9062957540263543,
602
- "recall": 0.8514442916093535,
603
- "support": 727.0
604
  },
605
  "eval_I-Claim": {
606
- "f1-score": 0.6615143369175628,
607
- "precision": 0.6171368861024034,
608
- "recall": 0.7127685252232682,
609
- "support": 4143.0
610
  },
611
  "eval_I-MajorClaim": {
612
- "f1-score": 0.9019424292066465,
613
- "precision": 0.9264423076923077,
614
- "recall": 0.8787049703602371,
615
- "support": 2193.0
616
  },
617
  "eval_I-Premise": {
618
- "f1-score": 0.8934422897101273,
619
- "precision": 0.9130796077779625,
620
- "recall": 0.8746318554485394,
621
- "support": 12563.0
622
  },
623
  "eval_O": {
624
- "f1-score": 0.9999601132782897,
625
- "precision": 1.0,
626
- "recall": 0.9999202297383536,
627
- "support": 12536.0
628
  },
629
- "eval_accuracy": 0.9008370895041854,
630
- "eval_loss": 0.49426138401031494,
631
  "eval_macro avg": {
632
- "f1-score": 0.8455569323161993,
633
- "precision": 0.8459089208536871,
634
- "recall": 0.8483453266496436,
635
- "support": 32613.0
636
- },
637
- "eval_runtime": 4.962,
638
- "eval_samples_per_second": 16.324,
639
- "eval_steps_per_second": 2.217,
640
  "eval_weighted avg": {
641
- "f1-score": 0.9032978864532892,
642
- "precision": 0.9071693470023781,
643
- "recall": 0.9008370895041854,
644
- "support": 32613.0
645
  },
646
  "step": 810
647
  },
648
  {
649
  "epoch": 11.0,
650
  "eval_B-Claim": {
651
- "f1-score": 0.6654991243432573,
652
- "precision": 0.6884057971014492,
653
- "recall": 0.6440677966101694,
654
- "support": 295.0
655
  },
656
  "eval_B-MajorClaim": {
657
- "f1-score": 0.8888888888888888,
658
- "precision": 0.9361702127659575,
659
- "recall": 0.8461538461538461,
660
- "support": 156.0
661
  },
662
  "eval_B-Premise": {
663
- "f1-score": 0.8911290322580646,
664
- "precision": 0.871222076215506,
665
- "recall": 0.9119669876203577,
666
- "support": 727.0
667
  },
668
  "eval_I-Claim": {
669
- "f1-score": 0.6376628304034401,
670
- "precision": 0.6697662061636557,
671
- "recall": 0.6084962587496983,
672
- "support": 4143.0
673
  },
674
  "eval_I-MajorClaim": {
675
- "f1-score": 0.8884068450228971,
676
- "precision": 0.9422290388548057,
677
- "recall": 0.8404012767897857,
678
- "support": 2193.0
679
  },
680
  "eval_I-Premise": {
681
- "f1-score": 0.904972804972805,
682
- "precision": 0.8838885937618578,
683
- "recall": 0.9270874791053092,
684
- "support": 12563.0
685
  },
686
  "eval_O": {
687
- "f1-score": 0.9999202361011406,
688
- "precision": 0.9998404849258254,
689
- "recall": 1.0,
690
- "support": 12536.0
691
  },
692
- "eval_accuracy": 0.9055284702419281,
693
- "eval_loss": 0.5303362607955933,
694
  "eval_macro avg": {
695
- "f1-score": 0.8394971088557848,
696
- "precision": 0.8559317728270083,
697
- "recall": 0.8254533778613096,
698
- "support": 32613.0
699
- },
700
- "eval_runtime": 4.9749,
701
- "eval_samples_per_second": 16.282,
702
- "eval_steps_per_second": 2.211,
703
  "eval_weighted avg": {
704
- "f1-score": 0.9038459038093231,
705
- "precision": 0.9033803424876604,
706
- "recall": 0.9055284702419281,
707
- "support": 32613.0
708
  },
709
  "step": 891
710
  },
711
  {
712
  "epoch": 12.0,
713
  "eval_B-Claim": {
714
- "f1-score": 0.6374045801526718,
715
- "precision": 0.7292576419213974,
716
- "recall": 0.5661016949152542,
717
- "support": 295.0
718
  },
719
  "eval_B-MajorClaim": {
720
- "f1-score": 0.9114754098360656,
721
- "precision": 0.9328859060402684,
722
- "recall": 0.8910256410256411,
723
- "support": 156.0
724
  },
725
  "eval_B-Premise": {
726
- "f1-score": 0.8913612565445026,
727
- "precision": 0.850187265917603,
728
- "recall": 0.936726272352132,
729
- "support": 727.0
730
  },
731
  "eval_I-Claim": {
732
- "f1-score": 0.5983549421441516,
733
- "precision": 0.7082508250825083,
734
- "recall": 0.5179821385469466,
735
- "support": 4143.0
736
  },
737
  "eval_I-MajorClaim": {
738
- "f1-score": 0.9206057824690226,
739
- "precision": 0.9265588914549654,
740
- "recall": 0.9147286821705426,
741
- "support": 2193.0
742
  },
743
  "eval_I-Premise": {
744
- "f1-score": 0.9017204628501827,
745
- "precision": 0.8640309285870595,
746
- "recall": 0.9428480458489215,
747
- "support": 12563.0
748
  },
749
  "eval_O": {
750
- "f1-score": 0.9997606319317004,
751
  "precision": 1.0,
752
- "recall": 0.9995213784301212,
753
- "support": 12536.0
754
  },
755
- "eval_accuracy": 0.9049765430963113,
756
- "eval_loss": 0.5691992044448853,
757
  "eval_macro avg": {
758
- "f1-score": 0.8372404379897568,
759
- "precision": 0.8587387798576859,
760
- "recall": 0.8241334076127942,
761
- "support": 32613.0
762
- },
763
- "eval_runtime": 4.9632,
764
- "eval_samples_per_second": 16.32,
765
- "eval_steps_per_second": 2.216,
766
  "eval_weighted avg": {
767
- "f1-score": 0.8995624631750451,
768
- "precision": 0.8995123024912762,
769
- "recall": 0.9049765430963113,
770
- "support": 32613.0
771
  },
772
  "step": 972
773
  },
774
  {
775
  "epoch": 12.35,
776
- "grad_norm": 8.504526138305664,
777
  "learning_rate": 1.506172839506173e-05,
778
- "loss": 0.0319,
779
  "step": 1000
780
  },
781
  {
782
  "epoch": 13.0,
783
  "eval_B-Claim": {
784
- "f1-score": 0.6779661016949152,
785
- "precision": 0.6779661016949152,
786
- "recall": 0.6779661016949152,
787
- "support": 295.0
788
  },
789
  "eval_B-MajorClaim": {
790
- "f1-score": 0.9096774193548387,
791
- "precision": 0.9155844155844156,
792
- "recall": 0.9038461538461539,
793
- "support": 156.0
794
  },
795
  "eval_B-Premise": {
796
- "f1-score": 0.8873626373626373,
797
- "precision": 0.8861454046639232,
798
- "recall": 0.8885832187070152,
799
- "support": 727.0
800
  },
801
  "eval_I-Claim": {
802
- "f1-score": 0.663892288861689,
803
- "precision": 0.6734541842562701,
804
- "recall": 0.6545981173062998,
805
- "support": 4143.0
806
  },
807
  "eval_I-MajorClaim": {
808
- "f1-score": 0.9122568973315242,
809
- "precision": 0.9048900852400179,
810
- "recall": 0.9197446420428637,
811
- "support": 2193.0
812
  },
813
  "eval_I-Premise": {
814
- "f1-score": 0.9055930186433956,
815
- "precision": 0.9025855934213648,
816
- "recall": 0.9086205524158243,
817
- "support": 12563.0
818
  },
819
  "eval_O": {
820
- "f1-score": 0.9998404340194671,
821
  "precision": 1.0,
822
- "recall": 0.9996809189534142,
823
- "support": 12536.0
824
  },
825
- "eval_accuracy": 0.9095452733572502,
826
- "eval_loss": 0.5995727777481079,
827
  "eval_macro avg": {
828
- "f1-score": 0.8509412567526382,
829
- "precision": 0.8515179692658439,
830
- "recall": 0.8504342435666409,
831
- "support": 32613.0
832
- },
833
- "eval_runtime": 4.967,
834
- "eval_samples_per_second": 16.308,
835
- "eval_steps_per_second": 2.215,
836
  "eval_weighted avg": {
837
- "f1-score": 0.9091182110651892,
838
- "precision": 0.908741493590885,
839
- "recall": 0.9095452733572502,
840
- "support": 32613.0
841
  },
842
  "step": 1053
843
  }
@@ -847,7 +847,7 @@
847
  "num_input_tokens_seen": 0,
848
  "num_train_epochs": 50,
849
  "save_steps": 500,
850
- "total_flos": 1863475346187000.0,
851
  "train_batch_size": 4,
852
  "trial_name": null,
853
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
+ "f1-score": 0.18764302059496568,
15
+ "precision": 0.3416666666666667,
16
+ "recall": 0.12933753943217666,
17
+ "support": 317.0
18
  },
19
  "eval_B-MajorClaim": {
20
+ "f1-score": 0.19540229885057472,
21
+ "precision": 0.8947368421052632,
22
+ "recall": 0.10967741935483871,
23
+ "support": 155.0
24
  },
25
  "eval_B-Premise": {
26
+ "f1-score": 0.8476339053562143,
27
+ "precision": 0.740909090909091,
28
+ "recall": 0.9902794653705954,
29
+ "support": 823.0
30
  },
31
  "eval_I-Claim": {
32
+ "f1-score": 0.4315998169894769,
33
+ "precision": 0.6394035246272028,
34
+ "recall": 0.3257366482504604,
35
+ "support": 4344.0
36
  },
37
  "eval_I-MajorClaim": {
38
+ "f1-score": 0.7779799818016379,
39
+ "precision": 0.7493426818580193,
40
+ "recall": 0.8088930936613056,
41
+ "support": 2114.0
42
  },
43
  "eval_I-Premise": {
44
+ "f1-score": 0.8948683715038855,
45
+ "precision": 0.8376057421174058,
46
+ "recall": 0.9605350187403542,
47
+ "support": 13607.0
48
  },
49
  "eval_O": {
50
+ "f1-score": 0.9980147987727848,
51
+ "precision": 0.9970251509961237,
52
+ "recall": 0.9990064131514769,
53
+ "support": 11071.0
54
  },
55
+ "eval_accuracy": 0.8673183065585397,
56
+ "eval_loss": 0.3193369209766388,
57
  "eval_macro avg": {
58
+ "f1-score": 0.6190203134099342,
59
+ "precision": 0.7429556713256817,
60
+ "recall": 0.6176379425658869,
61
+ "support": 32431.0
62
+ },
63
+ "eval_runtime": 4.9083,
64
+ "eval_samples_per_second": 16.299,
65
+ "eval_steps_per_second": 2.037,
66
  "eval_weighted avg": {
67
+ "f1-score": 0.8489527906249091,
68
+ "precision": 0.8526967051825685,
69
+ "recall": 0.8673183065585397,
70
+ "support": 32431.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
+ "f1-score": 0.583941605839416,
78
+ "precision": 0.6926406926406926,
79
+ "recall": 0.5047318611987381,
80
+ "support": 317.0
81
  },
82
  "eval_B-MajorClaim": {
83
+ "f1-score": 0.8117647058823529,
84
+ "precision": 0.745945945945946,
85
+ "recall": 0.8903225806451613,
86
+ "support": 155.0
87
  },
88
  "eval_B-Premise": {
89
+ "f1-score": 0.8943661971830986,
90
+ "precision": 0.8649262202043133,
91
+ "recall": 0.9258809234507898,
92
+ "support": 823.0
93
  },
94
  "eval_I-Claim": {
95
+ "f1-score": 0.6197740770932805,
96
+ "precision": 0.6271506009898656,
97
+ "recall": 0.6125690607734806,
98
+ "support": 4344.0
99
  },
100
  "eval_I-MajorClaim": {
101
+ "f1-score": 0.8298737427776589,
102
+ "precision": 0.7577178585384916,
103
+ "recall": 0.9172185430463576,
104
+ "support": 2114.0
105
  },
106
  "eval_I-Premise": {
107
+ "f1-score": 0.9013697613125952,
108
+ "precision": 0.9106660666066607,
109
+ "recall": 0.8922613360770192,
110
+ "support": 13607.0
111
  },
112
  "eval_O": {
113
+ "f1-score": 0.9961487925331883,
114
+ "precision": 0.9993636363636363,
115
+ "recall": 0.9929545659831993,
116
+ "support": 11071.0
117
  },
118
+ "eval_accuracy": 0.887854213561099,
119
+ "eval_loss": 0.26407390832901,
120
  "eval_macro avg": {
121
+ "f1-score": 0.8053198403745129,
122
+ "precision": 0.7997730030413723,
123
+ "recall": 0.8194198387392494,
124
+ "support": 32431.0
125
+ },
126
+ "eval_runtime": 4.9329,
127
+ "eval_samples_per_second": 16.218,
128
+ "eval_steps_per_second": 2.027,
129
  "eval_weighted avg": {
130
+ "f1-score": 0.8876368199002541,
131
+ "precision": 0.8889201304482091,
132
+ "recall": 0.887854213561099,
133
+ "support": 32431.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
+ "f1-score": 0.5895522388059701,
141
+ "precision": 0.7214611872146118,
142
+ "recall": 0.49842271293375395,
143
+ "support": 317.0
144
  },
145
  "eval_B-MajorClaim": {
146
+ "f1-score": 0.8668941979522183,
147
+ "precision": 0.9202898550724637,
148
+ "recall": 0.8193548387096774,
149
+ "support": 155.0
150
  },
151
  "eval_B-Premise": {
152
+ "f1-score": 0.8929178470254956,
153
+ "precision": 0.8365180467091295,
154
+ "recall": 0.9574726609963548,
155
+ "support": 823.0
156
  },
157
  "eval_I-Claim": {
158
+ "f1-score": 0.6548140728684112,
159
+ "precision": 0.7178149876475433,
160
+ "recall": 0.6019797421731123,
161
+ "support": 4344.0
162
  },
163
  "eval_I-MajorClaim": {
164
+ "f1-score": 0.8815622622368754,
165
+ "precision": 0.9502460360852925,
166
+ "recall": 0.8221381267738883,
167
+ "support": 2114.0
168
  },
169
  "eval_I-Premise": {
170
+ "f1-score": 0.9169943670953343,
171
+ "precision": 0.8852257181942544,
172
+ "recall": 0.9511280958330272,
173
+ "support": 13607.0
174
  },
175
  "eval_O": {
176
+ "f1-score": 0.9984170774727511,
177
+ "precision": 0.9998188405797102,
178
+ "recall": 0.9970192394544305,
179
+ "support": 11071.0
180
  },
181
+ "eval_accuracy": 0.9067250470229101,
182
+ "eval_loss": 0.25323203206062317,
183
  "eval_macro avg": {
184
+ "f1-score": 0.8287360090652937,
185
+ "precision": 0.8616249530718579,
186
+ "recall": 0.8067879166963207,
187
+ "support": 32431.0
188
+ },
189
+ "eval_runtime": 4.9025,
190
+ "eval_samples_per_second": 16.318,
191
+ "eval_steps_per_second": 2.04,
192
  "eval_weighted avg": {
193
+ "f1-score": 0.9033110548907601,
194
+ "precision": 0.9034896128542845,
195
+ "recall": 0.9067250470229101,
196
+ "support": 32431.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
+ "f1-score": 0.6806136680613668,
204
+ "precision": 0.61,
205
+ "recall": 0.7697160883280757,
206
+ "support": 317.0
207
  },
208
  "eval_B-MajorClaim": {
209
+ "f1-score": 0.8861538461538462,
210
+ "precision": 0.8470588235294118,
211
+ "recall": 0.9290322580645162,
212
+ "support": 155.0
213
  },
214
  "eval_B-Premise": {
215
+ "f1-score": 0.8687258687258688,
216
+ "precision": 0.9233926128590971,
217
+ "recall": 0.8201701093560145,
218
+ "support": 823.0
219
  },
220
  "eval_I-Claim": {
221
+ "f1-score": 0.656155854589275,
222
+ "precision": 0.5819380121125757,
223
+ "recall": 0.7520718232044199,
224
+ "support": 4344.0
225
  },
226
  "eval_I-MajorClaim": {
227
+ "f1-score": 0.8818897637795275,
228
+ "precision": 0.8408408408408409,
229
+ "recall": 0.9271523178807947,
230
+ "support": 2114.0
231
  },
232
  "eval_I-Premise": {
233
+ "f1-score": 0.8829758349171871,
234
+ "precision": 0.9349433218334154,
235
+ "recall": 0.836481222899978,
236
+ "support": 13607.0
237
  },
238
  "eval_O": {
239
+ "f1-score": 0.9971017118014671,
240
+ "precision": 0.9998183634547271,
241
+ "recall": 0.9943997832174148,
242
+ "support": 11071.0
243
  },
244
+ "eval_accuracy": 0.8843698930036077,
245
+ "eval_loss": 0.2791365385055542,
246
  "eval_macro avg": {
247
+ "f1-score": 0.8362309354326484,
248
+ "precision": 0.8197131392328668,
249
+ "recall": 0.8612890861358877,
250
+ "support": 32431.0
251
+ },
252
+ "eval_runtime": 4.9234,
253
+ "eval_samples_per_second": 16.249,
254
+ "eval_steps_per_second": 2.031,
255
  "eval_weighted avg": {
256
+ "f1-score": 0.889158243622726,
257
+ "precision": 0.8997827813567029,
258
+ "recall": 0.8843698930036077,
259
+ "support": 32431.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
+ "f1-score": 0.6078799249530958,
267
+ "precision": 0.75,
268
+ "recall": 0.5110410094637224,
269
+ "support": 317.0
270
  },
271
  "eval_B-MajorClaim": {
272
+ "f1-score": 0.8971962616822431,
273
+ "precision": 0.8674698795180723,
274
+ "recall": 0.9290322580645162,
275
+ "support": 155.0
276
  },
277
  "eval_B-Premise": {
278
+ "f1-score": 0.893739230327398,
279
+ "precision": 0.8474945533769063,
280
+ "recall": 0.945321992709599,
281
+ "support": 823.0
282
  },
283
  "eval_I-Claim": {
284
+ "f1-score": 0.6119813135476778,
285
+ "precision": 0.7590320381731425,
286
+ "recall": 0.5126611418047882,
287
+ "support": 4344.0
288
  },
289
  "eval_I-MajorClaim": {
290
+ "f1-score": 0.8889393595275948,
291
+ "precision": 0.854958497160332,
292
+ "recall": 0.9257332071901608,
293
+ "support": 2114.0
294
  },
295
  "eval_I-Premise": {
296
+ "f1-score": 0.9137743463765573,
297
+ "precision": 0.874462654486835,
298
+ "recall": 0.956786947894466,
299
+ "support": 13607.0
300
  },
301
  "eval_O": {
302
+ "f1-score": 0.9976913675252366,
303
+ "precision": 1.0,
304
+ "recall": 0.9953933700659381,
305
+ "support": 11071.0
306
  },
307
+ "eval_accuracy": 0.9036724121982054,
308
+ "eval_loss": 0.3500836193561554,
309
  "eval_macro avg": {
310
+ "f1-score": 0.8301716862771149,
311
+ "precision": 0.8504882318164697,
312
+ "recall": 0.8251385610275985,
313
+ "support": 32431.0
314
+ },
315
+ "eval_runtime": 4.9239,
316
+ "eval_samples_per_second": 16.247,
317
+ "eval_steps_per_second": 2.031,
318
  "eval_weighted avg": {
319
+ "f1-score": 0.8968007164885599,
320
+ "precision": 0.8986502613295283,
321
+ "recall": 0.9036724121982054,
322
+ "support": 32431.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
+ "f1-score": 0.672077922077922,
330
+ "precision": 0.6923076923076923,
331
+ "recall": 0.6529968454258676,
332
+ "support": 317.0
333
  },
334
  "eval_B-MajorClaim": {
335
+ "f1-score": 0.8910256410256411,
336
+ "precision": 0.8853503184713376,
337
+ "recall": 0.896774193548387,
338
+ "support": 155.0
339
  },
340
  "eval_B-Premise": {
341
+ "f1-score": 0.8907563025210083,
342
+ "precision": 0.8801897983392646,
343
+ "recall": 0.9015795868772782,
344
+ "support": 823.0
345
  },
346
  "eval_I-Claim": {
347
+ "f1-score": 0.6666666666666667,
348
+ "precision": 0.6998228296633764,
349
+ "recall": 0.6365101289134438,
350
+ "support": 4344.0
351
  },
352
  "eval_I-MajorClaim": {
353
+ "f1-score": 0.8853370396108408,
354
+ "precision": 0.867453472537449,
355
+ "recall": 0.9039735099337748,
356
+ "support": 2114.0
357
  },
358
  "eval_I-Premise": {
359
+ "f1-score": 0.9098497495826378,
360
+ "precision": 0.8987595898759589,
361
+ "recall": 0.9212170206511354,
362
+ "support": 13607.0
363
  },
364
  "eval_O": {
365
+ "f1-score": 0.998190209030857,
366
+ "precision": 1.0,
367
+ "recall": 0.9963869569144612,
368
+ "support": 11071.0
369
  },
370
+ "eval_accuracy": 0.9043816101877833,
371
+ "eval_loss": 0.30672115087509155,
372
  "eval_macro avg": {
373
+ "f1-score": 0.8448433615022248,
374
+ "precision": 0.8462691001707254,
375
+ "recall": 0.8442054631806213,
376
+ "support": 32431.0
377
+ },
378
+ "eval_runtime": 4.9018,
379
+ "eval_samples_per_second": 16.321,
380
+ "eval_steps_per_second": 2.04,
381
  "eval_weighted avg": {
382
+ "f1-score": 0.9029367568413356,
383
+ "precision": 0.9020793621628027,
384
+ "recall": 0.9043816101877833,
385
+ "support": 32431.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
+ "grad_norm": 0.43724608421325684,
392
  "learning_rate": 1.7530864197530865e-05,
393
  "loss": 0.2434,
394
  "step": 500
 
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
+ "f1-score": 0.6763285024154589,
400
+ "precision": 0.6907894736842105,
401
+ "recall": 0.6624605678233438,
402
+ "support": 317.0
403
  },
404
  "eval_B-MajorClaim": {
405
+ "f1-score": 0.8952380952380952,
406
+ "precision": 0.88125,
407
+ "recall": 0.9096774193548387,
408
+ "support": 155.0
409
  },
410
  "eval_B-Premise": {
411
+ "f1-score": 0.8941034897713597,
412
+ "precision": 0.8855780691299165,
413
+ "recall": 0.9027946537059538,
414
+ "support": 823.0
415
  },
416
  "eval_I-Claim": {
417
+ "f1-score": 0.6960393719240684,
418
+ "precision": 0.7088305489260143,
419
+ "recall": 0.6837016574585635,
420
+ "support": 4344.0
421
  },
422
  "eval_I-MajorClaim": {
423
+ "f1-score": 0.9140037593984962,
424
+ "precision": 0.9080298786181139,
425
+ "recall": 0.9200567644276254,
426
+ "support": 2114.0
427
  },
428
  "eval_I-Premise": {
429
+ "f1-score": 0.9164355568530141,
430
+ "precision": 0.9101253895774444,
431
+ "recall": 0.9228338355258323,
432
+ "support": 13607.0
433
  },
434
  "eval_O": {
435
+ "f1-score": 0.9967376529225193,
436
  "precision": 1.0,
437
+ "recall": 0.9934965224460302,
438
+ "support": 11071.0
439
  },
440
+ "eval_accuracy": 0.9116277635595572,
441
+ "eval_loss": 0.34577852487564087,
442
  "eval_macro avg": {
443
+ "f1-score": 0.855555204074716,
444
+ "precision": 0.8549433371336713,
445
+ "recall": 0.8564316315345982,
446
+ "support": 32431.0
447
+ },
448
+ "eval_runtime": 4.8989,
449
+ "eval_samples_per_second": 16.33,
450
+ "eval_steps_per_second": 2.041,
451
  "eval_weighted avg": {
452
+ "f1-score": 0.9111536914902465,
453
+ "precision": 0.910801887328957,
454
+ "recall": 0.9116277635595572,
455
+ "support": 32431.0
456
  },
457
  "step": 567
458
  },
459
  {
460
  "epoch": 8.0,
461
  "eval_B-Claim": {
462
+ "f1-score": 0.701923076923077,
463
+ "precision": 0.7133550488599348,
464
+ "recall": 0.6908517350157729,
465
+ "support": 317.0
466
  },
467
  "eval_B-MajorClaim": {
468
+ "f1-score": 0.9260450160771705,
469
+ "precision": 0.9230769230769231,
470
+ "recall": 0.9290322580645162,
471
+ "support": 155.0
472
  },
473
  "eval_B-Premise": {
474
+ "f1-score": 0.8985507246376812,
475
+ "precision": 0.8931572629051621,
476
+ "recall": 0.9040097205346294,
477
+ "support": 823.0
478
  },
479
  "eval_I-Claim": {
480
+ "f1-score": 0.6915371329879102,
481
+ "precision": 0.6917760884588804,
482
+ "recall": 0.6912983425414365,
483
+ "support": 4344.0
484
  },
485
  "eval_I-MajorClaim": {
486
+ "f1-score": 0.9147286821705426,
487
+ "precision": 0.9085394307046197,
488
+ "recall": 0.9210028382213813,
489
+ "support": 2114.0
490
  },
491
  "eval_I-Premise": {
492
+ "f1-score": 0.912640635340834,
493
+ "precision": 0.9131778382753293,
494
+ "recall": 0.9121040640846623,
495
+ "support": 13607.0
496
  },
497
  "eval_O": {
498
+ "f1-score": 0.9993222177036736,
499
+ "precision": 0.9998191681735985,
500
+ "recall": 0.9988257609971999,
501
+ "support": 11071.0
502
  },
503
+ "eval_accuracy": 0.9104252104467947,
504
+ "eval_loss": 0.3896510601043701,
505
  "eval_macro avg": {
506
+ "f1-score": 0.8635353551201269,
507
+ "precision": 0.863271680064921,
508
+ "recall": 0.8638749599227997,
509
+ "support": 32431.0
510
+ },
511
+ "eval_runtime": 4.9786,
512
+ "eval_samples_per_second": 16.069,
513
+ "eval_steps_per_second": 2.009,
514
  "eval_weighted avg": {
515
+ "f1-score": 0.9103982293589928,
516
+ "precision": 0.9103825180220693,
517
+ "recall": 0.9104252104467947,
518
+ "support": 32431.0
519
  },
520
  "step": 648
521
  },
522
  {
523
  "epoch": 9.0,
524
  "eval_B-Claim": {
525
+ "f1-score": 0.6862745098039215,
526
+ "precision": 0.6171284634760705,
527
+ "recall": 0.7728706624605678,
528
+ "support": 317.0
529
  },
530
  "eval_B-MajorClaim": {
531
+ "f1-score": 0.9119496855345912,
532
+ "precision": 0.8895705521472392,
533
+ "recall": 0.9354838709677419,
534
+ "support": 155.0
535
  },
536
  "eval_B-Premise": {
537
+ "f1-score": 0.8690095846645367,
538
+ "precision": 0.9164420485175202,
539
+ "recall": 0.8262454434993924,
540
+ "support": 823.0
541
  },
542
  "eval_I-Claim": {
543
+ "f1-score": 0.6841291120965247,
544
+ "precision": 0.619652531290865,
545
+ "recall": 0.7635819521178637,
546
+ "support": 4344.0
547
  },
548
  "eval_I-MajorClaim": {
549
+ "f1-score": 0.9025385312783318,
550
+ "precision": 0.8664055700609226,
551
+ "recall": 0.9418164616840113,
552
+ "support": 2114.0
553
  },
554
  "eval_I-Premise": {
555
+ "f1-score": 0.8949834831374357,
556
+ "precision": 0.9374748531423513,
557
+ "recall": 0.8561769677371941,
558
+ "support": 13607.0
559
  },
560
  "eval_O": {
561
+ "f1-score": 0.9989151071331707,
562
+ "precision": 0.9998190209030857,
563
+ "recall": 0.9980128263029536,
564
+ "support": 11071.0
565
  },
566
+ "eval_accuracy": 0.8965804323024267,
567
+ "eval_loss": 0.4196818470954895,
568
  "eval_macro avg": {
569
+ "f1-score": 0.8496857162355017,
570
+ "precision": 0.8352132913625792,
571
+ "recall": 0.8705983121099606,
572
+ "support": 32431.0
573
+ },
574
+ "eval_runtime": 4.9285,
575
+ "eval_samples_per_second": 16.232,
576
+ "eval_steps_per_second": 2.029,
577
  "eval_weighted avg": {
578
+ "f1-score": 0.9000940098435676,
579
+ "precision": 0.9076597589527613,
580
+ "recall": 0.8965804323024267,
581
+ "support": 32431.0
582
  },
583
  "step": 729
584
  },
585
  {
586
  "epoch": 10.0,
587
  "eval_B-Claim": {
588
+ "f1-score": 0.6905537459283387,
589
+ "precision": 0.7138047138047138,
590
+ "recall": 0.668769716088328,
591
+ "support": 317.0
592
  },
593
  "eval_B-MajorClaim": {
594
+ "f1-score": 0.9142857142857143,
595
+ "precision": 0.9,
596
+ "recall": 0.9290322580645162,
597
+ "support": 155.0
598
  },
599
  "eval_B-Premise": {
600
+ "f1-score": 0.894484412470024,
601
+ "precision": 0.8828402366863906,
602
+ "recall": 0.9064398541919806,
603
+ "support": 823.0
604
  },
605
  "eval_I-Claim": {
606
+ "f1-score": 0.681682755333095,
607
+ "precision": 0.7066963182604399,
608
+ "recall": 0.6583793738489871,
609
+ "support": 4344.0
610
  },
611
  "eval_I-MajorClaim": {
612
+ "f1-score": 0.9083702494754021,
613
+ "precision": 0.895632183908046,
614
+ "recall": 0.9214758751182592,
615
+ "support": 2114.0
616
  },
617
  "eval_I-Premise": {
618
+ "f1-score": 0.9129089784183134,
619
+ "precision": 0.904253785147801,
620
+ "recall": 0.9217314617476299,
621
+ "support": 13607.0
622
  },
623
  "eval_O": {
624
+ "f1-score": 0.9983716301791207,
625
+ "precision": 0.999909395669113,
626
+ "recall": 0.9968385873001535,
627
+ "support": 11071.0
628
  },
629
+ "eval_accuracy": 0.9092534920292313,
630
+ "eval_loss": 0.5311424136161804,
631
  "eval_macro avg": {
632
+ "f1-score": 0.8572367837271441,
633
+ "precision": 0.8575909476395005,
634
+ "recall": 0.857523875194265,
635
+ "support": 32431.0
636
+ },
637
+ "eval_runtime": 4.8937,
638
+ "eval_samples_per_second": 16.348,
639
+ "eval_steps_per_second": 2.043,
640
  "eval_weighted avg": {
641
+ "f1-score": 0.9081813659228948,
642
+ "precision": 0.9074585127295745,
643
+ "recall": 0.9092534920292313,
644
+ "support": 32431.0
645
  },
646
  "step": 810
647
  },
648
  {
649
  "epoch": 11.0,
650
  "eval_B-Claim": {
651
+ "f1-score": 0.680921052631579,
652
+ "precision": 0.711340206185567,
653
+ "recall": 0.6529968454258676,
654
+ "support": 317.0
655
  },
656
  "eval_B-MajorClaim": {
657
+ "f1-score": 0.9009584664536742,
658
+ "precision": 0.8924050632911392,
659
+ "recall": 0.9096774193548387,
660
+ "support": 155.0
661
  },
662
  "eval_B-Premise": {
663
+ "f1-score": 0.8942020322773461,
664
+ "precision": 0.88,
665
+ "recall": 0.9088699878493317,
666
+ "support": 823.0
667
  },
668
  "eval_I-Claim": {
669
+ "f1-score": 0.6838570400575678,
670
+ "precision": 0.713820731096645,
671
+ "recall": 0.6563075506445673,
672
+ "support": 4344.0
673
  },
674
  "eval_I-MajorClaim": {
675
+ "f1-score": 0.902502979737783,
676
+ "precision": 0.9096588178760211,
677
+ "recall": 0.8954588457899716,
678
+ "support": 2114.0
679
  },
680
  "eval_I-Premise": {
681
+ "f1-score": 0.9134434303925829,
682
+ "precision": 0.9004641199571581,
683
+ "recall": 0.9268023811273609,
684
+ "support": 13607.0
685
  },
686
  "eval_O": {
687
+ "f1-score": 0.9988699543461556,
688
+ "precision": 0.999728555917481,
689
+ "recall": 0.9980128263029536,
690
+ "support": 11071.0
691
  },
692
+ "eval_accuracy": 0.9096235083716198,
693
+ "eval_loss": 0.49213308095932007,
694
  "eval_macro avg": {
695
+ "f1-score": 0.8535364222709555,
696
+ "precision": 0.8582024991891446,
697
+ "recall": 0.8497322652135558,
698
+ "support": 32431.0
699
+ },
700
+ "eval_runtime": 4.9244,
701
+ "eval_samples_per_second": 16.246,
702
+ "eval_steps_per_second": 2.031,
703
  "eval_weighted avg": {
704
+ "f1-score": 0.9083194817194259,
705
+ "precision": 0.9075428987655072,
706
+ "recall": 0.9096235083716198,
707
+ "support": 32431.0
708
  },
709
  "step": 891
710
  },
711
  {
712
  "epoch": 12.0,
713
  "eval_B-Claim": {
714
+ "f1-score": 0.6577181208053691,
715
+ "precision": 0.7025089605734767,
716
+ "recall": 0.6182965299684543,
717
+ "support": 317.0
718
  },
719
  "eval_B-MajorClaim": {
720
+ "f1-score": 0.8978328173374612,
721
+ "precision": 0.8630952380952381,
722
+ "recall": 0.9354838709677419,
723
+ "support": 155.0
724
  },
725
  "eval_B-Premise": {
726
+ "f1-score": 0.8902147971360382,
727
+ "precision": 0.8745603751465416,
728
+ "recall": 0.9064398541919806,
729
+ "support": 823.0
730
  },
731
  "eval_I-Claim": {
732
+ "f1-score": 0.6518804243008679,
733
+ "precision": 0.6842105263157895,
734
+ "recall": 0.6224677716390423,
735
+ "support": 4344.0
736
  },
737
  "eval_I-MajorClaim": {
738
+ "f1-score": 0.8919829328542555,
739
+ "precision": 0.8490808037622916,
740
+ "recall": 0.9394512771996215,
741
+ "support": 2114.0
742
  },
743
  "eval_I-Premise": {
744
+ "f1-score": 0.9075452422650322,
745
+ "precision": 0.9011665821317296,
746
+ "recall": 0.9140148453002132,
747
+ "support": 13607.0
748
  },
749
  "eval_O": {
750
+ "f1-score": 0.9985526910900046,
751
  "precision": 1.0,
752
+ "recall": 0.9971095655315689,
753
+ "support": 11071.0
754
  },
755
+ "eval_accuracy": 0.9020073386574574,
756
+ "eval_loss": 0.5465030670166016,
757
  "eval_macro avg": {
758
+ "f1-score": 0.8422467179698613,
759
+ "precision": 0.8392317837178667,
760
+ "recall": 0.8476091021140889,
761
+ "support": 32431.0
762
+ },
763
+ "eval_runtime": 4.8988,
764
+ "eval_samples_per_second": 16.331,
765
+ "eval_steps_per_second": 2.041,
766
  "eval_weighted avg": {
767
+ "f1-score": 0.9004246846165758,
768
+ "precision": 0.8996509302731299,
769
+ "recall": 0.9020073386574574,
770
+ "support": 32431.0
771
  },
772
  "step": 972
773
  },
774
  {
775
  "epoch": 12.35,
776
+ "grad_norm": 0.1318705528974533,
777
  "learning_rate": 1.506172839506173e-05,
778
+ "loss": 0.0258,
779
  "step": 1000
780
  },
781
  {
782
  "epoch": 13.0,
783
  "eval_B-Claim": {
784
+ "f1-score": 0.6857142857142857,
785
+ "precision": 0.6551724137931034,
786
+ "recall": 0.7192429022082019,
787
+ "support": 317.0
788
  },
789
  "eval_B-MajorClaim": {
790
+ "f1-score": 0.9022082018927444,
791
+ "precision": 0.8827160493827161,
792
+ "recall": 0.9225806451612903,
793
+ "support": 155.0
794
  },
795
  "eval_B-Premise": {
796
+ "f1-score": 0.8858560794044665,
797
+ "precision": 0.9049429657794676,
798
+ "recall": 0.8675577156743621,
799
+ "support": 823.0
800
  },
801
  "eval_I-Claim": {
802
+ "f1-score": 0.6909411894516165,
803
+ "precision": 0.6634880271243908,
804
+ "recall": 0.7207642725598526,
805
+ "support": 4344.0
806
  },
807
  "eval_I-MajorClaim": {
808
+ "f1-score": 0.9048498845265589,
809
+ "precision": 0.8840252707581228,
810
+ "recall": 0.9266792809839167,
811
+ "support": 2114.0
812
  },
813
  "eval_I-Premise": {
814
+ "f1-score": 0.906777130915062,
815
+ "precision": 0.9211464098870271,
816
+ "recall": 0.8928492687587272,
817
+ "support": 13607.0
818
  },
819
  "eval_O": {
820
+ "f1-score": 0.9971466099008107,
821
  "precision": 1.0,
822
+ "recall": 0.9943094571402764,
823
+ "support": 11071.0
824
  },
825
+ "eval_accuracy": 0.9044432795781814,
826
+ "eval_loss": 0.5739177465438843,
827
  "eval_macro avg": {
828
+ "f1-score": 0.8533561974007922,
829
+ "precision": 0.8444987338178326,
830
+ "recall": 0.8634262203552325,
831
+ "support": 32431.0
832
+ },
833
+ "eval_runtime": 4.929,
834
+ "eval_samples_per_second": 16.231,
835
+ "eval_steps_per_second": 2.029,
836
  "eval_weighted avg": {
837
+ "f1-score": 0.9058771229581195,
838
+ "precision": 0.9079380628166581,
839
+ "recall": 0.9044432795781814,
840
+ "support": 32431.0
841
  },
842
  "step": 1053
843
  }
 
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,
852
  "trial_name": null,
853
  "trial_params": null
checkpoint-1053/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5eff186592c44100b7fd5c301c21b9eebfc7f8bd335567cdc6ffe4a4d32ec927
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2f0bdafd19c0347fe87a3f27a2ea54a847df1f75da4e3bbd7bf587c614e2949
3
  size 4920