anoaky commited on
Commit
f4e8fc1
·
verified ·
1 Parent(s): 2c53f7c

Training in progress, epoch 2, checkpoint

Browse files
checkpoint-1110/config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/bart-large-mnli",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BartForSequenceClassification"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.0,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 1024,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 4096,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
+ "encoder_attention_heads": 16,
22
+ "encoder_ffn_dim": 4096,
23
+ "encoder_layerdrop": 0.0,
24
+ "encoder_layers": 12,
25
+ "eos_token_id": 2,
26
+ "forced_eos_token_id": 2,
27
+ "gradient_checkpointing": false,
28
+ "id2label": {
29
+ "0": "LABEL_0",
30
+ "1": "LABEL_1"
31
+ },
32
+ "init_std": 0.02,
33
+ "is_encoder_decoder": true,
34
+ "label2id": {
35
+ "LABEL_0": 0,
36
+ "LABEL_1": 1
37
+ },
38
+ "max_position_embeddings": 1024,
39
+ "model_type": "bart",
40
+ "normalize_before": false,
41
+ "num_hidden_layers": 12,
42
+ "output_past": false,
43
+ "pad_token_id": 1,
44
+ "scale_embedding": false,
45
+ "torch_dtype": "float32",
46
+ "transformers_version": "4.47.0",
47
+ "use_cache": true,
48
+ "vocab_size": 50265
49
+ }
checkpoint-1110/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:188baab2aaa5966f543ecb0e243aeddb55f5cea36563403397fa5b2138fd77a9
3
+ size 1629432864
checkpoint-1110/trainer_state.json ADDED
@@ -0,0 +1,840 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 693,
6
+ "global_step": 1110,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0,
13
+ "eval_f1": 0.6193506281612009,
14
+ "eval_fn": 512,
15
+ "eval_fp": 1821,
16
+ "eval_loss": 0.736859142780304,
17
+ "eval_precision": 0.5103522452272116,
18
+ "eval_recall": 0.787551867219917,
19
+ "eval_runtime": 69.9291,
20
+ "eval_samples_per_second": 66.825,
21
+ "eval_steps_per_second": 8.366,
22
+ "eval_tn": 442,
23
+ "eval_tp": 1898,
24
+ "step": 0
25
+ },
26
+ {
27
+ "epoch": 0.018026137899954935,
28
+ "grad_norm": 32.56014633178711,
29
+ "learning_rate": 4.981949458483755e-05,
30
+ "loss": 0.8382,
31
+ "step": 10
32
+ },
33
+ {
34
+ "epoch": 0.03605227579990987,
35
+ "grad_norm": 27.728681564331055,
36
+ "learning_rate": 4.963898916967509e-05,
37
+ "loss": 0.5565,
38
+ "step": 20
39
+ },
40
+ {
41
+ "epoch": 0.054078413699864804,
42
+ "grad_norm": 26.70035171508789,
43
+ "learning_rate": 4.945848375451264e-05,
44
+ "loss": 0.5545,
45
+ "step": 30
46
+ },
47
+ {
48
+ "epoch": 0.07210455159981974,
49
+ "grad_norm": 23.187734603881836,
50
+ "learning_rate": 4.927797833935018e-05,
51
+ "loss": 0.6027,
52
+ "step": 40
53
+ },
54
+ {
55
+ "epoch": 0.09013068949977468,
56
+ "grad_norm": 35.976619720458984,
57
+ "learning_rate": 4.909747292418773e-05,
58
+ "loss": 0.5741,
59
+ "step": 50
60
+ },
61
+ {
62
+ "epoch": 0.10815682739972961,
63
+ "grad_norm": 27.495447158813477,
64
+ "learning_rate": 4.891696750902527e-05,
65
+ "loss": 0.4941,
66
+ "step": 60
67
+ },
68
+ {
69
+ "epoch": 0.12618296529968454,
70
+ "grad_norm": 25.021883010864258,
71
+ "learning_rate": 4.873646209386282e-05,
72
+ "loss": 0.457,
73
+ "step": 70
74
+ },
75
+ {
76
+ "epoch": 0.14420910319963948,
77
+ "grad_norm": 29.016551971435547,
78
+ "learning_rate": 4.855595667870036e-05,
79
+ "loss": 0.4215,
80
+ "step": 80
81
+ },
82
+ {
83
+ "epoch": 0.16223524109959442,
84
+ "grad_norm": 25.652786254882812,
85
+ "learning_rate": 4.837545126353791e-05,
86
+ "loss": 0.4292,
87
+ "step": 90
88
+ },
89
+ {
90
+ "epoch": 0.18026137899954936,
91
+ "grad_norm": 24.254980087280273,
92
+ "learning_rate": 4.819494584837546e-05,
93
+ "loss": 0.4559,
94
+ "step": 100
95
+ },
96
+ {
97
+ "epoch": 0.19828751689950427,
98
+ "grad_norm": 43.745643615722656,
99
+ "learning_rate": 4.8014440433213e-05,
100
+ "loss": 0.4536,
101
+ "step": 110
102
+ },
103
+ {
104
+ "epoch": 0.21631365479945922,
105
+ "grad_norm": 19.265901565551758,
106
+ "learning_rate": 4.783393501805055e-05,
107
+ "loss": 0.4634,
108
+ "step": 120
109
+ },
110
+ {
111
+ "epoch": 0.23433979269941416,
112
+ "grad_norm": 75.13768768310547,
113
+ "learning_rate": 4.765342960288809e-05,
114
+ "loss": 0.4067,
115
+ "step": 130
116
+ },
117
+ {
118
+ "epoch": 0.25236593059936907,
119
+ "grad_norm": 24.337844848632812,
120
+ "learning_rate": 4.747292418772563e-05,
121
+ "loss": 0.4483,
122
+ "step": 140
123
+ },
124
+ {
125
+ "epoch": 0.270392068499324,
126
+ "grad_norm": 27.518775939941406,
127
+ "learning_rate": 4.7292418772563177e-05,
128
+ "loss": 0.4524,
129
+ "step": 150
130
+ },
131
+ {
132
+ "epoch": 0.28841820639927895,
133
+ "grad_norm": 23.281097412109375,
134
+ "learning_rate": 4.711191335740072e-05,
135
+ "loss": 0.4443,
136
+ "step": 160
137
+ },
138
+ {
139
+ "epoch": 0.3064443442992339,
140
+ "grad_norm": 12.933683395385742,
141
+ "learning_rate": 4.693140794223827e-05,
142
+ "loss": 0.472,
143
+ "step": 170
144
+ },
145
+ {
146
+ "epoch": 0.32447048219918884,
147
+ "grad_norm": 16.49010467529297,
148
+ "learning_rate": 4.675090252707581e-05,
149
+ "loss": 0.4549,
150
+ "step": 180
151
+ },
152
+ {
153
+ "epoch": 0.3424966200991438,
154
+ "grad_norm": 19.209016799926758,
155
+ "learning_rate": 4.657039711191336e-05,
156
+ "loss": 0.4628,
157
+ "step": 190
158
+ },
159
+ {
160
+ "epoch": 0.3605227579990987,
161
+ "grad_norm": 19.71845054626465,
162
+ "learning_rate": 4.63898916967509e-05,
163
+ "loss": 0.4281,
164
+ "step": 200
165
+ },
166
+ {
167
+ "epoch": 0.3785488958990536,
168
+ "grad_norm": 19.700483322143555,
169
+ "learning_rate": 4.620938628158845e-05,
170
+ "loss": 0.4427,
171
+ "step": 210
172
+ },
173
+ {
174
+ "epoch": 0.39657503379900855,
175
+ "grad_norm": 19.06282615661621,
176
+ "learning_rate": 4.602888086642599e-05,
177
+ "loss": 0.4214,
178
+ "step": 220
179
+ },
180
+ {
181
+ "epoch": 0.4146011716989635,
182
+ "grad_norm": 14.681564331054688,
183
+ "learning_rate": 4.584837545126354e-05,
184
+ "loss": 0.4279,
185
+ "step": 230
186
+ },
187
+ {
188
+ "epoch": 0.43262730959891843,
189
+ "grad_norm": 28.783143997192383,
190
+ "learning_rate": 4.566787003610109e-05,
191
+ "loss": 0.4475,
192
+ "step": 240
193
+ },
194
+ {
195
+ "epoch": 0.45065344749887337,
196
+ "grad_norm": 14.686984062194824,
197
+ "learning_rate": 4.548736462093863e-05,
198
+ "loss": 0.4232,
199
+ "step": 250
200
+ },
201
+ {
202
+ "epoch": 0.4686795853988283,
203
+ "grad_norm": 15.651535987854004,
204
+ "learning_rate": 4.530685920577618e-05,
205
+ "loss": 0.4026,
206
+ "step": 260
207
+ },
208
+ {
209
+ "epoch": 0.48670572329878325,
210
+ "grad_norm": 17.506153106689453,
211
+ "learning_rate": 4.5126353790613716e-05,
212
+ "loss": 0.416,
213
+ "step": 270
214
+ },
215
+ {
216
+ "epoch": 0.5047318611987381,
217
+ "grad_norm": 23.381561279296875,
218
+ "learning_rate": 4.494584837545127e-05,
219
+ "loss": 0.385,
220
+ "step": 280
221
+ },
222
+ {
223
+ "epoch": 0.5227579990986931,
224
+ "grad_norm": 18.458227157592773,
225
+ "learning_rate": 4.4765342960288806e-05,
226
+ "loss": 0.4352,
227
+ "step": 290
228
+ },
229
+ {
230
+ "epoch": 0.540784136998648,
231
+ "grad_norm": 16.285743713378906,
232
+ "learning_rate": 4.458483754512636e-05,
233
+ "loss": 0.4191,
234
+ "step": 300
235
+ },
236
+ {
237
+ "epoch": 0.558810274898603,
238
+ "grad_norm": 22.14523696899414,
239
+ "learning_rate": 4.44043321299639e-05,
240
+ "loss": 0.4525,
241
+ "step": 310
242
+ },
243
+ {
244
+ "epoch": 0.5768364127985579,
245
+ "grad_norm": 12.94579029083252,
246
+ "learning_rate": 4.422382671480145e-05,
247
+ "loss": 0.4335,
248
+ "step": 320
249
+ },
250
+ {
251
+ "epoch": 0.5948625506985128,
252
+ "grad_norm": 29.218290328979492,
253
+ "learning_rate": 4.404332129963899e-05,
254
+ "loss": 0.4194,
255
+ "step": 330
256
+ },
257
+ {
258
+ "epoch": 0.6128886885984678,
259
+ "grad_norm": 20.067014694213867,
260
+ "learning_rate": 4.386281588447654e-05,
261
+ "loss": 0.4016,
262
+ "step": 340
263
+ },
264
+ {
265
+ "epoch": 0.6309148264984227,
266
+ "grad_norm": 15.315394401550293,
267
+ "learning_rate": 4.368231046931408e-05,
268
+ "loss": 0.4293,
269
+ "step": 350
270
+ },
271
+ {
272
+ "epoch": 0.6489409643983777,
273
+ "grad_norm": 28.87151527404785,
274
+ "learning_rate": 4.350180505415163e-05,
275
+ "loss": 0.3952,
276
+ "step": 360
277
+ },
278
+ {
279
+ "epoch": 0.6669671022983326,
280
+ "grad_norm": 12.91943645477295,
281
+ "learning_rate": 4.332129963898917e-05,
282
+ "loss": 0.4276,
283
+ "step": 370
284
+ },
285
+ {
286
+ "epoch": 0.6849932401982876,
287
+ "grad_norm": 18.917442321777344,
288
+ "learning_rate": 4.314079422382672e-05,
289
+ "loss": 0.3843,
290
+ "step": 380
291
+ },
292
+ {
293
+ "epoch": 0.7030193780982424,
294
+ "grad_norm": 16.344675064086914,
295
+ "learning_rate": 4.296028880866426e-05,
296
+ "loss": 0.4283,
297
+ "step": 390
298
+ },
299
+ {
300
+ "epoch": 0.7210455159981974,
301
+ "grad_norm": 27.420021057128906,
302
+ "learning_rate": 4.277978339350181e-05,
303
+ "loss": 0.4663,
304
+ "step": 400
305
+ },
306
+ {
307
+ "epoch": 0.7390716538981523,
308
+ "grad_norm": 37.13134765625,
309
+ "learning_rate": 4.259927797833935e-05,
310
+ "loss": 0.4338,
311
+ "step": 410
312
+ },
313
+ {
314
+ "epoch": 0.7570977917981072,
315
+ "grad_norm": 23.59578514099121,
316
+ "learning_rate": 4.24187725631769e-05,
317
+ "loss": 0.3714,
318
+ "step": 420
319
+ },
320
+ {
321
+ "epoch": 0.7751239296980622,
322
+ "grad_norm": 30.009662628173828,
323
+ "learning_rate": 4.223826714801444e-05,
324
+ "loss": 0.3775,
325
+ "step": 430
326
+ },
327
+ {
328
+ "epoch": 0.7931500675980171,
329
+ "grad_norm": 13.575807571411133,
330
+ "learning_rate": 4.205776173285199e-05,
331
+ "loss": 0.4104,
332
+ "step": 440
333
+ },
334
+ {
335
+ "epoch": 0.8111762054979721,
336
+ "grad_norm": 18.727609634399414,
337
+ "learning_rate": 4.187725631768953e-05,
338
+ "loss": 0.3804,
339
+ "step": 450
340
+ },
341
+ {
342
+ "epoch": 0.829202343397927,
343
+ "grad_norm": 14.367807388305664,
344
+ "learning_rate": 4.169675090252708e-05,
345
+ "loss": 0.4124,
346
+ "step": 460
347
+ },
348
+ {
349
+ "epoch": 0.847228481297882,
350
+ "grad_norm": 12.945828437805176,
351
+ "learning_rate": 4.151624548736462e-05,
352
+ "loss": 0.4149,
353
+ "step": 470
354
+ },
355
+ {
356
+ "epoch": 0.8652546191978369,
357
+ "grad_norm": 24.02052116394043,
358
+ "learning_rate": 4.1335740072202167e-05,
359
+ "loss": 0.4093,
360
+ "step": 480
361
+ },
362
+ {
363
+ "epoch": 0.8832807570977917,
364
+ "grad_norm": 15.609790802001953,
365
+ "learning_rate": 4.115523465703972e-05,
366
+ "loss": 0.3874,
367
+ "step": 490
368
+ },
369
+ {
370
+ "epoch": 0.9013068949977467,
371
+ "grad_norm": 25.04174041748047,
372
+ "learning_rate": 4.0974729241877256e-05,
373
+ "loss": 0.391,
374
+ "step": 500
375
+ },
376
+ {
377
+ "epoch": 0.9193330328977016,
378
+ "grad_norm": 20.739681243896484,
379
+ "learning_rate": 4.079422382671481e-05,
380
+ "loss": 0.3939,
381
+ "step": 510
382
+ },
383
+ {
384
+ "epoch": 0.9373591707976566,
385
+ "grad_norm": 16.13037872314453,
386
+ "learning_rate": 4.0613718411552346e-05,
387
+ "loss": 0.4004,
388
+ "step": 520
389
+ },
390
+ {
391
+ "epoch": 0.9553853086976115,
392
+ "grad_norm": 21.17761993408203,
393
+ "learning_rate": 4.043321299638989e-05,
394
+ "loss": 0.3978,
395
+ "step": 530
396
+ },
397
+ {
398
+ "epoch": 0.9734114465975665,
399
+ "grad_norm": 27.03806495666504,
400
+ "learning_rate": 4.0252707581227436e-05,
401
+ "loss": 0.4391,
402
+ "step": 540
403
+ },
404
+ {
405
+ "epoch": 0.9914375844975214,
406
+ "grad_norm": 15.054065704345703,
407
+ "learning_rate": 4.007220216606498e-05,
408
+ "loss": 0.4187,
409
+ "step": 550
410
+ },
411
+ {
412
+ "epoch": 1.0090130689499774,
413
+ "grad_norm": 22.64523696899414,
414
+ "learning_rate": 3.989169675090253e-05,
415
+ "loss": 0.3444,
416
+ "step": 560
417
+ },
418
+ {
419
+ "epoch": 1.0270392068499323,
420
+ "grad_norm": 21.697904586791992,
421
+ "learning_rate": 3.971119133574007e-05,
422
+ "loss": 0.2976,
423
+ "step": 570
424
+ },
425
+ {
426
+ "epoch": 1.0450653447498874,
427
+ "grad_norm": 14.764150619506836,
428
+ "learning_rate": 3.953068592057762e-05,
429
+ "loss": 0.3295,
430
+ "step": 580
431
+ },
432
+ {
433
+ "epoch": 1.0630914826498423,
434
+ "grad_norm": 25.92884635925293,
435
+ "learning_rate": 3.935018050541516e-05,
436
+ "loss": 0.2984,
437
+ "step": 590
438
+ },
439
+ {
440
+ "epoch": 1.0811176205497972,
441
+ "grad_norm": 23.327699661254883,
442
+ "learning_rate": 3.916967509025271e-05,
443
+ "loss": 0.3668,
444
+ "step": 600
445
+ },
446
+ {
447
+ "epoch": 1.099143758449752,
448
+ "grad_norm": 27.034189224243164,
449
+ "learning_rate": 3.898916967509025e-05,
450
+ "loss": 0.2852,
451
+ "step": 610
452
+ },
453
+ {
454
+ "epoch": 1.117169896349707,
455
+ "grad_norm": 29.098979949951172,
456
+ "learning_rate": 3.88086642599278e-05,
457
+ "loss": 0.2913,
458
+ "step": 620
459
+ },
460
+ {
461
+ "epoch": 1.135196034249662,
462
+ "grad_norm": 21.875049591064453,
463
+ "learning_rate": 3.862815884476535e-05,
464
+ "loss": 0.2853,
465
+ "step": 630
466
+ },
467
+ {
468
+ "epoch": 1.153222172149617,
469
+ "grad_norm": 15.405293464660645,
470
+ "learning_rate": 3.844765342960289e-05,
471
+ "loss": 0.3296,
472
+ "step": 640
473
+ },
474
+ {
475
+ "epoch": 1.1712483100495719,
476
+ "grad_norm": 21.292566299438477,
477
+ "learning_rate": 3.826714801444044e-05,
478
+ "loss": 0.2739,
479
+ "step": 650
480
+ },
481
+ {
482
+ "epoch": 1.1892744479495267,
483
+ "grad_norm": 17.985347747802734,
484
+ "learning_rate": 3.8086642599277976e-05,
485
+ "loss": 0.3034,
486
+ "step": 660
487
+ },
488
+ {
489
+ "epoch": 1.2073005858494819,
490
+ "grad_norm": 16.177711486816406,
491
+ "learning_rate": 3.790613718411553e-05,
492
+ "loss": 0.3032,
493
+ "step": 670
494
+ },
495
+ {
496
+ "epoch": 1.2253267237494367,
497
+ "grad_norm": 38.261051177978516,
498
+ "learning_rate": 3.7725631768953066e-05,
499
+ "loss": 0.2819,
500
+ "step": 680
501
+ },
502
+ {
503
+ "epoch": 1.2433528616493916,
504
+ "grad_norm": 24.247678756713867,
505
+ "learning_rate": 3.754512635379062e-05,
506
+ "loss": 0.3527,
507
+ "step": 690
508
+ },
509
+ {
510
+ "epoch": 1.2487607030193781,
511
+ "eval_f1": 0.8497959183673469,
512
+ "eval_fn": 328,
513
+ "eval_fp": 408,
514
+ "eval_loss": 0.36085668206214905,
515
+ "eval_precision": 0.8361445783132531,
516
+ "eval_recall": 0.8639004149377594,
517
+ "eval_runtime": 63.9221,
518
+ "eval_samples_per_second": 73.105,
519
+ "eval_steps_per_second": 9.152,
520
+ "eval_tn": 1855,
521
+ "eval_tp": 2082,
522
+ "step": 693
523
+ },
524
+ {
525
+ "epoch": 1.2613789995493465,
526
+ "grad_norm": 26.0637149810791,
527
+ "learning_rate": 3.7364620938628155e-05,
528
+ "loss": 0.2519,
529
+ "step": 700
530
+ },
531
+ {
532
+ "epoch": 1.2794051374493014,
533
+ "grad_norm": 22.626008987426758,
534
+ "learning_rate": 3.718411552346571e-05,
535
+ "loss": 0.2764,
536
+ "step": 710
537
+ },
538
+ {
539
+ "epoch": 1.2974312753492563,
540
+ "grad_norm": 24.119760513305664,
541
+ "learning_rate": 3.700361010830325e-05,
542
+ "loss": 0.3114,
543
+ "step": 720
544
+ },
545
+ {
546
+ "epoch": 1.3154574132492114,
547
+ "grad_norm": 25.81672477722168,
548
+ "learning_rate": 3.68231046931408e-05,
549
+ "loss": 0.2792,
550
+ "step": 730
551
+ },
552
+ {
553
+ "epoch": 1.3334835511491663,
554
+ "grad_norm": 15.591266632080078,
555
+ "learning_rate": 3.664259927797834e-05,
556
+ "loss": 0.2857,
557
+ "step": 740
558
+ },
559
+ {
560
+ "epoch": 1.3515096890491212,
561
+ "grad_norm": 21.272573471069336,
562
+ "learning_rate": 3.646209386281589e-05,
563
+ "loss": 0.2894,
564
+ "step": 750
565
+ },
566
+ {
567
+ "epoch": 1.3695358269490763,
568
+ "grad_norm": 24.08788299560547,
569
+ "learning_rate": 3.628158844765343e-05,
570
+ "loss": 0.2806,
571
+ "step": 760
572
+ },
573
+ {
574
+ "epoch": 1.3875619648490312,
575
+ "grad_norm": 20.197420120239258,
576
+ "learning_rate": 3.610108303249098e-05,
577
+ "loss": 0.3313,
578
+ "step": 770
579
+ },
580
+ {
581
+ "epoch": 1.405588102748986,
582
+ "grad_norm": 22.551584243774414,
583
+ "learning_rate": 3.592057761732852e-05,
584
+ "loss": 0.3261,
585
+ "step": 780
586
+ },
587
+ {
588
+ "epoch": 1.423614240648941,
589
+ "grad_norm": 15.885344505310059,
590
+ "learning_rate": 3.574007220216607e-05,
591
+ "loss": 0.3198,
592
+ "step": 790
593
+ },
594
+ {
595
+ "epoch": 1.4416403785488958,
596
+ "grad_norm": 30.44365882873535,
597
+ "learning_rate": 3.555956678700361e-05,
598
+ "loss": 0.3143,
599
+ "step": 800
600
+ },
601
+ {
602
+ "epoch": 1.4596665164488507,
603
+ "grad_norm": 18.411624908447266,
604
+ "learning_rate": 3.537906137184116e-05,
605
+ "loss": 0.2787,
606
+ "step": 810
607
+ },
608
+ {
609
+ "epoch": 1.4776926543488058,
610
+ "grad_norm": 18.90251922607422,
611
+ "learning_rate": 3.51985559566787e-05,
612
+ "loss": 0.2858,
613
+ "step": 820
614
+ },
615
+ {
616
+ "epoch": 1.4957187922487607,
617
+ "grad_norm": 28.24523162841797,
618
+ "learning_rate": 3.5018050541516247e-05,
619
+ "loss": 0.2856,
620
+ "step": 830
621
+ },
622
+ {
623
+ "epoch": 1.5137449301487156,
624
+ "grad_norm": 20.0897216796875,
625
+ "learning_rate": 3.483754512635379e-05,
626
+ "loss": 0.324,
627
+ "step": 840
628
+ },
629
+ {
630
+ "epoch": 1.5317710680486707,
631
+ "grad_norm": 20.818204879760742,
632
+ "learning_rate": 3.4657039711191336e-05,
633
+ "loss": 0.2933,
634
+ "step": 850
635
+ },
636
+ {
637
+ "epoch": 1.5497972059486256,
638
+ "grad_norm": 25.051939010620117,
639
+ "learning_rate": 3.447653429602888e-05,
640
+ "loss": 0.2948,
641
+ "step": 860
642
+ },
643
+ {
644
+ "epoch": 1.5678233438485805,
645
+ "grad_norm": 32.370933532714844,
646
+ "learning_rate": 3.4296028880866426e-05,
647
+ "loss": 0.2847,
648
+ "step": 870
649
+ },
650
+ {
651
+ "epoch": 1.5858494817485354,
652
+ "grad_norm": 34.086158752441406,
653
+ "learning_rate": 3.411552346570397e-05,
654
+ "loss": 0.2805,
655
+ "step": 880
656
+ },
657
+ {
658
+ "epoch": 1.6038756196484902,
659
+ "grad_norm": 17.575464248657227,
660
+ "learning_rate": 3.3935018050541516e-05,
661
+ "loss": 0.3013,
662
+ "step": 890
663
+ },
664
+ {
665
+ "epoch": 1.6219017575484451,
666
+ "grad_norm": 16.147525787353516,
667
+ "learning_rate": 3.375451263537907e-05,
668
+ "loss": 0.3257,
669
+ "step": 900
670
+ },
671
+ {
672
+ "epoch": 1.6399278954484002,
673
+ "grad_norm": 23.968366622924805,
674
+ "learning_rate": 3.3574007220216606e-05,
675
+ "loss": 0.3042,
676
+ "step": 910
677
+ },
678
+ {
679
+ "epoch": 1.6579540333483551,
680
+ "grad_norm": 24.20454216003418,
681
+ "learning_rate": 3.339350180505416e-05,
682
+ "loss": 0.3174,
683
+ "step": 920
684
+ },
685
+ {
686
+ "epoch": 1.67598017124831,
687
+ "grad_norm": 14.595170021057129,
688
+ "learning_rate": 3.3212996389891696e-05,
689
+ "loss": 0.305,
690
+ "step": 930
691
+ },
692
+ {
693
+ "epoch": 1.694006309148265,
694
+ "grad_norm": 46.18140411376953,
695
+ "learning_rate": 3.303249097472924e-05,
696
+ "loss": 0.3085,
697
+ "step": 940
698
+ },
699
+ {
700
+ "epoch": 1.71203244704822,
701
+ "grad_norm": 25.23206329345703,
702
+ "learning_rate": 3.2851985559566786e-05,
703
+ "loss": 0.2921,
704
+ "step": 950
705
+ },
706
+ {
707
+ "epoch": 1.7300585849481749,
708
+ "grad_norm": 18.097789764404297,
709
+ "learning_rate": 3.267148014440433e-05,
710
+ "loss": 0.2792,
711
+ "step": 960
712
+ },
713
+ {
714
+ "epoch": 1.7480847228481298,
715
+ "grad_norm": 25.6617488861084,
716
+ "learning_rate": 3.249097472924188e-05,
717
+ "loss": 0.3027,
718
+ "step": 970
719
+ },
720
+ {
721
+ "epoch": 1.7661108607480847,
722
+ "grad_norm": 33.289459228515625,
723
+ "learning_rate": 3.231046931407942e-05,
724
+ "loss": 0.2854,
725
+ "step": 980
726
+ },
727
+ {
728
+ "epoch": 1.7841369986480395,
729
+ "grad_norm": 20.507280349731445,
730
+ "learning_rate": 3.212996389891697e-05,
731
+ "loss": 0.3646,
732
+ "step": 990
733
+ },
734
+ {
735
+ "epoch": 1.8021631365479944,
736
+ "grad_norm": 17.23845100402832,
737
+ "learning_rate": 3.194945848375451e-05,
738
+ "loss": 0.2958,
739
+ "step": 1000
740
+ },
741
+ {
742
+ "epoch": 1.8201892744479495,
743
+ "grad_norm": 23.676586151123047,
744
+ "learning_rate": 3.176895306859206e-05,
745
+ "loss": 0.2935,
746
+ "step": 1010
747
+ },
748
+ {
749
+ "epoch": 1.8382154123479044,
750
+ "grad_norm": 12.614605903625488,
751
+ "learning_rate": 3.15884476534296e-05,
752
+ "loss": 0.2903,
753
+ "step": 1020
754
+ },
755
+ {
756
+ "epoch": 1.8562415502478595,
757
+ "grad_norm": 42.599510192871094,
758
+ "learning_rate": 3.140794223826715e-05,
759
+ "loss": 0.3415,
760
+ "step": 1030
761
+ },
762
+ {
763
+ "epoch": 1.8742676881478144,
764
+ "grad_norm": 31.382732391357422,
765
+ "learning_rate": 3.12274368231047e-05,
766
+ "loss": 0.31,
767
+ "step": 1040
768
+ },
769
+ {
770
+ "epoch": 1.8922938260477693,
771
+ "grad_norm": 41.608890533447266,
772
+ "learning_rate": 3.104693140794224e-05,
773
+ "loss": 0.2515,
774
+ "step": 1050
775
+ },
776
+ {
777
+ "epoch": 1.9103199639477242,
778
+ "grad_norm": 28.570653915405273,
779
+ "learning_rate": 3.086642599277979e-05,
780
+ "loss": 0.325,
781
+ "step": 1060
782
+ },
783
+ {
784
+ "epoch": 1.928346101847679,
785
+ "grad_norm": 28.502283096313477,
786
+ "learning_rate": 3.0685920577617325e-05,
787
+ "loss": 0.2759,
788
+ "step": 1070
789
+ },
790
+ {
791
+ "epoch": 1.946372239747634,
792
+ "grad_norm": 16.24626350402832,
793
+ "learning_rate": 3.0505415162454877e-05,
794
+ "loss": 0.2995,
795
+ "step": 1080
796
+ },
797
+ {
798
+ "epoch": 1.9643983776475888,
799
+ "grad_norm": 27.27639389038086,
800
+ "learning_rate": 3.032490974729242e-05,
801
+ "loss": 0.2787,
802
+ "step": 1090
803
+ },
804
+ {
805
+ "epoch": 1.982424515547544,
806
+ "grad_norm": 27.928380966186523,
807
+ "learning_rate": 3.0144404332129967e-05,
808
+ "loss": 0.2762,
809
+ "step": 1100
810
+ },
811
+ {
812
+ "epoch": 2.0,
813
+ "grad_norm": 22.15119743347168,
814
+ "learning_rate": 2.996389891696751e-05,
815
+ "loss": 0.2623,
816
+ "step": 1110
817
+ }
818
+ ],
819
+ "logging_steps": 10,
820
+ "max_steps": 2770,
821
+ "num_input_tokens_seen": 0,
822
+ "num_train_epochs": 5,
823
+ "save_steps": 500,
824
+ "stateful_callbacks": {
825
+ "TrainerControl": {
826
+ "args": {
827
+ "should_epoch_stop": false,
828
+ "should_evaluate": false,
829
+ "should_log": false,
830
+ "should_save": true,
831
+ "should_training_stop": false
832
+ },
833
+ "attributes": {}
834
+ }
835
+ },
836
+ "total_flos": 3.858514256176742e+16,
837
+ "train_batch_size": 8,
838
+ "trial_name": null,
839
+ "trial_params": null
840
+ }
checkpoint-1110/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bac6e5d54145c75fc6db520c88eaa2ca8beb7fe2c31ead4c74ee603f7febb3f
3
+ size 5432