Theoreticallyhugo commited on
Commit
94f4372
1 Parent(s): b97d030

Training in progress, epoch 10, checkpoint

Browse files
checkpoint-810/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aa9d3ff67d83e34c49ed1e06d08d76aed3baa9d5cc1e49f4f7482415d187dfca
3
  size 592318676
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0411c659a26be1ad2d23c65920d06dc51eb3ae9dd61c2659d4e4037062b1f2d
3
  size 592318676
checkpoint-810/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bb1073e16f8a72946914f45479a4deed25eb31c5be8b87f5b5f840cfd7a45579
3
  size 1014645498
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1db891b8390791da69022f012d643ff62cc5fd2b0c1ccb270c4222a1275733a0
3
  size 1014645498
checkpoint-810/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:516a7ea2b4f58aa1466b3499947a191d75f003f2e001b6d4bbb7a888f04d977f
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05f741fe22f4ca5655fda1f9ef1ecba0015571f3b8afe49bf335811d1e94a138
3
  size 14244
checkpoint-810/trainer_state.json CHANGED
@@ -11,397 +11,397 @@
11
  {
12
  "epoch": 1.0,
13
  "eval_B": {
14
- "f1-score": 0.8266764382557715,
15
- "precision": 0.7866108786610879,
16
- "recall": 0.871042471042471,
17
- "support": 1295.0
18
  },
19
  "eval_I": {
20
- "f1-score": 0.9410658617447671,
21
- "precision": 0.9344077040239768,
22
- "recall": 0.9478195863443808,
23
- "support": 20065.0
24
  },
25
  "eval_O": {
26
- "f1-score": 0.8562443302086484,
27
- "precision": 0.8789421405512788,
28
- "recall": 0.8346893055064262,
29
- "support": 8481.0
30
  },
31
- "eval_accuracy": 0.912335377500754,
32
- "eval_loss": 0.2450982630252838,
33
  "eval_macro avg": {
34
- "f1-score": 0.8746622100697291,
35
- "precision": 0.8666535744121145,
36
- "recall": 0.8845171209644261,
37
- "support": 29841.0
38
- },
39
- "eval_runtime": 4.6269,
40
- "eval_samples_per_second": 17.29,
41
- "eval_steps_per_second": 2.161,
42
  "eval_weighted avg": {
43
- "f1-score": 0.9119949287205363,
44
- "precision": 0.9122301519092053,
45
- "recall": 0.912335377500754,
46
- "support": 29841.0
47
  },
48
  "step": 81
49
  },
50
  {
51
  "epoch": 2.0,
52
  "eval_B": {
53
- "f1-score": 0.8833395452851286,
54
- "precision": 0.8537463976945245,
55
- "recall": 0.915057915057915,
56
- "support": 1295.0
57
  },
58
  "eval_I": {
59
- "f1-score": 0.9549518227873249,
60
- "precision": 0.9516903430183636,
61
- "recall": 0.958235733864939,
62
- "support": 20065.0
63
  },
64
  "eval_O": {
65
- "f1-score": 0.88924750463212,
66
- "precision": 0.9016969696969697,
67
- "recall": 0.877137130055418,
68
- "support": 8481.0
69
  },
70
- "eval_accuracy": 0.9333132267685399,
71
- "eval_loss": 0.20076116919517517,
72
  "eval_macro avg": {
73
- "f1-score": 0.9091796242348579,
74
- "precision": 0.9023779034699526,
75
- "recall": 0.916810259659424,
76
- "support": 29841.0
77
- },
78
- "eval_runtime": 4.6524,
79
- "eval_samples_per_second": 17.195,
80
- "eval_steps_per_second": 2.149,
81
  "eval_weighted avg": {
82
- "f1-score": 0.9331705077630417,
83
- "precision": 0.9332314707173981,
84
- "recall": 0.9333132267685399,
85
- "support": 29841.0
86
  },
87
  "step": 162
88
  },
89
  {
90
  "epoch": 3.0,
91
  "eval_B": {
92
- "f1-score": 0.8927335640138409,
93
- "precision": 0.8889739663093414,
94
- "recall": 0.8965250965250965,
95
- "support": 1295.0
96
  },
97
  "eval_I": {
98
- "f1-score": 0.9590235396687009,
99
- "precision": 0.9586653386454184,
100
- "recall": 0.9593820084724645,
101
- "support": 20065.0
102
  },
103
  "eval_O": {
104
- "f1-score": 0.9006849315068494,
105
- "precision": 0.9020697811945594,
106
- "recall": 0.8993043273198915,
107
- "support": 8481.0
108
  },
109
- "eval_accuracy": 0.9395797727958178,
110
- "eval_loss": 0.19863653182983398,
111
  "eval_macro avg": {
112
- "f1-score": 0.9174806783964637,
113
- "precision": 0.9165696953831063,
114
- "recall": 0.9184038107724842,
115
- "support": 29841.0
116
- },
117
- "eval_runtime": 4.655,
118
- "eval_samples_per_second": 17.186,
119
- "eval_steps_per_second": 2.148,
120
  "eval_weighted avg": {
121
- "f1-score": 0.9395665759512081,
122
- "precision": 0.9395561516236713,
123
- "recall": 0.9395797727958178,
124
- "support": 29841.0
125
  },
126
  "step": 243
127
  },
128
  {
129
  "epoch": 4.0,
130
  "eval_B": {
131
- "f1-score": 0.8999616711383672,
132
- "precision": 0.893455098934551,
133
- "recall": 0.9065637065637066,
134
- "support": 1295.0
135
  },
136
  "eval_I": {
137
- "f1-score": 0.9612657113512985,
138
- "precision": 0.9436362763454799,
139
- "recall": 0.9795664091701969,
140
- "support": 20065.0
141
  },
142
  "eval_O": {
143
- "f1-score": 0.8996847765622102,
144
- "precision": 0.9454403741231489,
145
- "recall": 0.8581535196321188,
146
- "support": 8481.0
147
  },
148
- "eval_accuracy": 0.9418920277470594,
149
- "eval_loss": 0.19743724167346954,
150
  "eval_macro avg": {
151
- "f1-score": 0.920304053017292,
152
- "precision": 0.9275105831343932,
153
- "recall": 0.9147612117886741,
154
- "support": 29841.0
155
- },
156
- "eval_runtime": 4.6615,
157
- "eval_samples_per_second": 17.162,
158
- "eval_steps_per_second": 2.145,
159
  "eval_weighted avg": {
160
- "f1-score": 0.9411036309913238,
161
- "precision": 0.9419713163409645,
162
- "recall": 0.9418920277470594,
163
- "support": 29841.0
164
  },
165
  "step": 324
166
  },
167
  {
168
  "epoch": 5.0,
169
  "eval_B": {
170
- "f1-score": 0.8953846153846153,
171
- "precision": 0.8919540229885058,
172
- "recall": 0.8988416988416988,
173
- "support": 1295.0
174
  },
175
  "eval_I": {
176
- "f1-score": 0.9602426713742261,
177
- "precision": 0.9581224570804803,
178
- "recall": 0.9623722900573137,
179
- "support": 20065.0
180
  },
181
  "eval_O": {
182
- "f1-score": 0.9022119433078337,
183
- "precision": 0.9075399665950847,
184
- "recall": 0.8969461148449476,
185
- "support": 8481.0
186
  },
187
- "eval_accuracy": 0.9410207432726785,
188
- "eval_loss": 0.2340461015701294,
189
  "eval_macro avg": {
190
- "f1-score": 0.9192797433555584,
191
- "precision": 0.9192054822213569,
192
- "recall": 0.9193867012479867,
193
- "support": 29841.0
194
- },
195
- "eval_runtime": 4.6719,
196
- "eval_samples_per_second": 17.124,
197
- "eval_steps_per_second": 2.14,
198
  "eval_weighted avg": {
199
- "f1-score": 0.9409353496612266,
200
- "precision": 0.9408751053176122,
201
- "recall": 0.9410207432726785,
202
- "support": 29841.0
203
  },
204
  "step": 405
205
  },
206
  {
207
  "epoch": 6.0,
208
  "eval_B": {
209
- "f1-score": 0.8929831438651511,
210
- "precision": 0.9068471337579618,
211
- "recall": 0.8795366795366796,
212
- "support": 1295.0
213
  },
214
  "eval_I": {
215
- "f1-score": 0.9522293798155866,
216
- "precision": 0.9655225409836066,
217
- "recall": 0.9392972838275604,
218
- "support": 20065.0
219
  },
220
  "eval_O": {
221
- "f1-score": 0.8885216003647556,
222
- "precision": 0.8599007170435742,
223
- "recall": 0.919113312109421,
224
- "support": 8481.0
225
  },
226
- "eval_accuracy": 0.930967460875976,
227
- "eval_loss": 0.21774999797344208,
228
  "eval_macro avg": {
229
- "f1-score": 0.9112447080151643,
230
- "precision": 0.9107567972617142,
231
- "recall": 0.9126490918245537,
232
- "support": 29841.0
233
- },
234
- "eval_runtime": 4.647,
235
- "eval_samples_per_second": 17.215,
236
- "eval_steps_per_second": 2.152,
237
  "eval_weighted avg": {
238
- "f1-score": 0.9315521386682287,
239
- "precision": 0.9329578366776978,
240
- "recall": 0.930967460875976,
241
- "support": 29841.0
242
  },
243
  "step": 486
244
  },
245
  {
246
  "epoch": 6.17,
247
- "grad_norm": 9.310327529907227,
248
  "learning_rate": 1.7530864197530865e-05,
249
- "loss": 0.1627,
250
  "step": 500
251
  },
252
  {
253
  "epoch": 7.0,
254
  "eval_B": {
255
- "f1-score": 0.8936490041337843,
256
- "precision": 0.8704245973645681,
257
- "recall": 0.9181467181467181,
258
- "support": 1295.0
259
  },
260
  "eval_I": {
261
- "f1-score": 0.9599271850630982,
262
- "precision": 0.9477800446905664,
263
- "recall": 0.9723897333665587,
264
- "support": 20065.0
265
  },
266
  "eval_O": {
267
- "f1-score": 0.8979841172877214,
268
- "precision": 0.9316770186335404,
269
- "recall": 0.8666430845419172,
270
- "support": 8481.0
271
  },
272
- "eval_accuracy": 0.9399819040916859,
273
- "eval_loss": 0.27376389503479004,
274
  "eval_macro avg": {
275
- "f1-score": 0.9171867688282013,
276
- "precision": 0.9166272202295582,
277
- "recall": 0.9190598453517312,
278
- "support": 29841.0
279
- },
280
- "eval_runtime": 4.6671,
281
- "eval_samples_per_second": 17.141,
282
- "eval_steps_per_second": 2.143,
283
  "eval_weighted avg": {
284
- "f1-score": 0.9394463230910989,
285
- "precision": 0.9398464945991887,
286
- "recall": 0.9399819040916859,
287
- "support": 29841.0
288
  },
289
  "step": 567
290
  },
291
  {
292
  "epoch": 8.0,
293
  "eval_B": {
294
- "f1-score": 0.8948545861297539,
295
- "precision": 0.8651766402307137,
296
- "recall": 0.9266409266409267,
297
- "support": 1295.0
298
  },
299
  "eval_I": {
300
- "f1-score": 0.9590139880661254,
301
- "precision": 0.9414702069429106,
302
- "recall": 0.9772240219287316,
303
- "support": 20065.0
304
  },
305
  "eval_O": {
306
- "f1-score": 0.8927241122423641,
307
- "precision": 0.9427035531663825,
308
- "recall": 0.8477773847423653,
309
- "support": 8481.0
310
  },
311
- "eval_accuracy": 0.9382393351429241,
312
- "eval_loss": 0.3103167414665222,
313
  "eval_macro avg": {
314
- "f1-score": 0.9155308954794145,
315
- "precision": 0.916450133446669,
316
- "recall": 0.9172141111040079,
317
- "support": 29841.0
318
- },
319
- "eval_runtime": 4.6339,
320
- "eval_samples_per_second": 17.264,
321
- "eval_steps_per_second": 2.158,
322
  "eval_weighted avg": {
323
- "f1-score": 0.9373896838414373,
324
- "precision": 0.9385098450391195,
325
- "recall": 0.9382393351429241,
326
- "support": 29841.0
327
  },
328
  "step": 648
329
  },
330
  {
331
  "epoch": 9.0,
332
  "eval_B": {
333
- "f1-score": 0.9011235955056179,
334
- "precision": 0.8749090909090909,
335
- "recall": 0.9289575289575289,
336
- "support": 1295.0
337
  },
338
  "eval_I": {
339
- "f1-score": 0.9615119937082186,
340
- "precision": 0.9485040973670175,
341
- "recall": 0.9748816346872664,
342
- "support": 20065.0
343
  },
344
  "eval_O": {
345
- "f1-score": 0.9013722126929674,
346
- "precision": 0.9380339155935229,
347
- "recall": 0.8674684589081476,
348
- "support": 8481.0
349
  },
350
- "eval_accuracy": 0.9423611809255722,
351
- "eval_loss": 0.32883673906326294,
352
  "eval_macro avg": {
353
- "f1-score": 0.9213359339689347,
354
- "precision": 0.9204823679565438,
355
- "recall": 0.9237692075176476,
356
- "support": 29841.0
357
- },
358
- "eval_runtime": 4.6714,
359
- "eval_samples_per_second": 17.126,
360
- "eval_steps_per_second": 2.141,
361
  "eval_weighted avg": {
362
- "f1-score": 0.9417992341337168,
363
- "precision": 0.9423346276781992,
364
- "recall": 0.9423611809255722,
365
- "support": 29841.0
366
  },
367
  "step": 729
368
  },
369
  {
370
  "epoch": 10.0,
371
  "eval_B": {
372
- "f1-score": 0.8935361216730039,
373
- "precision": 0.8801498127340824,
374
- "recall": 0.9073359073359073,
375
- "support": 1295.0
376
  },
377
  "eval_I": {
378
- "f1-score": 0.9632332861646931,
379
- "precision": 0.9596359319351009,
380
- "recall": 0.9668577124345876,
381
- "support": 20065.0
382
  },
383
  "eval_O": {
384
- "f1-score": 0.9095462405342556,
385
- "precision": 0.9200241254523522,
386
- "recall": 0.8993043273198915,
387
- "support": 8481.0
388
  },
389
- "eval_accuracy": 0.9450755671726819,
390
- "eval_loss": 0.3041447699069977,
391
  "eval_macro avg": {
392
- "f1-score": 0.9221052161239841,
393
- "precision": 0.9199366233738452,
394
- "recall": 0.9244993156967954,
395
- "support": 29841.0
396
- },
397
- "eval_runtime": 4.6532,
398
- "eval_samples_per_second": 17.193,
399
- "eval_steps_per_second": 2.149,
400
  "eval_weighted avg": {
401
- "f1-score": 0.9449504651463465,
402
- "precision": 0.9449285744355027,
403
- "recall": 0.9450755671726819,
404
- "support": 29841.0
405
  },
406
  "step": 810
407
  }
@@ -411,7 +411,7 @@
411
  "num_input_tokens_seen": 0,
412
  "num_train_epochs": 50,
413
  "save_steps": 500,
414
- "total_flos": 1437866527356000.0,
415
  "train_batch_size": 4,
416
  "trial_name": null,
417
  "trial_params": null
 
11
  {
12
  "epoch": 1.0,
13
  "eval_B": {
14
+ "f1-score": 0.7905866302864938,
15
+ "precision": 0.7768096514745308,
16
+ "recall": 0.8048611111111111,
17
+ "support": 1440.0
18
  },
19
  "eval_I": {
20
+ "f1-score": 0.9005331227169513,
21
+ "precision": 0.9637593111099372,
22
+ "recall": 0.8450919534905267,
23
+ "support": 21587.0
24
  },
25
  "eval_O": {
26
+ "f1-score": 0.8288892663043479,
27
+ "precision": 0.7463108800367001,
28
+ "recall": 0.9320156593144275,
29
+ "support": 10473.0
30
  },
31
+ "eval_accuracy": 0.8705373134328358,
32
+ "eval_loss": 0.32930415868759155,
33
  "eval_macro avg": {
34
+ "f1-score": 0.8400030064359308,
35
+ "precision": 0.8289599475403894,
36
+ "recall": 0.8606562413053552,
37
+ "support": 33500.0
38
+ },
39
+ "eval_runtime": 4.7349,
40
+ "eval_samples_per_second": 17.107,
41
+ "eval_steps_per_second": 2.323,
42
  "eval_weighted avg": {
43
+ "f1-score": 0.8734092702599646,
44
+ "precision": 0.8877430445874,
45
+ "recall": 0.8705373134328358,
46
+ "support": 33500.0
47
  },
48
  "step": 81
49
  },
50
  {
51
  "epoch": 2.0,
52
  "eval_B": {
53
+ "f1-score": 0.8656218739579861,
54
+ "precision": 0.8325849903784477,
55
+ "recall": 0.9013888888888889,
56
+ "support": 1440.0
57
  },
58
  "eval_I": {
59
+ "f1-score": 0.9547168948727912,
60
+ "precision": 0.9423736462093862,
61
+ "recall": 0.9673877796822161,
62
+ "support": 21587.0
63
  },
64
  "eval_O": {
65
+ "f1-score": 0.9013528191962081,
66
+ "precision": 0.9332379102341274,
67
+ "recall": 0.8715745249689678,
68
+ "support": 10473.0
69
  },
70
+ "eval_accuracy": 0.9345970149253732,
71
+ "eval_loss": 0.2162298709154129,
72
  "eval_macro avg": {
73
+ "f1-score": 0.9072305293423284,
74
+ "precision": 0.902732182273987,
75
+ "recall": 0.9134503978466909,
76
+ "support": 33500.0
77
+ },
78
+ "eval_runtime": 4.753,
79
+ "eval_samples_per_second": 17.042,
80
+ "eval_steps_per_second": 2.314,
81
  "eval_weighted avg": {
82
+ "f1-score": 0.9342040950316517,
83
+ "precision": 0.9347982961417612,
84
+ "recall": 0.9345970149253732,
85
+ "support": 33500.0
86
  },
87
  "step": 162
88
  },
89
  {
90
  "epoch": 3.0,
91
  "eval_B": {
92
+ "f1-score": 0.8603587759409076,
93
+ "precision": 0.8717034925160371,
94
+ "recall": 0.8493055555555555,
95
+ "support": 1440.0
96
  },
97
  "eval_I": {
98
+ "f1-score": 0.9434195218082959,
99
+ "precision": 0.9677057963955188,
100
+ "recall": 0.9203224162690509,
101
+ "support": 21587.0
102
  },
103
  "eval_O": {
104
+ "f1-score": 0.8879310344827586,
105
+ "precision": 0.8459410391631365,
106
+ "recall": 0.9343072663038289,
107
+ "support": 10473.0
108
  },
109
+ "eval_accuracy": 0.9216417910447762,
110
+ "eval_loss": 0.24082203209400177,
111
  "eval_macro avg": {
112
+ "f1-score": 0.8972364440773206,
113
+ "precision": 0.8951167760248975,
114
+ "recall": 0.9013117460428117,
115
+ "support": 33500.0
116
+ },
117
+ "eval_runtime": 4.7554,
118
+ "eval_samples_per_second": 17.033,
119
+ "eval_steps_per_second": 2.313,
120
  "eval_weighted avg": {
121
+ "f1-score": 0.9225019575751797,
122
+ "precision": 0.9255121957960801,
123
+ "recall": 0.9216417910447762,
124
+ "support": 33500.0
125
  },
126
  "step": 243
127
  },
128
  {
129
  "epoch": 4.0,
130
  "eval_B": {
131
+ "f1-score": 0.8804020100502512,
132
+ "precision": 0.8504854368932039,
133
+ "recall": 0.9125,
134
+ "support": 1440.0
135
  },
136
  "eval_I": {
137
+ "f1-score": 0.9589184992350025,
138
+ "precision": 0.945595388218339,
139
+ "recall": 0.9726224116366332,
140
+ "support": 21587.0
141
  },
142
  "eval_O": {
143
+ "f1-score": 0.9091178797468354,
144
+ "precision": 0.9427750999897446,
145
+ "recall": 0.877780960565263,
146
+ "support": 10473.0
147
  },
148
+ "eval_accuracy": 0.9403880597014925,
149
+ "eval_loss": 0.22083505988121033,
150
  "eval_macro avg": {
151
+ "f1-score": 0.9161461296773631,
152
+ "precision": 0.9129519750337626,
153
+ "recall": 0.9209677907339654,
154
+ "support": 33500.0
155
+ },
156
+ "eval_runtime": 4.7449,
157
+ "eval_samples_per_second": 17.071,
158
+ "eval_steps_per_second": 2.318,
159
  "eval_weighted avg": {
160
+ "f1-score": 0.9399744505088647,
161
+ "precision": 0.9406253819936745,
162
+ "recall": 0.9403880597014925,
163
+ "support": 33500.0
164
  },
165
  "step": 324
166
  },
167
  {
168
  "epoch": 5.0,
169
  "eval_B": {
170
+ "f1-score": 0.8879781420765027,
171
+ "precision": 0.8736559139784946,
172
+ "recall": 0.9027777777777778,
173
+ "support": 1440.0
174
  },
175
  "eval_I": {
176
+ "f1-score": 0.9613201259451634,
177
+ "precision": 0.9539317642765919,
178
+ "recall": 0.9688238291564367,
179
+ "support": 21587.0
180
  },
181
  "eval_O": {
182
+ "f1-score": 0.916006030835076,
183
+ "precision": 0.9334853291038858,
184
+ "recall": 0.899169292466342,
185
+ "support": 10473.0
186
  },
187
+ "eval_accuracy": 0.9442089552238806,
188
+ "eval_loss": 0.2257838249206543,
189
  "eval_macro avg": {
190
+ "f1-score": 0.921768099618914,
191
+ "precision": 0.9203576691196576,
192
+ "recall": 0.9235902998001855,
193
+ "support": 33500.0
194
+ },
195
+ "eval_runtime": 4.7845,
196
+ "eval_samples_per_second": 16.93,
197
+ "eval_steps_per_second": 2.299,
198
  "eval_weighted avg": {
199
+ "f1-score": 0.9440011117702735,
200
+ "precision": 0.9440889959185916,
201
+ "recall": 0.9442089552238806,
202
+ "support": 33500.0
203
  },
204
  "step": 405
205
  },
206
  {
207
  "epoch": 6.0,
208
  "eval_B": {
209
+ "f1-score": 0.8873579056148813,
210
+ "precision": 0.8803827751196173,
211
+ "recall": 0.8944444444444445,
212
+ "support": 1440.0
213
  },
214
  "eval_I": {
215
+ "f1-score": 0.9622060984570168,
216
+ "precision": 0.9539267015706806,
217
+ "recall": 0.9706304720433594,
218
+ "support": 21587.0
219
  },
220
  "eval_O": {
221
+ "f1-score": 0.9182769530299342,
222
+ "precision": 0.9365567911040509,
223
+ "recall": 0.9006970304592762,
224
+ "support": 10473.0
225
  },
226
+ "eval_accuracy": 0.9454925373134329,
227
+ "eval_loss": 0.2242102175951004,
228
  "eval_macro avg": {
229
+ "f1-score": 0.9226136523672773,
230
+ "precision": 0.923622089264783,
231
+ "recall": 0.9219239823156933,
232
+ "support": 33500.0
233
+ },
234
+ "eval_runtime": 4.7715,
235
+ "eval_samples_per_second": 16.976,
236
+ "eval_steps_per_second": 2.305,
237
  "eval_weighted avg": {
238
+ "f1-score": 0.945255312255509,
239
+ "precision": 0.9453351097376493,
240
+ "recall": 0.9454925373134329,
241
+ "support": 33500.0
242
  },
243
  "step": 486
244
  },
245
  {
246
  "epoch": 6.17,
247
+ "grad_norm": 1.8705081939697266,
248
  "learning_rate": 1.7530864197530865e-05,
249
+ "loss": 0.1601,
250
  "step": 500
251
  },
252
  {
253
  "epoch": 7.0,
254
  "eval_B": {
255
+ "f1-score": 0.8876749743939911,
256
+ "precision": 0.8730691739422431,
257
+ "recall": 0.9027777777777778,
258
+ "support": 1440.0
259
  },
260
  "eval_I": {
261
+ "f1-score": 0.9611415113943341,
262
+ "precision": 0.9581108451482232,
263
+ "recall": 0.9641914114976606,
264
+ "support": 21587.0
265
  },
266
  "eval_O": {
267
+ "f1-score": 0.9164739884393064,
268
+ "precision": 0.9247594050743657,
269
+ "recall": 0.9083357204239473,
270
+ "support": 10473.0
271
  },
272
+ "eval_accuracy": 0.944089552238806,
273
+ "eval_loss": 0.22517083585262299,
274
  "eval_macro avg": {
275
+ "f1-score": 0.9217634914092105,
276
+ "precision": 0.9186464747216107,
277
+ "recall": 0.9251016365664619,
278
+ "support": 33500.0
279
+ },
280
+ "eval_runtime": 4.7926,
281
+ "eval_samples_per_second": 16.901,
282
+ "eval_steps_per_second": 2.295,
283
  "eval_weighted avg": {
284
+ "f1-score": 0.9440192791200506,
285
+ "precision": 0.9440287663891153,
286
+ "recall": 0.944089552238806,
287
+ "support": 33500.0
288
  },
289
  "step": 567
290
  },
291
  {
292
  "epoch": 8.0,
293
  "eval_B": {
294
+ "f1-score": 0.894540522210919,
295
+ "precision": 0.8740888005301524,
296
+ "recall": 0.9159722222222222,
297
+ "support": 1440.0
298
  },
299
  "eval_I": {
300
+ "f1-score": 0.9638018385639943,
301
+ "precision": 0.9540255968049378,
302
+ "recall": 0.9737805160513272,
303
+ "support": 21587.0
304
  },
305
  "eval_O": {
306
+ "f1-score": 0.9203132648066569,
307
+ "precision": 0.9441598875163202,
308
+ "recall": 0.8976415544734078,
309
+ "support": 10473.0
310
  },
311
+ "eval_accuracy": 0.9474925373134329,
312
+ "eval_loss": 0.24869142472743988,
313
  "eval_macro avg": {
314
+ "f1-score": 0.9262185418605234,
315
+ "precision": 0.9240914282838034,
316
+ "recall": 0.9291314309156524,
317
+ "support": 33500.0
318
+ },
319
+ "eval_runtime": 4.7743,
320
+ "eval_samples_per_second": 16.966,
321
+ "eval_steps_per_second": 2.304,
322
  "eval_weighted avg": {
323
+ "f1-score": 0.947228939205516,
324
+ "precision": 0.9475052218791054,
325
+ "recall": 0.9474925373134329,
326
+ "support": 33500.0
327
  },
328
  "step": 648
329
  },
330
  {
331
  "epoch": 9.0,
332
  "eval_B": {
333
+ "f1-score": 0.8851236868858015,
334
+ "precision": 0.8643282594308405,
335
+ "recall": 0.9069444444444444,
336
+ "support": 1440.0
337
  },
338
  "eval_I": {
339
+ "f1-score": 0.9617001723147618,
340
+ "precision": 0.9540067462849849,
341
+ "recall": 0.9695186918052532,
342
+ "support": 21587.0
343
  },
344
  "eval_O": {
345
+ "f1-score": 0.9165854609237966,
346
+ "precision": 0.9358272808675754,
347
+ "recall": 0.8981189725961998,
348
+ "support": 10473.0
349
  },
350
+ "eval_accuracy": 0.9445074626865672,
351
+ "eval_loss": 0.3045330345630646,
352
  "eval_macro avg": {
353
+ "f1-score": 0.9211364400414532,
354
+ "precision": 0.9180540955278002,
355
+ "recall": 0.9248607029486324,
356
+ "support": 33500.0
357
+ },
358
+ "eval_runtime": 4.7599,
359
+ "eval_samples_per_second": 17.017,
360
+ "eval_steps_per_second": 2.311,
361
  "eval_weighted avg": {
362
+ "f1-score": 0.9443044555560967,
363
+ "precision": 0.9444685205421044,
364
+ "recall": 0.9445074626865672,
365
+ "support": 33500.0
366
  },
367
  "step": 729
368
  },
369
  {
370
  "epoch": 10.0,
371
  "eval_B": {
372
+ "f1-score": 0.8815068493150685,
373
+ "precision": 0.8695945945945946,
374
+ "recall": 0.89375,
375
+ "support": 1440.0
376
  },
377
  "eval_I": {
378
+ "f1-score": 0.956297959947595,
379
+ "precision": 0.9660159757999717,
380
+ "recall": 0.9467735211006625,
381
+ "support": 21587.0
382
  },
383
  "eval_O": {
384
+ "f1-score": 0.9101049868766404,
385
+ "precision": 0.8937678357728068,
386
+ "recall": 0.9270505108373914,
387
+ "support": 10473.0
388
  },
389
+ "eval_accuracy": 0.9383283582089552,
390
+ "eval_loss": 0.32627996802330017,
391
  "eval_macro avg": {
392
+ "f1-score": 0.9159699320464346,
393
+ "precision": 0.909792802055791,
394
+ "recall": 0.9225246773126846,
395
+ "support": 33500.0
396
+ },
397
+ "eval_runtime": 4.7534,
398
+ "eval_samples_per_second": 17.04,
399
+ "eval_steps_per_second": 2.314,
400
  "eval_weighted avg": {
401
+ "f1-score": 0.9386418940884026,
402
+ "precision": 0.9392845859659347,
403
+ "recall": 0.9383283582089552,
404
+ "support": 33500.0
405
  },
406
  "step": 810
407
  }
 
411
  "num_input_tokens_seen": 0,
412
  "num_train_epochs": 50,
413
  "save_steps": 500,
414
+ "total_flos": 1433401103358000.0,
415
  "train_batch_size": 4,
416
  "trial_name": null,
417
  "trial_params": null
checkpoint-810/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d229d7bc9127d222c880edfd3fa9418a9ccede656fe27f7f0aa04990c9605be4
3
  size 4856
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8ca363922308e3892528dd031742ffedc372734b464a042d09444af9700901e
3
  size 4856