Theoreticallyhugo commited on
Commit
fb9f59e
1 Parent(s): 7fd424b

Training in progress, epoch 11, checkpoint

Browse files
checkpoint-891/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:08ee1b848b028c6fad033d9e4da3a6d47e348ee80522894a6a58dd5d0e0b98b4
3
  size 592324828
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bab4ec2e30336e4d01f9d4e3daef4f19d3868176376ae26f795c6c66ab40c99
3
  size 592324828
checkpoint-891/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fc84929172846bc56238eb205ab49a6401efe231994793b80ff34fc98c056ea7
3
  size 1014657786
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:afd931442b94db0c3898dd957552e64e03d70fec9206cb6f8ad404a6c2ca3ac9
3
  size 1014657786
checkpoint-891/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0bb7364fd68593f88aeabea076c26b75fece7518f250d191da2d08668972c0aa
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93f9c336c989c02801f193daf66231cc7b0216d42e4d84579a8bf465e525943e
3
  size 14244
checkpoint-891/trainer_state.json CHANGED
@@ -11,502 +11,502 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_Claim": {
14
- "f1-score": 0.46243291592128805,
15
- "precision": 0.6588785046728972,
16
- "recall": 0.3562241616903996,
17
- "support": 4354.0
18
  },
19
  "eval_MajorClaim": {
20
- "f1-score": 0.8342621259029929,
21
- "precision": 0.8096955128205128,
22
- "recall": 0.8603661132396765,
23
- "support": 2349.0
24
  },
25
  "eval_O": {
26
- "f1-score": 0.9993622448979592,
27
- "precision": 0.9987253027405991,
28
  "recall": 1.0,
29
- "support": 12536.0
30
  },
31
  "eval_Premise": {
32
- "f1-score": 0.8941052999825083,
33
- "precision": 0.8401157057392676,
34
- "recall": 0.955510692388216,
35
- "support": 13374.0
36
  },
37
- "eval_accuracy": 0.8857510808573268,
38
- "eval_loss": 0.2754688858985901,
39
  "eval_macro avg": {
40
- "f1-score": 0.7975406466761871,
41
- "precision": 0.8268537564933192,
42
- "recall": 0.793025241829573,
43
- "support": 32613.0
44
- },
45
- "eval_runtime": 4.8781,
46
- "eval_samples_per_second": 16.605,
47
- "eval_steps_per_second": 2.255,
48
  "eval_weighted avg": {
49
- "f1-score": 0.8726239240080429,
50
- "precision": 0.8746959682542021,
51
- "recall": 0.8857510808573268,
52
- "support": 32613.0
53
  },
54
  "step": 81
55
  },
56
  {
57
  "epoch": 2.0,
58
  "eval_Claim": {
59
- "f1-score": 0.6446312571646924,
60
- "precision": 0.5518482172064115,
61
- "recall": 0.77491961414791,
62
- "support": 4354.0
63
  },
64
  "eval_MajorClaim": {
65
- "f1-score": 0.8912341158733577,
66
- "precision": 0.9019180470793374,
67
- "recall": 0.8808003405704555,
68
- "support": 2349.0
69
  },
70
  "eval_O": {
71
- "f1-score": 0.9998005504806733,
72
- "precision": 0.9999202106439001,
73
- "recall": 0.9996809189534142,
74
- "support": 12536.0
75
  },
76
  "eval_Premise": {
77
- "f1-score": 0.8690409646250897,
78
- "precision": 0.9324023303632625,
79
- "recall": 0.8137430835950351,
80
- "support": 13374.0
81
  },
82
- "eval_accuracy": 0.8848618649004998,
83
- "eval_loss": 0.25178349018096924,
84
  "eval_macro avg": {
85
- "f1-score": 0.8511767220359533,
86
- "precision": 0.8465222013232279,
87
- "recall": 0.8672859893167036,
88
- "support": 32613.0
89
- },
90
- "eval_runtime": 4.907,
91
- "eval_samples_per_second": 16.507,
92
- "eval_steps_per_second": 2.242,
93
  "eval_weighted avg": {
94
- "f1-score": 0.8909418634778542,
95
- "precision": 0.9053537287957648,
96
- "recall": 0.8848618649004998,
97
- "support": 32613.0
98
  },
99
  "step": 162
100
  },
101
  {
102
  "epoch": 3.0,
103
  "eval_Claim": {
104
- "f1-score": 0.6598764019226366,
105
- "precision": 0.6576186131386861,
106
- "recall": 0.6621497473587505,
107
- "support": 4354.0
108
  },
109
  "eval_MajorClaim": {
110
- "f1-score": 0.9062119366626066,
111
- "precision": 0.8661233993015134,
112
- "recall": 0.9501915708812261,
113
- "support": 2349.0
114
  },
115
  "eval_O": {
116
- "f1-score": 0.9998404722022812,
117
- "precision": 0.9997607273887382,
118
- "recall": 0.9999202297383536,
119
- "support": 12536.0
120
  },
121
  "eval_Premise": {
122
- "f1-score": 0.9007852612503775,
123
- "precision": 0.9097148086014946,
124
- "recall": 0.8920293106026619,
125
- "support": 13374.0
126
  },
127
- "eval_accuracy": 0.9070002759635728,
128
- "eval_loss": 0.23778581619262695,
129
  "eval_macro avg": {
130
- "f1-score": 0.8666785180094754,
131
- "precision": 0.8583043871076081,
132
- "recall": 0.876072714645248,
133
- "support": 32613.0
134
- },
135
- "eval_runtime": 4.9168,
136
- "eval_samples_per_second": 16.474,
137
- "eval_steps_per_second": 2.237,
138
  "eval_weighted avg": {
139
- "f1-score": 0.907089072967282,
140
- "precision": 0.9075314026721463,
141
- "recall": 0.9070002759635728,
142
- "support": 32613.0
143
  },
144
  "step": 243
145
  },
146
  {
147
  "epoch": 4.0,
148
  "eval_Claim": {
149
- "f1-score": 0.6604639922667954,
150
- "precision": 0.696838347781744,
151
- "recall": 0.6276986678915939,
152
- "support": 4354.0
153
  },
154
  "eval_MajorClaim": {
155
- "f1-score": 0.9140214420853479,
156
- "precision": 0.9028239202657807,
157
- "recall": 0.9255002128565347,
158
- "support": 2349.0
159
  },
160
  "eval_O": {
161
- "f1-score": 0.9994417862838917,
162
- "precision": 0.9991230867346939,
163
- "recall": 0.9997606892150607,
164
- "support": 12536.0
165
  },
166
  "eval_Premise": {
167
- "f1-score": 0.9084203149780549,
168
- "precision": 0.8963534463934785,
169
- "recall": 0.9208165096455809,
170
- "support": 13374.0
171
  },
172
- "eval_accuracy": 0.912366234323736,
173
- "eval_loss": 0.2598011791706085,
174
  "eval_macro avg": {
175
- "f1-score": 0.8705868839035225,
176
- "precision": 0.8737847002939243,
177
- "recall": 0.8684440199021926,
178
- "support": 32613.0
179
- },
180
- "eval_runtime": 4.8872,
181
- "eval_samples_per_second": 16.574,
182
- "eval_steps_per_second": 2.251,
183
  "eval_weighted avg": {
184
- "f1-score": 0.9107077581074873,
185
- "precision": 0.9096864919608297,
186
- "recall": 0.912366234323736,
187
- "support": 32613.0
188
  },
189
  "step": 324
190
  },
191
  {
192
  "epoch": 5.0,
193
  "eval_Claim": {
194
- "f1-score": 0.6731001206272618,
195
- "precision": 0.6440713536201469,
196
- "recall": 0.7048690858980248,
197
- "support": 4354.0
198
  },
199
  "eval_MajorClaim": {
200
- "f1-score": 0.8871466886734063,
201
- "precision": 0.8606885508406725,
202
- "recall": 0.9152830991911451,
203
- "support": 2349.0
204
  },
205
  "eval_O": {
206
- "f1-score": 0.9998005345673595,
207
- "precision": 1.0,
208
- "recall": 0.9996011486917677,
209
- "support": 12536.0
210
  },
211
  "eval_Premise": {
212
- "f1-score": 0.9039056236399038,
213
- "precision": 0.9234729698104376,
214
- "recall": 0.8851502916105877,
215
- "support": 13374.0
216
  },
217
- "eval_accuracy": 0.9072455769171802,
218
- "eval_loss": 0.3166182339191437,
219
  "eval_macro avg": {
220
- "f1-score": 0.8659882418769828,
221
- "precision": 0.8570582185678143,
222
- "recall": 0.8762259063478812,
223
- "support": 32613.0
224
- },
225
- "eval_runtime": 4.9018,
226
- "eval_samples_per_second": 16.525,
227
- "eval_steps_per_second": 2.244,
228
  "eval_weighted avg": {
229
- "f1-score": 0.9087455557232214,
230
- "precision": 0.9110652677714914,
231
- "recall": 0.9072455769171802,
232
- "support": 32613.0
233
  },
234
  "step": 405
235
  },
236
  {
237
  "epoch": 6.0,
238
  "eval_Claim": {
239
- "f1-score": 0.634920634920635,
240
- "precision": 0.7171775592828225,
241
- "recall": 0.5695911805236564,
242
- "support": 4354.0
243
  },
244
  "eval_MajorClaim": {
245
- "f1-score": 0.9061357171881348,
246
- "precision": 0.8666925767586475,
247
- "recall": 0.9493401447424435,
248
- "support": 2349.0
249
  },
250
  "eval_O": {
251
- "f1-score": 0.9998404722022812,
252
- "precision": 0.9997607273887382,
253
- "recall": 0.9999202297383536,
254
- "support": 12536.0
255
  },
256
  "eval_Premise": {
257
- "f1-score": 0.9096214165876431,
258
- "precision": 0.8879236684705212,
259
- "recall": 0.9324061612083147,
260
- "support": 13374.0
261
  },
262
- "eval_accuracy": 0.9111397295556987,
263
- "eval_loss": 0.32181796431541443,
264
  "eval_macro avg": {
265
- "f1-score": 0.8626295602246735,
266
- "precision": 0.8678886329751823,
267
- "recall": 0.8628144290531921,
268
- "support": 32613.0
269
- },
270
- "eval_runtime": 4.8856,
271
- "eval_samples_per_second": 16.579,
272
- "eval_steps_per_second": 2.252,
273
  "eval_weighted avg": {
274
- "f1-score": 0.907375409471386,
275
- "precision": 0.9065876667768511,
276
- "recall": 0.9111397295556987,
277
- "support": 32613.0
278
  },
279
  "step": 486
280
  },
281
  {
282
  "epoch": 6.17,
283
- "grad_norm": 7.612313747406006,
284
  "learning_rate": 1.7530864197530865e-05,
285
- "loss": 0.2088,
286
  "step": 500
287
  },
288
  {
289
  "epoch": 7.0,
290
  "eval_Claim": {
291
- "f1-score": 0.6538414409998775,
292
- "precision": 0.7008142894667717,
293
- "recall": 0.6127698667891593,
294
- "support": 4354.0
295
  },
296
  "eval_MajorClaim": {
297
- "f1-score": 0.8866711022399646,
298
- "precision": 0.9254629629629629,
299
- "recall": 0.8510004257130694,
300
- "support": 2349.0
301
  },
302
  "eval_O": {
303
- "f1-score": 0.999880349379811,
304
- "precision": 0.9998404722022812,
305
- "recall": 0.9999202297383536,
306
- "support": 12536.0
307
  },
308
  "eval_Premise": {
309
- "f1-score": 0.9127824473310774,
310
- "precision": 0.8890070167977886,
311
- "recall": 0.9378645132346344,
312
- "support": 13374.0
313
  },
314
- "eval_accuracy": 0.9120596081317266,
315
- "eval_loss": 0.4053190350532532,
316
  "eval_macro avg": {
317
- "f1-score": 0.8632938349876826,
318
- "precision": 0.8787811853574512,
319
- "recall": 0.8503887588688042,
320
- "support": 32613.0
321
- },
322
- "eval_runtime": 4.9061,
323
- "eval_samples_per_second": 16.51,
324
- "eval_steps_per_second": 2.242,
325
  "eval_weighted avg": {
326
- "f1-score": 0.909811074225195,
327
- "precision": 0.9091110268457285,
328
- "recall": 0.9120596081317266,
329
- "support": 32613.0
330
  },
331
  "step": 567
332
  },
333
  {
334
  "epoch": 8.0,
335
  "eval_Claim": {
336
- "f1-score": 0.668355545764952,
337
- "precision": 0.6421164021164021,
338
- "recall": 0.6968305006890216,
339
- "support": 4354.0
340
  },
341
  "eval_MajorClaim": {
342
- "f1-score": 0.8960263537162857,
343
- "precision": 0.8676236044657097,
344
- "recall": 0.9263516389953171,
345
- "support": 2349.0
346
  },
347
  "eval_O": {
348
- "f1-score": 1.0,
349
- "precision": 1.0,
350
- "recall": 1.0,
351
- "support": 12536.0
352
  },
353
  "eval_Premise": {
354
- "f1-score": 0.9012129071630178,
355
- "precision": 0.9198069137340392,
356
- "recall": 0.8833557649170032,
357
- "support": 13374.0
358
  },
359
- "eval_accuracy": 0.9063870235795541,
360
- "eval_loss": 0.4514281749725342,
361
  "eval_macro avg": {
362
- "f1-score": 0.8663987016610639,
363
- "precision": 0.8573867300790377,
364
- "recall": 0.8766344761503355,
365
- "support": 32613.0
366
- },
367
- "eval_runtime": 4.8816,
368
- "eval_samples_per_second": 16.593,
369
- "eval_steps_per_second": 2.253,
370
  "eval_weighted avg": {
371
- "f1-score": 0.9077241398073883,
372
- "precision": 0.9098003963445191,
373
- "recall": 0.9063870235795541,
374
- "support": 32613.0
375
  },
376
  "step": 648
377
  },
378
  {
379
  "epoch": 9.0,
380
  "eval_Claim": {
381
- "f1-score": 0.6913841807909605,
382
- "precision": 0.7090777402221149,
383
- "recall": 0.674552135966927,
384
- "support": 4354.0
385
  },
386
  "eval_MajorClaim": {
387
- "f1-score": 0.9055304497525284,
388
- "precision": 0.9155787641427328,
389
- "recall": 0.8957002979991485,
390
- "support": 2349.0
391
  },
392
  "eval_O": {
393
- "f1-score": 0.9998404340194671,
394
- "precision": 1.0,
395
- "recall": 0.9996809189534142,
396
- "support": 12536.0
397
  },
398
  "eval_Premise": {
399
- "f1-score": 0.9162317231167869,
400
- "precision": 0.9072648632798181,
401
- "recall": 0.9253775983251085,
402
- "support": 13374.0
403
  },
404
- "eval_accuracy": 0.9183147824487168,
405
- "eval_loss": 0.4471658766269684,
406
  "eval_macro avg": {
407
- "f1-score": 0.8782466969199356,
408
- "precision": 0.8829803419111664,
409
- "recall": 0.8738277378111496,
410
- "support": 32613.0
411
- },
412
- "eval_runtime": 4.8813,
413
- "eval_samples_per_second": 16.594,
414
- "eval_steps_per_second": 2.253,
415
  "eval_weighted avg": {
416
- "f1-score": 0.9175807345372852,
417
- "precision": 0.9170508471898524,
418
- "recall": 0.9183147824487168,
419
- "support": 32613.0
420
  },
421
  "step": 729
422
  },
423
  {
424
  "epoch": 10.0,
425
  "eval_Claim": {
426
- "f1-score": 0.6829727187206021,
427
- "precision": 0.6997590361445784,
428
- "recall": 0.6669728984841525,
429
- "support": 4354.0
430
  },
431
  "eval_MajorClaim": {
432
- "f1-score": 0.908567906307787,
433
- "precision": 0.8780778395552026,
434
- "recall": 0.9412515964240102,
435
- "support": 2349.0
436
  },
437
  "eval_O": {
438
- "f1-score": 0.9999202233745512,
439
- "precision": 1.0,
440
- "recall": 0.9998404594767071,
441
- "support": 12536.0
442
  },
443
  "eval_Premise": {
444
- "f1-score": 0.9131976852716073,
445
- "precision": 0.9119379613749907,
446
- "recall": 0.914460894272469,
447
- "support": 13374.0
448
  },
449
- "eval_accuracy": 0.9161683991046515,
450
- "eval_loss": 0.45130759477615356,
451
  "eval_macro avg": {
452
- "f1-score": 0.8761646334186369,
453
- "precision": 0.8724437092686929,
454
- "recall": 0.8806314621643347,
455
- "support": 32613.0
456
- },
457
- "eval_runtime": 4.8986,
458
- "eval_samples_per_second": 16.535,
459
- "eval_steps_per_second": 2.246,
460
  "eval_weighted avg": {
461
- "f1-score": 0.9154630053129839,
462
- "precision": 0.9150220459300829,
463
- "recall": 0.9161683991046515,
464
- "support": 32613.0
465
  },
466
  "step": 810
467
  },
468
  {
469
  "epoch": 11.0,
470
  "eval_Claim": {
471
- "f1-score": 0.6209771986970684,
472
- "precision": 0.717554953327311,
473
- "recall": 0.5473128158015618,
474
- "support": 4354.0
475
  },
476
  "eval_MajorClaim": {
477
- "f1-score": 0.8710538016896399,
478
- "precision": 0.9115867845509539,
479
- "recall": 0.8339719029374202,
480
- "support": 2349.0
481
  },
482
  "eval_O": {
483
- "f1-score": 1.0,
484
- "precision": 1.0,
485
  "recall": 1.0,
486
- "support": 12536.0
487
  },
488
  "eval_Premise": {
489
- "f1-score": 0.9144776812837283,
490
- "precision": 0.8758814267132197,
491
- "recall": 0.9566322715717063,
492
- "support": 13374.0
493
  },
494
- "eval_accuracy": 0.9098212369300586,
495
- "eval_loss": 0.5217961668968201,
496
  "eval_macro avg": {
497
- "f1-score": 0.8516271704176092,
498
- "precision": 0.8762557911478711,
499
- "recall": 0.8344792475776721,
500
- "support": 32613.0
501
- },
502
- "eval_runtime": 4.8915,
503
- "eval_samples_per_second": 16.559,
504
- "eval_steps_per_second": 2.249,
505
  "eval_weighted avg": {
506
- "f1-score": 0.9050398495319223,
507
- "precision": 0.9050252912813879,
508
- "recall": 0.9098212369300586,
509
- "support": 32613.0
510
  },
511
  "step": 891
512
  }
@@ -516,7 +516,7 @@
516
  "num_input_tokens_seen": 0,
517
  "num_train_epochs": 50,
518
  "save_steps": 500,
519
- "total_flos": 1576764022541400.0,
520
  "train_batch_size": 4,
521
  "trial_name": null,
522
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_Claim": {
14
+ "f1-score": 0.5724454649827784,
15
+ "precision": 0.6002889477486154,
16
+ "recall": 0.5470704410796576,
17
+ "support": 4557.0
18
  },
19
  "eval_MajorClaim": {
20
+ "f1-score": 0.7860447185813415,
21
+ "precision": 0.698526892771497,
22
+ "recall": 0.8986337593653592,
23
+ "support": 2269.0
24
  },
25
  "eval_O": {
26
+ "f1-score": 0.9987820830889982,
27
+ "precision": 0.9975671292124707,
28
  "recall": 1.0,
29
+ "support": 11071.0
30
  },
31
  "eval_Premise": {
32
+ "f1-score": 0.8965445389824622,
33
+ "precision": 0.9051258677512096,
34
+ "recall": 0.8881243979633962,
35
+ "support": 14534.0
36
  },
37
+ "eval_accuracy": 0.8791279948197712,
38
+ "eval_loss": 0.3157936632633209,
39
  "eval_macro avg": {
40
+ "f1-score": 0.8134542014088951,
41
+ "precision": 0.8003772093709481,
42
+ "recall": 0.8334571496021033,
43
+ "support": 32431.0
44
+ },
45
+ "eval_runtime": 4.8184,
46
+ "eval_samples_per_second": 16.603,
47
+ "eval_steps_per_second": 2.075,
48
  "eval_weighted avg": {
49
+ "f1-score": 0.8781740995293698,
50
+ "precision": 0.8793943851252293,
51
+ "recall": 0.8791279948197712,
52
+ "support": 32431.0
53
  },
54
  "step": 81
55
  },
56
  {
57
  "epoch": 2.0,
58
  "eval_Claim": {
59
+ "f1-score": 0.6090644692195831,
60
+ "precision": 0.680108254397835,
61
+ "recall": 0.5514592933947773,
62
+ "support": 4557.0
63
  },
64
  "eval_MajorClaim": {
65
+ "f1-score": 0.8585858585858586,
66
+ "precision": 0.8960229995208433,
67
+ "recall": 0.8241516086381666,
68
+ "support": 2269.0
69
  },
70
  "eval_O": {
71
+ "f1-score": 0.999458141425088,
72
+ "precision": 0.9992776523702032,
73
+ "recall": 0.9996386956914461,
74
+ "support": 11071.0
75
  },
76
  "eval_Premise": {
77
+ "f1-score": 0.9105885478942475,
78
+ "precision": 0.8801849235905997,
79
+ "recall": 0.9431677445988716,
80
+ "support": 14534.0
81
  },
82
+ "eval_accuracy": 0.8990780426135487,
83
+ "eval_loss": 0.24451124668121338,
84
  "eval_macro avg": {
85
+ "f1-score": 0.8444242542811943,
86
+ "precision": 0.8638984574698704,
87
+ "recall": 0.8296043355808155,
88
+ "support": 32431.0
89
+ },
90
+ "eval_runtime": 4.8529,
91
+ "eval_samples_per_second": 16.485,
92
+ "eval_steps_per_second": 2.061,
93
  "eval_weighted avg": {
94
+ "f1-score": 0.8949194640367888,
95
+ "precision": 0.8938342965082798,
96
+ "recall": 0.8990780426135487,
97
+ "support": 32431.0
98
  },
99
  "step": 162
100
  },
101
  {
102
  "epoch": 3.0,
103
  "eval_Claim": {
104
+ "f1-score": 0.6542320859402667,
105
+ "precision": 0.7079713847726111,
106
+ "recall": 0.60807548825982,
107
+ "support": 4557.0
108
  },
109
  "eval_MajorClaim": {
110
+ "f1-score": 0.8915187376725839,
111
+ "precision": 0.8866608544027899,
112
+ "recall": 0.8964301454385192,
113
+ "support": 2269.0
114
  },
115
  "eval_O": {
116
+ "f1-score": 0.9995936611133684,
117
+ "precision": 0.9992778479870013,
118
+ "recall": 0.9999096739228616,
119
+ "support": 11071.0
120
  },
121
  "eval_Premise": {
122
+ "f1-score": 0.9157990498332153,
123
+ "precision": 0.8973258501155497,
124
+ "recall": 0.935048850970139,
125
+ "support": 14534.0
126
  },
127
+ "eval_accuracy": 0.9085442940396534,
128
+ "eval_loss": 0.22665414214134216,
129
  "eval_macro avg": {
130
+ "f1-score": 0.8652858836398586,
131
+ "precision": 0.872808984319488,
132
+ "recall": 0.8598660396478349,
133
+ "support": 32431.0
134
+ },
135
+ "eval_runtime": 4.8494,
136
+ "eval_samples_per_second": 16.497,
137
+ "eval_steps_per_second": 2.062,
138
  "eval_weighted avg": {
139
+ "f1-score": 0.9059516032151627,
140
+ "precision": 0.9047762338408377,
141
+ "recall": 0.9085442940396534,
142
+ "support": 32431.0
143
  },
144
  "step": 243
145
  },
146
  {
147
  "epoch": 4.0,
148
  "eval_Claim": {
149
+ "f1-score": 0.6861642294713161,
150
+ "precision": 0.6424741478360781,
151
+ "recall": 0.7362299758613122,
152
+ "support": 4557.0
153
  },
154
  "eval_MajorClaim": {
155
+ "f1-score": 0.8995515695067264,
156
+ "precision": 0.9155636695572797,
157
+ "recall": 0.884089907448215,
158
+ "support": 2269.0
159
  },
160
  "eval_O": {
161
+ "f1-score": 0.9976481230212574,
162
+ "precision": 0.9990941208442794,
163
+ "recall": 0.9962063047601842,
164
+ "support": 11071.0
165
  },
166
  "eval_Premise": {
167
+ "f1-score": 0.9062532879739067,
168
+ "precision": 0.9242435081193219,
169
+ "recall": 0.8889500481629283,
170
+ "support": 14534.0
171
  },
172
+ "eval_accuracy": 0.9037649162838025,
173
+ "eval_loss": 0.2452327460050583,
174
  "eval_macro avg": {
175
+ "f1-score": 0.8724043024933017,
176
+ "precision": 0.8703438615892398,
177
+ "recall": 0.8763690590581599,
178
+ "support": 32431.0
179
+ },
180
+ "eval_runtime": 4.8393,
181
+ "eval_samples_per_second": 16.531,
182
+ "eval_steps_per_second": 2.066,
183
  "eval_weighted avg": {
184
+ "f1-score": 0.9060584182508294,
185
+ "precision": 0.9095955973231696,
186
+ "recall": 0.9037649162838025,
187
+ "support": 32431.0
188
  },
189
  "step": 324
190
  },
191
  {
192
  "epoch": 5.0,
193
  "eval_Claim": {
194
+ "f1-score": 0.6640249759846301,
195
+ "precision": 0.7332272606735614,
196
+ "recall": 0.6067588325652842,
197
+ "support": 4557.0
198
  },
199
  "eval_MajorClaim": {
200
+ "f1-score": 0.8848007008322383,
201
+ "precision": 0.8794079233783195,
202
+ "recall": 0.8902600264433671,
203
+ "support": 2269.0
204
  },
205
  "eval_O": {
206
+ "f1-score": 0.999367774566474,
207
+ "precision": 0.9992775219001174,
208
+ "recall": 0.9994580435371692,
209
+ "support": 11071.0
210
  },
211
  "eval_Premise": {
212
+ "f1-score": 0.9201314377682404,
213
+ "precision": 0.8973839110529758,
214
+ "recall": 0.9440621989816981,
215
+ "support": 14534.0
216
  },
217
+ "eval_accuracy": 0.9118127717307515,
218
+ "eval_loss": 0.28411003947257996,
219
  "eval_macro avg": {
220
+ "f1-score": 0.8670812222878956,
221
+ "precision": 0.8773241542512435,
222
+ "recall": 0.8601347753818797,
223
+ "support": 32431.0
224
+ },
225
+ "eval_runtime": 4.8228,
226
+ "eval_samples_per_second": 16.588,
227
+ "eval_steps_per_second": 2.073,
228
  "eval_weighted avg": {
229
+ "f1-score": 0.9087220731552942,
230
+ "precision": 0.9078434958291443,
231
+ "recall": 0.9118127717307515,
232
+ "support": 32431.0
233
  },
234
  "step": 405
235
  },
236
  {
237
  "epoch": 6.0,
238
  "eval_Claim": {
239
+ "f1-score": 0.6496087442553721,
240
+ "precision": 0.7484258729250143,
241
+ "recall": 0.5738424402018872,
242
+ "support": 4557.0
243
  },
244
  "eval_MajorClaim": {
245
+ "f1-score": 0.8865800865800866,
246
+ "precision": 0.8711186729051468,
247
+ "recall": 0.9026002644336713,
248
+ "support": 2269.0
249
  },
250
  "eval_O": {
251
+ "f1-score": 0.9986892655367232,
252
+ "precision": 0.9994572100597069,
253
+ "recall": 0.9979225002258152,
254
+ "support": 11071.0
255
  },
256
  "eval_Premise": {
257
+ "f1-score": 0.9191778088205947,
258
+ "precision": 0.889647180015452,
259
+ "recall": 0.9507362047612495,
260
+ "support": 14534.0
261
  },
262
+ "eval_accuracy": 0.9105177145323918,
263
+ "eval_loss": 0.3082066774368286,
264
  "eval_macro avg": {
265
+ "f1-score": 0.8635139762981943,
266
+ "precision": 0.87716223397633,
267
+ "recall": 0.8562753524056559,
268
+ "support": 32431.0
269
+ },
270
+ "eval_runtime": 4.8389,
271
+ "eval_samples_per_second": 16.533,
272
+ "eval_steps_per_second": 2.067,
273
  "eval_weighted avg": {
274
+ "f1-score": 0.9061618943658084,
275
+ "precision": 0.9059932736781678,
276
+ "recall": 0.9105177145323918,
277
+ "support": 32431.0
278
  },
279
  "step": 486
280
  },
281
  {
282
  "epoch": 6.17,
283
+ "grad_norm": 0.8612700700759888,
284
  "learning_rate": 1.7530864197530865e-05,
285
+ "loss": 0.2223,
286
  "step": 500
287
  },
288
  {
289
  "epoch": 7.0,
290
  "eval_Claim": {
291
+ "f1-score": 0.6977381768334475,
292
+ "precision": 0.629950495049505,
293
+ "recall": 0.7818740399385561,
294
+ "support": 4557.0
295
  },
296
  "eval_MajorClaim": {
297
+ "f1-score": 0.8892209178228387,
298
+ "precision": 0.8621688741721855,
299
+ "recall": 0.9180255619215514,
300
+ "support": 2269.0
301
  },
302
  "eval_O": {
303
+ "f1-score": 0.9988698521766648,
304
+ "precision": 0.9998190045248869,
305
+ "recall": 0.9979225002258152,
306
+ "support": 11071.0
307
  },
308
  "eval_Premise": {
309
+ "f1-score": 0.9034227633516503,
310
+ "precision": 0.9449996243143738,
311
+ "recall": 0.8653502132929682,
312
+ "support": 14534.0
313
  },
314
+ "eval_accuracy": 0.9025623631710401,
315
+ "eval_loss": 0.3512111008167267,
316
  "eval_macro avg": {
317
+ "f1-score": 0.8723129275461503,
318
+ "precision": 0.8592344995152377,
319
+ "recall": 0.8907930788447227,
320
+ "support": 32431.0
321
+ },
322
+ "eval_runtime": 4.8174,
323
+ "eval_samples_per_second": 16.607,
324
+ "eval_steps_per_second": 2.076,
325
  "eval_weighted avg": {
326
+ "f1-score": 0.9061105026169647,
327
+ "precision": 0.9136494810618672,
328
+ "recall": 0.9025623631710401,
329
+ "support": 32431.0
330
  },
331
  "step": 567
332
  },
333
  {
334
  "epoch": 8.0,
335
  "eval_Claim": {
336
+ "f1-score": 0.6987032799389779,
337
+ "precision": 0.6177710335525206,
338
+ "recall": 0.80403774412991,
339
+ "support": 4557.0
340
  },
341
  "eval_MajorClaim": {
342
+ "f1-score": 0.9015364639688379,
343
+ "precision": 0.8856292517006803,
344
+ "recall": 0.9180255619215514,
345
+ "support": 2269.0
346
  },
347
  "eval_O": {
348
+ "f1-score": 0.998915499322187,
349
+ "precision": 0.9994574554661362,
350
+ "recall": 0.9983741306115076,
351
+ "support": 11071.0
352
  },
353
  "eval_Premise": {
354
+ "f1-score": 0.8976577489773016,
355
+ "precision": 0.9472075788830315,
356
+ "recall": 0.8530342644832806,
357
+ "support": 14534.0
358
  },
359
+ "eval_accuracy": 0.9003114304215103,
360
+ "eval_loss": 0.3705728054046631,
361
  "eval_macro avg": {
362
+ "f1-score": 0.8742032480518261,
363
+ "precision": 0.8625163299005921,
364
+ "recall": 0.8933679252865624,
365
+ "support": 32431.0
366
+ },
367
+ "eval_runtime": 4.8542,
368
+ "eval_samples_per_second": 16.48,
369
+ "eval_steps_per_second": 2.06,
370
  "eval_weighted avg": {
371
+ "f1-score": 0.9045397397569994,
372
+ "precision": 0.9144455555782818,
373
+ "recall": 0.9003114304215103,
374
+ "support": 32431.0
375
  },
376
  "step": 648
377
  },
378
  {
379
  "epoch": 9.0,
380
  "eval_Claim": {
381
+ "f1-score": 0.6657139454567107,
382
+ "precision": 0.7278645833333334,
383
+ "recall": 0.6133421110379635,
384
+ "support": 4557.0
385
  },
386
  "eval_MajorClaim": {
387
+ "f1-score": 0.8882521489971346,
388
+ "precision": 0.8884479717813051,
389
+ "recall": 0.8880564125165271,
390
+ "support": 2269.0
391
  },
392
  "eval_O": {
393
+ "f1-score": 0.9987339482727438,
394
+ "precision": 0.9999094612947035,
395
+ "recall": 0.9975611959172613,
396
+ "support": 11071.0
397
  },
398
  "eval_Premise": {
399
+ "f1-score": 0.9178854152690192,
400
+ "precision": 0.8955360649299646,
401
+ "recall": 0.9413788358332187,
402
+ "support": 14534.0
403
  },
404
+ "eval_accuracy": 0.9107335573987851,
405
+ "eval_loss": 0.41826051473617554,
406
  "eval_macro avg": {
407
+ "f1-score": 0.8676463644989021,
408
+ "precision": 0.8779395203348266,
409
+ "recall": 0.8600846388262426,
410
+ "support": 32431.0
411
+ },
412
+ "eval_runtime": 4.809,
413
+ "eval_samples_per_second": 16.635,
414
+ "eval_steps_per_second": 2.079,
415
  "eval_weighted avg": {
416
+ "f1-score": 0.9079779452489347,
417
+ "precision": 0.9071100542045434,
418
+ "recall": 0.9107335573987851,
419
+ "support": 32431.0
420
  },
421
  "step": 729
422
  },
423
  {
424
  "epoch": 10.0,
425
  "eval_Claim": {
426
+ "f1-score": 0.6738416222952305,
427
+ "precision": 0.6964871194379391,
428
+ "recall": 0.652622339258284,
429
+ "support": 4557.0
430
  },
431
  "eval_MajorClaim": {
432
+ "f1-score": 0.8951019550136641,
433
+ "precision": 0.8557073954983923,
434
+ "recall": 0.9382988100484795,
435
+ "support": 2269.0
436
  },
437
  "eval_O": {
438
+ "f1-score": 0.9989605459393501,
439
+ "precision": 0.9996382054992764,
440
+ "recall": 0.998283804534369,
441
+ "support": 11071.0
442
  },
443
  "eval_Premise": {
444
+ "f1-score": 0.9109121470961545,
445
+ "precision": 0.9083259218717931,
446
+ "recall": 0.9135131415990092,
447
+ "support": 14534.0
448
  },
449
+ "eval_accuracy": 0.9075267490980852,
450
+ "eval_loss": 0.5459077954292297,
451
  "eval_macro avg": {
452
+ "f1-score": 0.8697040675860997,
453
+ "precision": 0.8650396605768502,
454
+ "recall": 0.8756795238600354,
455
+ "support": 32431.0
456
+ },
457
+ "eval_runtime": 4.8478,
458
+ "eval_samples_per_second": 16.502,
459
+ "eval_steps_per_second": 2.063,
460
  "eval_weighted avg": {
461
+ "f1-score": 0.9065515080853328,
462
+ "precision": 0.9060496255197702,
463
+ "recall": 0.9075267490980852,
464
+ "support": 32431.0
465
  },
466
  "step": 810
467
  },
468
  {
469
  "epoch": 11.0,
470
  "eval_Claim": {
471
+ "f1-score": 0.6581417175035196,
472
+ "precision": 0.7070834383665239,
473
+ "recall": 0.6155365371955234,
474
+ "support": 4557.0
475
  },
476
  "eval_MajorClaim": {
477
+ "f1-score": 0.881979417560762,
478
+ "precision": 0.876414273281114,
479
+ "recall": 0.8876156897311591,
480
+ "support": 2269.0
481
  },
482
  "eval_O": {
483
+ "f1-score": 0.9999548390010388,
484
+ "precision": 0.9999096820809249,
485
  "recall": 1.0,
486
+ "support": 11071.0
487
  },
488
  "eval_Premise": {
489
+ "f1-score": 0.9173079519373565,
490
+ "precision": 0.9002915065588976,
491
+ "recall": 0.9349800467868447,
492
+ "support": 14534.0
493
  },
494
+ "eval_accuracy": 0.9089759797724399,
495
+ "eval_loss": 0.5726844668388367,
496
  "eval_macro avg": {
497
+ "f1-score": 0.8643459815006692,
498
+ "precision": 0.8709247250718651,
499
+ "recall": 0.8595330684283817,
500
+ "support": 32431.0
501
+ },
502
+ "eval_runtime": 4.8447,
503
+ "eval_samples_per_second": 16.513,
504
+ "eval_steps_per_second": 2.064,
505
  "eval_weighted avg": {
506
+ "f1-score": 0.9066330640790278,
507
+ "precision": 0.9054793272287636,
508
+ "recall": 0.9089759797724399,
509
+ "support": 32431.0
510
  },
511
  "step": 891
512
  }
 
516
  "num_input_tokens_seen": 0,
517
  "num_train_epochs": 50,
518
  "save_steps": 500,
519
+ "total_flos": 1581676059994800.0,
520
  "train_batch_size": 4,
521
  "trial_name": null,
522
  "trial_params": null
checkpoint-891/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3f40e108617f6f8082cc0f1a47e89121d861cfb2e77ef6b602a909d9b6ba8d15
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2dd7b387b7349843b62a9d5b4facde1d14a6a35003d55f9e21a477105ecd1d3
3
  size 4856