Theoreticallyhugo commited on
Commit
548ae3a
1 Parent(s): 698c427

Training in progress, epoch 7, checkpoint

Browse files
checkpoint-567/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:74d9489dd63b713bd314f34ebe6b1960cc9739c40fdfadc0f2b8b8746a2bd3ff
3
  size 592330980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffcc31247bc4bd734dae08f5bfba6fd09758e2054ebe57ec5741c0447ddfcfdf
3
  size 592330980
checkpoint-567/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5addb116396c47167d7cb4b818727dceaeaab9d437b7a40f0c76f3c29913d742
3
  size 1014670074
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c09385272d7c58e4c8f7b1635705ebd9c0df2834ce5cb8acd60547b18494cbe0
3
  size 1014670074
checkpoint-567/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4d340499ea5f9e21fcf9c408e8be5415c449998f2ce5ff402242628e91c815b5
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff3bd6bda568947eb5d7a73d9adae4043762d42004c1ccb619b636603f95daa0
3
  size 1064
checkpoint-567/trainer_state.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
- "best_metric": null,
3
- "best_model_checkpoint": null,
4
  "epoch": 7.0,
5
  "eval_steps": 500,
6
  "global_step": 567,
@@ -11,456 +11,456 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
- "f1-score": 0.37160751565762,
15
- "precision": 0.42788461538461536,
16
- "recall": 0.3284132841328413,
17
- "support": 271.0
18
  },
19
  "eval_B-MajorClaim": {
20
- "f1-score": 0.3578947368421052,
21
- "precision": 0.6666666666666666,
22
- "recall": 0.2446043165467626,
23
- "support": 139.0
24
  },
25
  "eval_B-Premise": {
26
- "f1-score": 0.8640915593705293,
27
- "precision": 0.7895424836601307,
28
- "recall": 0.9541864139020537,
29
- "support": 633.0
30
  },
31
  "eval_I-Claim": {
32
- "f1-score": 0.5003402749421533,
33
- "precision": 0.5493126120741183,
34
- "recall": 0.4593851537115721,
35
- "support": 4001.0
36
  },
37
  "eval_I-MajorClaim": {
38
- "f1-score": 0.7718093699515347,
39
- "precision": 0.6502211636611093,
40
- "recall": 0.9493293591654247,
41
- "support": 2013.0
42
  },
43
  "eval_I-Premise": {
44
- "f1-score": 0.875016720916752,
45
- "precision": 0.8846812731043188,
46
- "recall": 0.865561044460127,
47
- "support": 11336.0
48
  },
49
  "eval_O": {
50
- "f1-score": 0.9992483530087988,
51
- "precision": 0.9995577178239717,
52
- "recall": 0.998939179632249,
53
- "support": 11312.0
54
  },
55
- "eval_accuracy": 0.8614038040733883,
56
- "eval_loss": 0.31713685393333435,
57
  "eval_macro avg": {
58
- "f1-score": 0.6771440758127848,
59
- "precision": 0.7096952189107044,
60
- "recall": 0.685774107364433,
61
- "support": 29705.0
62
- },
63
- "eval_runtime": 4.8338,
64
- "eval_samples_per_second": 16.55,
65
- "eval_steps_per_second": 2.069,
66
  "eval_weighted avg": {
67
- "f1-score": 0.8576207231627551,
68
- "precision": 0.8601529227027923,
69
- "recall": 0.8614038040733883,
70
- "support": 29705.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
- "f1-score": 0.4708624708624709,
78
- "precision": 0.6392405063291139,
79
- "recall": 0.3726937269372694,
80
- "support": 271.0
81
  },
82
  "eval_B-MajorClaim": {
83
- "f1-score": 0.796875,
84
- "precision": 0.8717948717948718,
85
- "recall": 0.7338129496402878,
86
- "support": 139.0
87
  },
88
  "eval_B-Premise": {
89
- "f1-score": 0.8736616702355461,
90
- "precision": 0.796875,
91
- "recall": 0.966824644549763,
92
- "support": 633.0
93
  },
94
  "eval_I-Claim": {
95
- "f1-score": 0.5100589925881107,
96
- "precision": 0.6459770114942529,
97
- "recall": 0.4213946513371657,
98
- "support": 4001.0
99
  },
100
  "eval_I-MajorClaim": {
101
- "f1-score": 0.8401387776888176,
102
- "precision": 0.9077277970011534,
103
- "recall": 0.7819175360158966,
104
- "support": 2013.0
105
  },
106
  "eval_I-Premise": {
107
- "f1-score": 0.8912891699864469,
108
- "precision": 0.8338584492430646,
109
- "recall": 0.9572159491884262,
110
- "support": 11336.0
111
  },
112
  "eval_O": {
113
- "f1-score": 0.9996904982977407,
114
- "precision": 1.0,
115
- "recall": 0.9993811881188119,
116
- "support": 11312.0
117
  },
118
- "eval_accuracy": 0.8830499915839084,
119
- "eval_loss": 0.2966194748878479,
120
  "eval_macro avg": {
121
- "f1-score": 0.7689395113798762,
122
- "precision": 0.8136390908374939,
123
- "recall": 0.7476058065410885,
124
- "support": 29705.0
125
- },
126
- "eval_runtime": 4.8625,
127
- "eval_samples_per_second": 16.452,
128
- "eval_steps_per_second": 2.057,
129
  "eval_weighted avg": {
130
- "f1-score": 0.8731020208182413,
131
- "precision": 0.874440834821272,
132
- "recall": 0.8830499915839084,
133
- "support": 29705.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
- "f1-score": 0.6085192697768763,
141
- "precision": 0.6756756756756757,
142
- "recall": 0.5535055350553506,
143
- "support": 271.0
144
  },
145
  "eval_B-MajorClaim": {
146
- "f1-score": 0.8571428571428571,
147
- "precision": 0.851063829787234,
148
- "recall": 0.8633093525179856,
149
- "support": 139.0
150
  },
151
  "eval_B-Premise": {
152
- "f1-score": 0.8834729626808834,
153
- "precision": 0.8529411764705882,
154
- "recall": 0.9162717219589257,
155
- "support": 633.0
156
  },
157
  "eval_I-Claim": {
158
- "f1-score": 0.5764474423833614,
159
- "precision": 0.6584269662921348,
160
- "recall": 0.5126218445388653,
161
- "support": 4001.0
162
  },
163
  "eval_I-MajorClaim": {
164
- "f1-score": 0.8581151832460733,
165
- "precision": 0.9070282235749861,
166
- "recall": 0.8142076502732241,
167
- "support": 2013.0
168
  },
169
  "eval_I-Premise": {
170
- "f1-score": 0.8959744247675935,
171
- "precision": 0.8563158317922328,
172
- "recall": 0.939484827099506,
173
- "support": 11336.0
174
  },
175
  "eval_O": {
176
- "f1-score": 0.9996020340481981,
177
- "precision": 1.0,
178
- "recall": 0.9992043847241867,
179
- "support": 11312.0
180
  },
181
- "eval_accuracy": 0.8918700555462044,
182
- "eval_loss": 0.2552729547023773,
183
  "eval_macro avg": {
184
- "f1-score": 0.811324882006549,
185
- "precision": 0.8287788147989789,
186
- "recall": 0.7998007594525776,
187
- "support": 29705.0
188
- },
189
- "eval_runtime": 4.8422,
190
- "eval_samples_per_second": 16.522,
191
- "eval_steps_per_second": 2.065,
192
  "eval_weighted avg": {
193
- "f1-score": 0.8867633844066056,
194
- "precision": 0.886070631898416,
195
- "recall": 0.8918700555462044,
196
- "support": 29705.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
- "f1-score": 0.6722408026755852,
204
- "precision": 0.6146788990825688,
205
- "recall": 0.7416974169741697,
206
- "support": 271.0
207
  },
208
  "eval_B-MajorClaim": {
209
- "f1-score": 0.8664259927797834,
210
- "precision": 0.8695652173913043,
211
- "recall": 0.8633093525179856,
212
- "support": 139.0
213
  },
214
  "eval_B-Premise": {
215
- "f1-score": 0.8687035507844755,
216
- "precision": 0.9100346020761245,
217
- "recall": 0.8309636650868878,
218
- "support": 633.0
219
  },
220
  "eval_I-Claim": {
221
- "f1-score": 0.6483151400094921,
222
- "precision": 0.6171222046532641,
223
- "recall": 0.6828292926768308,
224
- "support": 4001.0
225
  },
226
  "eval_I-MajorClaim": {
227
- "f1-score": 0.8696993060909791,
228
- "precision": 0.9009584664536742,
229
- "recall": 0.8405365126676602,
230
- "support": 2013.0
231
  },
232
  "eval_I-Premise": {
233
- "f1-score": 0.8906159274643798,
234
- "precision": 0.9020175517958925,
235
- "recall": 0.879498941425547,
236
- "support": 11336.0
237
  },
238
  "eval_O": {
239
- "f1-score": 0.999557835160948,
240
- "precision": 0.9999115357395613,
241
- "recall": 0.9992043847241867,
242
- "support": 11312.0
243
  },
244
- "eval_accuracy": 0.8935869382258879,
245
- "eval_loss": 0.29465603828430176,
246
  "eval_macro avg": {
247
- "f1-score": 0.8307940792808061,
248
- "precision": 0.8306126395989127,
249
- "recall": 0.8340056522961811,
250
- "support": 29705.0
251
- },
252
- "eval_runtime": 4.8494,
253
- "eval_samples_per_second": 16.497,
254
- "eval_steps_per_second": 2.062,
255
  "eval_weighted avg": {
256
- "f1-score": 0.8954766464091573,
257
- "precision": 0.8982496227307208,
258
- "recall": 0.8935869382258879,
259
- "support": 29705.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
- "f1-score": 0.6438095238095239,
267
- "precision": 0.6653543307086615,
268
- "recall": 0.6236162361623616,
269
- "support": 271.0
270
  },
271
  "eval_B-MajorClaim": {
272
- "f1-score": 0.8785714285714286,
273
- "precision": 0.8723404255319149,
274
- "recall": 0.8848920863309353,
275
- "support": 139.0
276
  },
277
  "eval_B-Premise": {
278
- "f1-score": 0.8783151326053042,
279
- "precision": 0.8674884437596302,
280
- "recall": 0.8894154818325435,
281
- "support": 633.0
282
  },
283
  "eval_I-Claim": {
284
- "f1-score": 0.6283729628640129,
285
- "precision": 0.6748923959827834,
286
- "recall": 0.5878530367408148,
287
- "support": 4001.0
288
  },
289
  "eval_I-MajorClaim": {
290
- "f1-score": 0.885450461692039,
291
- "precision": 0.8896690070210632,
292
- "recall": 0.8812717337307501,
293
- "support": 2013.0
294
  },
295
  "eval_I-Premise": {
296
- "f1-score": 0.8989315871101154,
297
- "precision": 0.878494442573257,
298
- "recall": 0.9203422724064926,
299
- "support": 11336.0
300
  },
301
  "eval_O": {
302
- "f1-score": 0.9997347245556637,
303
- "precision": 1.0,
304
- "recall": 0.9994695898161244,
305
- "support": 11312.0
306
  },
307
- "eval_accuracy": 0.8995118666891095,
308
- "eval_loss": 0.3176642060279846,
309
  "eval_macro avg": {
310
- "f1-score": 0.830455117315441,
311
- "precision": 0.8354627207967587,
312
- "recall": 0.8266943481457174,
313
- "support": 29705.0
314
- },
315
- "eval_runtime": 4.8682,
316
- "eval_samples_per_second": 16.433,
317
- "eval_steps_per_second": 2.054,
318
  "eval_weighted avg": {
319
- "f1-score": 0.8971010593476484,
320
- "precision": 0.8958911872123141,
321
- "recall": 0.8995118666891095,
322
- "support": 29705.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
- "f1-score": 0.6150712830957231,
330
- "precision": 0.6863636363636364,
331
- "recall": 0.5571955719557196,
332
- "support": 271.0
333
  },
334
  "eval_B-MajorClaim": {
335
- "f1-score": 0.8686131386861314,
336
- "precision": 0.8814814814814815,
337
- "recall": 0.8561151079136691,
338
- "support": 139.0
339
  },
340
  "eval_B-Premise": {
341
- "f1-score": 0.8804841149773072,
342
- "precision": 0.8447024673439768,
343
- "recall": 0.919431279620853,
344
- "support": 633.0
345
  },
346
  "eval_I-Claim": {
347
- "f1-score": 0.5758052970651396,
348
- "precision": 0.6739276139410187,
349
- "recall": 0.5026243439140214,
350
- "support": 4001.0
351
  },
352
  "eval_I-MajorClaim": {
353
- "f1-score": 0.8627552339105711,
354
- "precision": 0.8992456896551724,
355
- "recall": 0.829110779930452,
356
- "support": 2013.0
357
  },
358
  "eval_I-Premise": {
359
- "f1-score": 0.894107779408681,
360
- "precision": 0.8521864257734432,
361
- "recall": 0.9403669724770642,
362
- "support": 11336.0
363
  },
364
  "eval_O": {
365
- "f1-score": 1.0,
366
- "precision": 1.0,
367
- "recall": 1.0,
368
- "support": 11312.0
369
  },
370
- "eval_accuracy": 0.8922403635751557,
371
- "eval_loss": 0.41071853041648865,
372
  "eval_macro avg": {
373
- "f1-score": 0.8138338353062219,
374
- "precision": 0.8339867592226755,
375
- "recall": 0.8006920079731114,
376
- "support": 29705.0
377
- },
378
- "eval_runtime": 4.8571,
379
- "eval_samples_per_second": 16.471,
380
- "eval_steps_per_second": 2.059,
381
  "eval_weighted avg": {
382
- "f1-score": 0.8864802913843919,
383
- "precision": 0.8861194550557427,
384
- "recall": 0.8922403635751557,
385
- "support": 29705.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
- "grad_norm": 5.239528656005859,
392
- "learning_rate": 1.7530864197530865e-05,
393
- "loss": 0.2363,
394
  "step": 500
395
  },
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
- "f1-score": 0.6954954954954955,
400
- "precision": 0.6795774647887324,
401
- "recall": 0.7121771217712177,
402
- "support": 271.0
403
  },
404
  "eval_B-MajorClaim": {
405
- "f1-score": 0.8920863309352518,
406
- "precision": 0.8920863309352518,
407
- "recall": 0.8920863309352518,
408
- "support": 139.0
409
  },
410
  "eval_B-Premise": {
411
- "f1-score": 0.8876494023904382,
412
- "precision": 0.8954983922829582,
413
- "recall": 0.8799368088467614,
414
- "support": 633.0
415
  },
416
  "eval_I-Claim": {
417
- "f1-score": 0.6831545741324923,
418
- "precision": 0.6898572884811417,
419
- "recall": 0.6765808547863035,
420
- "support": 4001.0
421
  },
422
  "eval_I-MajorClaim": {
423
- "f1-score": 0.8984707946853848,
424
- "precision": 0.9068825910931174,
425
- "recall": 0.8902136115250869,
426
- "support": 2013.0
427
  },
428
  "eval_I-Premise": {
429
- "f1-score": 0.9068164859763859,
430
- "precision": 0.9024198479951079,
431
- "recall": 0.9112561750176429,
432
- "support": 11336.0
433
  },
434
  "eval_O": {
435
- "f1-score": 0.9999558011049724,
436
- "precision": 0.9999116061168567,
437
- "recall": 1.0,
438
- "support": 11312.0
439
  },
440
- "eval_accuracy": 0.9094428547382596,
441
- "eval_loss": 0.3520519435405731,
442
  "eval_macro avg": {
443
- "f1-score": 0.8519469835314888,
444
- "precision": 0.8523190745275951,
445
- "recall": 0.8517501289831806,
446
- "support": 29705.0
447
- },
448
- "eval_runtime": 4.8905,
449
- "eval_samples_per_second": 16.358,
450
- "eval_steps_per_second": 2.045,
451
  "eval_weighted avg": {
452
- "f1-score": 0.9091888981291354,
453
- "precision": 0.908989097041669,
454
- "recall": 0.9094428547382596,
455
- "support": 29705.0
456
  },
457
  "step": 567
458
  }
459
  ],
460
  "logging_steps": 500,
461
- "max_steps": 4050,
462
  "num_input_tokens_seen": 0,
463
- "num_train_epochs": 50,
464
  "save_steps": 500,
465
  "total_flos": 1006535689026000.0,
466
  "train_batch_size": 4,
 
1
  {
2
+ "best_metric": 0.276384562253952,
3
+ "best_model_checkpoint": "longformer-sep_tok_full_labels/checkpoint-324",
4
  "epoch": 7.0,
5
  "eval_steps": 500,
6
  "global_step": 567,
 
11
  {
12
  "epoch": 1.0,
13
  "eval_B-Claim": {
14
+ "f1-score": 0.21800947867298578,
15
+ "precision": 0.3333333333333333,
16
+ "recall": 0.1619718309859155,
17
+ "support": 284.0
18
  },
19
  "eval_B-MajorClaim": {
20
+ "f1-score": 0.0,
21
+ "precision": 0.0,
22
+ "recall": 0.0,
23
+ "support": 141.0
24
  },
25
  "eval_B-Premise": {
26
+ "f1-score": 0.8305489260143198,
27
+ "precision": 0.71900826446281,
28
+ "recall": 0.9830508474576272,
29
+ "support": 708.0
30
  },
31
  "eval_I-Claim": {
32
+ "f1-score": 0.5590492696211935,
33
+ "precision": 0.5643589102724319,
34
+ "recall": 0.5538386068187393,
35
+ "support": 4077.0
36
  },
37
  "eval_I-MajorClaim": {
38
+ "f1-score": 0.7399038461538461,
39
+ "precision": 0.7205056179775281,
40
+ "recall": 0.7603754940711462,
41
+ "support": 2024.0
42
  },
43
  "eval_I-Premise": {
44
+ "f1-score": 0.8867229093291834,
45
+ "precision": 0.8883965205974068,
46
+ "recall": 0.8850555918901243,
47
+ "support": 12232.0
48
  },
49
  "eval_O": {
50
+ "f1-score": 0.9976548035383666,
51
+ "precision": 0.9961383616794018,
52
+ "recall": 0.9991758694577221,
53
+ "support": 12134.0
54
  },
55
+ "eval_accuracy": 0.8699050632911393,
56
+ "eval_loss": 0.32671716809272766,
57
  "eval_macro avg": {
58
+ "f1-score": 0.6045556047614136,
59
+ "precision": 0.6031058583318446,
60
+ "recall": 0.6204954629544678,
61
+ "support": 31600.0
62
+ },
63
+ "eval_runtime": 6.234,
64
+ "eval_samples_per_second": 12.833,
65
+ "eval_steps_per_second": 1.604,
66
  "eval_weighted avg": {
67
+ "f1-score": 0.8664142595402331,
68
+ "precision": 0.8644597558999653,
69
+ "recall": 0.8699050632911393,
70
+ "support": 31600.0
71
  },
72
  "step": 81
73
  },
74
  {
75
  "epoch": 2.0,
76
  "eval_B-Claim": {
77
+ "f1-score": 0.5352697095435685,
78
+ "precision": 0.6515151515151515,
79
+ "recall": 0.45422535211267606,
80
+ "support": 284.0
81
  },
82
  "eval_B-MajorClaim": {
83
+ "f1-score": 0.7766990291262136,
84
+ "precision": 0.7142857142857143,
85
+ "recall": 0.851063829787234,
86
+ "support": 141.0
87
  },
88
  "eval_B-Premise": {
89
+ "f1-score": 0.8802721088435375,
90
+ "precision": 0.8490813648293963,
91
+ "recall": 0.9138418079096046,
92
+ "support": 708.0
93
  },
94
  "eval_I-Claim": {
95
+ "f1-score": 0.5747557221255521,
96
+ "precision": 0.6325869180907484,
97
+ "recall": 0.5266127054206524,
98
+ "support": 4077.0
99
  },
100
  "eval_I-MajorClaim": {
101
+ "f1-score": 0.7833260963960051,
102
+ "precision": 0.6986831913245546,
103
+ "recall": 0.8913043478260869,
104
+ "support": 2024.0
105
  },
106
  "eval_I-Premise": {
107
+ "f1-score": 0.8995515695067264,
108
+ "precision": 0.8971377459749553,
109
+ "recall": 0.9019784172661871,
110
+ "support": 12232.0
111
  },
112
  "eval_O": {
113
+ "f1-score": 0.9971231300345225,
114
+ "precision": 0.9945072962780783,
115
+ "recall": 0.9997527608373167,
116
+ "support": 12134.0
117
  },
118
+ "eval_accuracy": 0.8864240506329114,
119
+ "eval_loss": 0.28716403245925903,
120
  "eval_macro avg": {
121
+ "f1-score": 0.7781424807965893,
122
+ "precision": 0.7768281974712284,
123
+ "recall": 0.791254174451394,
124
+ "support": 31600.0
125
+ },
126
+ "eval_runtime": 6.2816,
127
+ "eval_samples_per_second": 12.736,
128
+ "eval_steps_per_second": 1.592,
129
  "eval_weighted avg": {
130
+ "f1-score": 0.8834146129726252,
131
+ "precision": 0.8835831101628018,
132
+ "recall": 0.8864240506329114,
133
+ "support": 31600.0
134
  },
135
  "step": 162
136
  },
137
  {
138
  "epoch": 3.0,
139
  "eval_B-Claim": {
140
+ "f1-score": 0.5580448065173116,
141
+ "precision": 0.6618357487922706,
142
+ "recall": 0.4823943661971831,
143
+ "support": 284.0
144
  },
145
  "eval_B-MajorClaim": {
146
+ "f1-score": 0.8046875,
147
+ "precision": 0.8956521739130435,
148
+ "recall": 0.7304964539007093,
149
+ "support": 141.0
150
  },
151
  "eval_B-Premise": {
152
+ "f1-score": 0.8868421052631579,
153
+ "precision": 0.8300492610837439,
154
+ "recall": 0.9519774011299436,
155
+ "support": 708.0
156
  },
157
  "eval_I-Claim": {
158
+ "f1-score": 0.5488426262920367,
159
+ "precision": 0.6751432664756447,
160
+ "recall": 0.4623497669855286,
161
+ "support": 4077.0
162
  },
163
  "eval_I-MajorClaim": {
164
+ "f1-score": 0.8110936682365254,
165
+ "precision": 0.8620689655172413,
166
+ "recall": 0.7658102766798419,
167
+ "support": 2024.0
168
  },
169
  "eval_I-Premise": {
170
+ "f1-score": 0.9039491427470623,
171
+ "precision": 0.854842235662756,
172
+ "recall": 0.9590418574231524,
173
+ "support": 12232.0
174
  },
175
  "eval_O": {
176
+ "f1-score": 0.9979824597521307,
177
+ "precision": 0.9972023368715544,
178
+ "recall": 0.9987638041865832,
179
+ "support": 12134.0
180
  },
181
+ "eval_accuracy": 0.892373417721519,
182
+ "eval_loss": 0.28271517157554626,
183
  "eval_macro avg": {
184
+ "f1-score": 0.7873489012583177,
185
+ "precision": 0.825256284045179,
186
+ "recall": 0.7644048466432775,
187
+ "support": 31600.0
188
+ },
189
+ "eval_runtime": 6.2071,
190
+ "eval_samples_per_second": 12.888,
191
+ "eval_steps_per_second": 1.611,
192
  "eval_weighted avg": {
193
+ "f1-score": 0.88435885840807,
194
+ "precision": 0.8846770016417851,
195
+ "recall": 0.892373417721519,
196
+ "support": 31600.0
197
  },
198
  "step": 243
199
  },
200
  {
201
  "epoch": 4.0,
202
  "eval_B-Claim": {
203
+ "f1-score": 0.659016393442623,
204
+ "precision": 0.6165644171779141,
205
+ "recall": 0.7077464788732394,
206
+ "support": 284.0
207
  },
208
  "eval_B-MajorClaim": {
209
+ "f1-score": 0.8675496688741722,
210
+ "precision": 0.8136645962732919,
211
+ "recall": 0.9290780141843972,
212
+ "support": 141.0
213
  },
214
  "eval_B-Premise": {
215
+ "f1-score": 0.8698224852071005,
216
+ "precision": 0.9130434782608695,
217
+ "recall": 0.8305084745762712,
218
+ "support": 708.0
219
  },
220
  "eval_I-Claim": {
221
+ "f1-score": 0.6472172351885099,
222
+ "precision": 0.5964839710444674,
223
+ "recall": 0.70738287956831,
224
+ "support": 4077.0
225
  },
226
  "eval_I-MajorClaim": {
227
+ "f1-score": 0.8549332083430982,
228
+ "precision": 0.8131966116807846,
229
+ "recall": 0.9011857707509882,
230
+ "support": 2024.0
231
  },
232
  "eval_I-Premise": {
233
+ "f1-score": 0.8876102800153433,
234
+ "precision": 0.9271658801531475,
235
+ "recall": 0.8512916939175932,
236
+ "support": 12232.0
237
  },
238
  "eval_O": {
239
+ "f1-score": 0.9981065283609122,
240
+ "precision": 0.9970394736842105,
241
+ "recall": 0.9991758694577221,
242
+ "support": 12134.0
243
  },
244
+ "eval_accuracy": 0.8912974683544304,
245
+ "eval_loss": 0.276384562253952,
246
  "eval_macro avg": {
247
+ "f1-score": 0.8263222570616798,
248
+ "precision": 0.8110226326106693,
249
+ "recall": 0.8466241687612174,
250
+ "support": 31600.0
251
+ },
252
+ "eval_runtime": 6.2529,
253
+ "eval_samples_per_second": 12.794,
254
+ "eval_steps_per_second": 1.599,
255
  "eval_weighted avg": {
256
+ "f1-score": 0.8943886873545748,
257
+ "precision": 0.9004180663566287,
258
+ "recall": 0.8912974683544304,
259
+ "support": 31600.0
260
  },
261
  "step": 324
262
  },
263
  {
264
  "epoch": 5.0,
265
  "eval_B-Claim": {
266
+ "f1-score": 0.665474060822898,
267
+ "precision": 0.6763636363636364,
268
+ "recall": 0.6549295774647887,
269
+ "support": 284.0
270
  },
271
  "eval_B-MajorClaim": {
272
+ "f1-score": 0.8896551724137931,
273
+ "precision": 0.8657718120805369,
274
+ "recall": 0.9148936170212766,
275
+ "support": 141.0
276
  },
277
  "eval_B-Premise": {
278
+ "f1-score": 0.8843441466854725,
279
+ "precision": 0.8830985915492958,
280
+ "recall": 0.885593220338983,
281
+ "support": 708.0
282
  },
283
  "eval_I-Claim": {
284
+ "f1-score": 0.6698067632850242,
285
+ "precision": 0.6597668332143707,
286
+ "recall": 0.6801569781702232,
287
+ "support": 4077.0
288
  },
289
  "eval_I-MajorClaim": {
290
+ "f1-score": 0.8809815950920246,
291
+ "precision": 0.8751828376401756,
292
+ "recall": 0.8868577075098815,
293
+ "support": 2024.0
294
  },
295
  "eval_I-Premise": {
296
+ "f1-score": 0.9024631353488756,
297
+ "precision": 0.9094305163539764,
298
+ "recall": 0.8956017004578156,
299
+ "support": 12232.0
300
  },
301
  "eval_O": {
302
+ "f1-score": 0.9980246913580246,
303
+ "precision": 0.9967121486108828,
304
+ "recall": 0.9993406955661777,
305
+ "support": 12134.0
306
  },
307
+ "eval_accuracy": 0.9047784810126582,
308
+ "eval_loss": 0.2893199920654297,
309
  "eval_macro avg": {
310
+ "f1-score": 0.8415356521437304,
311
+ "precision": 0.838046625116125,
312
+ "recall": 0.8453390709327352,
313
+ "support": 31600.0
314
+ },
315
+ "eval_runtime": 6.2466,
316
+ "eval_samples_per_second": 12.807,
317
+ "eval_steps_per_second": 1.601,
318
  "eval_weighted avg": {
319
+ "f1-score": 0.9051715590931133,
320
+ "precision": 0.9056611908459661,
321
+ "recall": 0.9047784810126582,
322
+ "support": 31600.0
323
  },
324
  "step": 405
325
  },
326
  {
327
  "epoch": 6.0,
328
  "eval_B-Claim": {
329
+ "f1-score": 0.6263345195729537,
330
+ "precision": 0.6330935251798561,
331
+ "recall": 0.6197183098591549,
332
+ "support": 284.0
333
  },
334
  "eval_B-MajorClaim": {
335
+ "f1-score": 0.8372093023255814,
336
+ "precision": 0.9230769230769231,
337
+ "recall": 0.7659574468085106,
338
+ "support": 141.0
339
  },
340
  "eval_B-Premise": {
341
+ "f1-score": 0.8784530386740332,
342
+ "precision": 0.8594594594594595,
343
+ "recall": 0.8983050847457628,
344
+ "support": 708.0
345
  },
346
  "eval_I-Claim": {
347
+ "f1-score": 0.6109127995920448,
348
+ "precision": 0.6360499070878683,
349
+ "recall": 0.5876870247731175,
350
+ "support": 4077.0
351
  },
352
  "eval_I-MajorClaim": {
353
+ "f1-score": 0.8409272581934453,
354
+ "precision": 0.9126662810873337,
355
+ "recall": 0.7796442687747036,
356
+ "support": 2024.0
357
  },
358
  "eval_I-Premise": {
359
+ "f1-score": 0.8964938902643559,
360
+ "precision": 0.8762685402029664,
361
+ "recall": 0.9176749509483323,
362
+ "support": 12232.0
363
  },
364
  "eval_O": {
365
+ "f1-score": 0.9987239122380934,
366
+ "precision": 0.9976971790443293,
367
+ "recall": 0.9997527608373167,
368
+ "support": 12134.0
369
  },
370
+ "eval_accuracy": 0.8939873417721519,
371
+ "eval_loss": 0.33532533049583435,
372
  "eval_macro avg": {
373
+ "f1-score": 0.8127221029800725,
374
+ "precision": 0.8340445450198194,
375
+ "recall": 0.7955342638209855,
376
+ "support": 31600.0
377
+ },
378
+ "eval_runtime": 6.2278,
379
+ "eval_samples_per_second": 12.846,
380
+ "eval_steps_per_second": 1.606,
381
  "eval_weighted avg": {
382
+ "f1-score": 0.8922477132246446,
383
+ "precision": 0.891880888702747,
384
+ "recall": 0.8939873417721519,
385
+ "support": 31600.0
386
  },
387
  "step": 486
388
  },
389
  {
390
  "epoch": 6.17,
391
+ "grad_norm": 4.557807445526123,
392
+ "learning_rate": 1.3827160493827162e-05,
393
+ "loss": 0.2608,
394
  "step": 500
395
  },
396
  {
397
  "epoch": 7.0,
398
  "eval_B-Claim": {
399
+ "f1-score": 0.5553235908141962,
400
+ "precision": 0.6820512820512821,
401
+ "recall": 0.46830985915492956,
402
+ "support": 284.0
403
  },
404
  "eval_B-MajorClaim": {
405
+ "f1-score": 0.8412698412698413,
406
+ "precision": 0.954954954954955,
407
+ "recall": 0.75177304964539,
408
+ "support": 141.0
409
  },
410
  "eval_B-Premise": {
411
+ "f1-score": 0.8846905537459283,
412
+ "precision": 0.8210399032648126,
413
+ "recall": 0.9590395480225988,
414
+ "support": 708.0
415
  },
416
  "eval_I-Claim": {
417
+ "f1-score": 0.5484584747012834,
418
+ "precision": 0.6880088823094005,
419
+ "recall": 0.45597252882021094,
420
+ "support": 4077.0
421
  },
422
  "eval_I-MajorClaim": {
423
+ "f1-score": 0.8530857454942655,
424
+ "precision": 0.9536019536019537,
425
+ "recall": 0.7717391304347826,
426
+ "support": 2024.0
427
  },
428
  "eval_I-Premise": {
429
+ "f1-score": 0.900171657448026,
430
+ "precision": 0.8438103411285132,
431
+ "recall": 0.9646010464355788,
432
+ "support": 12232.0
433
  },
434
  "eval_O": {
435
+ "f1-score": 0.9986819342614713,
436
+ "precision": 0.9982707509881423,
437
+ "recall": 0.9990934564034943,
438
+ "support": 12134.0
439
  },
440
+ "eval_accuracy": 0.8943354430379746,
441
+ "eval_loss": 0.45137375593185425,
442
  "eval_macro avg": {
443
+ "f1-score": 0.7973831139621446,
444
+ "precision": 0.8488197240427228,
445
+ "recall": 0.7672183741309979,
446
+ "support": 31600.0
447
+ },
448
+ "eval_runtime": 6.1951,
449
+ "eval_samples_per_second": 12.913,
450
+ "eval_steps_per_second": 1.614,
451
  "eval_weighted avg": {
452
+ "f1-score": 0.8858958517067363,
453
+ "precision": 0.8885840321741321,
454
+ "recall": 0.8943354430379746,
455
+ "support": 31600.0
456
  },
457
  "step": 567
458
  }
459
  ],
460
  "logging_steps": 500,
461
+ "max_steps": 1620,
462
  "num_input_tokens_seen": 0,
463
+ "num_train_epochs": 20,
464
  "save_steps": 500,
465
  "total_flos": 1006535689026000.0,
466
  "train_batch_size": 4,
checkpoint-567/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3aa7d1dcb1ecb96052506e9849f1d3fa19e688784f8ea37f784dfaca9546785
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a6eed06a4e3317966121cd80a8275e0b82aa59ebaa0185c3b4676ff92bfed33
3
  size 4920