Chua, Vui Seng commited on
Commit
6484a29
1 Parent(s): fb9d0c0

Initial commit

Browse files

all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.8411310008136696,
4
+ "eval_loss": 0.5979823470115662,
5
+ "eval_runtime": 14.7831,
6
+ "eval_samples": 9832,
7
+ "eval_samples_per_second": 665.086,
8
+ "eval_steps_per_second": 41.602,
9
+ "train_loss": 0.3792748264984585,
10
+ "train_runtime": 5994.6559,
11
+ "train_samples": 392702,
12
+ "train_samples_per_second": 196.526,
13
+ "train_steps_per_second": 12.283
14
+ }
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bert-base-uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "finetuning_task": "mnli",
9
+ "gradient_checkpointing": false,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "entailment",
15
+ "1": "neutral",
16
+ "2": "contradiction"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 3072,
20
+ "label2id": {
21
+ "contradiction": 2,
22
+ "entailment": 0,
23
+ "neutral": 1
24
+ },
25
+ "layer_norm_eps": 1e-12,
26
+ "max_position_embeddings": 512,
27
+ "model_type": "bert",
28
+ "num_attention_heads": 12,
29
+ "num_hidden_layers": 12,
30
+ "pad_token_id": 0,
31
+ "position_embedding_type": "absolute",
32
+ "problem_type": "single_label_classification",
33
+ "torch_dtype": "float32",
34
+ "transformers_version": "4.10.3",
35
+ "type_vocab_size": 2,
36
+ "use_cache": true,
37
+ "vocab_size": 30522
38
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.8411310008136696,
4
+ "eval_loss": 0.5979823470115662,
5
+ "eval_runtime": 14.7831,
6
+ "eval_samples": 9832,
7
+ "eval_samples_per_second": 665.086,
8
+ "eval_steps_per_second": 41.602
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:700ad3dcd617847dffe47936516873bd6e16af283fd83ed41904c75ae3e5df09
3
+ size 438022317
run.log ADDED
The diff for this file is too large to render. See raw diff
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.3792748264984585,
4
+ "train_runtime": 5994.6559,
5
+ "train_samples": 392702,
6
+ "train_samples_per_second": 196.526,
7
+ "train_steps_per_second": 12.283
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,907 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 73632,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 4.9660473707083876e-05,
13
+ "loss": 0.8682,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.04,
18
+ "learning_rate": 4.932094741416776e-05,
19
+ "loss": 0.7273,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.06,
24
+ "learning_rate": 4.898142112125163e-05,
25
+ "loss": 0.6857,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.08,
30
+ "learning_rate": 4.864189482833551e-05,
31
+ "loss": 0.6453,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.1,
36
+ "learning_rate": 4.8302368535419385e-05,
37
+ "loss": 0.6283,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.12,
42
+ "learning_rate": 4.7962842242503265e-05,
43
+ "loss": 0.6197,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.14,
48
+ "learning_rate": 4.762331594958714e-05,
49
+ "loss": 0.6029,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.16,
54
+ "learning_rate": 4.728378965667101e-05,
55
+ "loss": 0.5786,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.18,
60
+ "learning_rate": 4.694426336375489e-05,
61
+ "loss": 0.595,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.2,
66
+ "learning_rate": 4.660473707083877e-05,
67
+ "loss": 0.5902,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.22,
72
+ "learning_rate": 4.626521077792265e-05,
73
+ "loss": 0.5888,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.24,
78
+ "learning_rate": 4.592568448500652e-05,
79
+ "loss": 0.5708,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.26,
84
+ "learning_rate": 4.55861581920904e-05,
85
+ "loss": 0.5667,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.29,
90
+ "learning_rate": 4.5246631899174276e-05,
91
+ "loss": 0.5807,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.31,
96
+ "learning_rate": 4.490710560625815e-05,
97
+ "loss": 0.5721,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.33,
102
+ "learning_rate": 4.456757931334203e-05,
103
+ "loss": 0.5549,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.35,
108
+ "learning_rate": 4.4228053020425904e-05,
109
+ "loss": 0.546,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.37,
114
+ "learning_rate": 4.3888526727509784e-05,
115
+ "loss": 0.5376,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.39,
120
+ "learning_rate": 4.354900043459366e-05,
121
+ "loss": 0.5533,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.41,
126
+ "learning_rate": 4.320947414167754e-05,
127
+ "loss": 0.5457,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.43,
132
+ "learning_rate": 4.286994784876141e-05,
133
+ "loss": 0.551,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.45,
138
+ "learning_rate": 4.2530421555845286e-05,
139
+ "loss": 0.545,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.47,
144
+ "learning_rate": 4.219089526292917e-05,
145
+ "loss": 0.539,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.49,
150
+ "learning_rate": 4.185136897001304e-05,
151
+ "loss": 0.5354,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.51,
156
+ "learning_rate": 4.151184267709692e-05,
157
+ "loss": 0.5251,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.53,
162
+ "learning_rate": 4.1172316384180795e-05,
163
+ "loss": 0.5493,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.55,
168
+ "learning_rate": 4.0832790091264675e-05,
169
+ "loss": 0.5293,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.57,
174
+ "learning_rate": 4.049326379834855e-05,
175
+ "loss": 0.5349,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.59,
180
+ "learning_rate": 4.015373750543242e-05,
181
+ "loss": 0.5141,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.61,
186
+ "learning_rate": 3.98142112125163e-05,
187
+ "loss": 0.536,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.63,
192
+ "learning_rate": 3.947468491960018e-05,
193
+ "loss": 0.5209,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.65,
198
+ "learning_rate": 3.913515862668405e-05,
199
+ "loss": 0.5218,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.67,
204
+ "learning_rate": 3.879563233376793e-05,
205
+ "loss": 0.5294,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.69,
210
+ "learning_rate": 3.8456106040851805e-05,
211
+ "loss": 0.5154,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.71,
216
+ "learning_rate": 3.811657974793568e-05,
217
+ "loss": 0.524,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.73,
222
+ "learning_rate": 3.777705345501956e-05,
223
+ "loss": 0.5032,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.75,
228
+ "learning_rate": 3.743752716210343e-05,
229
+ "loss": 0.5083,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.77,
234
+ "learning_rate": 3.7098000869187314e-05,
235
+ "loss": 0.5192,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.79,
240
+ "learning_rate": 3.675847457627119e-05,
241
+ "loss": 0.5035,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.81,
246
+ "learning_rate": 3.641894828335506e-05,
247
+ "loss": 0.4907,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.84,
252
+ "learning_rate": 3.607942199043894e-05,
253
+ "loss": 0.51,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.86,
258
+ "learning_rate": 3.5739895697522816e-05,
259
+ "loss": 0.5032,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.88,
264
+ "learning_rate": 3.540036940460669e-05,
265
+ "loss": 0.5003,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.9,
270
+ "learning_rate": 3.506084311169057e-05,
271
+ "loss": 0.5053,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.92,
276
+ "learning_rate": 3.4721316818774444e-05,
277
+ "loss": 0.5073,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.94,
282
+ "learning_rate": 3.4381790525858324e-05,
283
+ "loss": 0.4997,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.96,
288
+ "learning_rate": 3.40422642329422e-05,
289
+ "loss": 0.4906,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.98,
294
+ "learning_rate": 3.370273794002607e-05,
295
+ "loss": 0.4974,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 1.0,
300
+ "learning_rate": 3.336321164710995e-05,
301
+ "loss": 0.4949,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 1.02,
306
+ "learning_rate": 3.3023685354193826e-05,
307
+ "loss": 0.3855,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 1.04,
312
+ "learning_rate": 3.2684159061277707e-05,
313
+ "loss": 0.3692,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 1.06,
318
+ "learning_rate": 3.234463276836158e-05,
319
+ "loss": 0.3659,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 1.08,
324
+ "learning_rate": 3.200510647544546e-05,
325
+ "loss": 0.3748,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 1.1,
330
+ "learning_rate": 3.1665580182529335e-05,
331
+ "loss": 0.3547,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 1.12,
336
+ "learning_rate": 3.132605388961321e-05,
337
+ "loss": 0.3658,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 1.14,
342
+ "learning_rate": 3.098652759669709e-05,
343
+ "loss": 0.3764,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 1.16,
348
+ "learning_rate": 3.064700130378096e-05,
349
+ "loss": 0.3796,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 1.18,
354
+ "learning_rate": 3.0307475010864843e-05,
355
+ "loss": 0.3742,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 1.2,
360
+ "learning_rate": 2.9967948717948717e-05,
361
+ "loss": 0.3798,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 1.22,
366
+ "learning_rate": 2.9628422425032598e-05,
367
+ "loss": 0.3776,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 1.24,
372
+ "learning_rate": 2.928889613211647e-05,
373
+ "loss": 0.3689,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 1.26,
378
+ "learning_rate": 2.8949369839200345e-05,
379
+ "loss": 0.3627,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 1.28,
384
+ "learning_rate": 2.8609843546284226e-05,
385
+ "loss": 0.3686,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 1.3,
390
+ "learning_rate": 2.82703172533681e-05,
391
+ "loss": 0.3748,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 1.32,
396
+ "learning_rate": 2.793079096045198e-05,
397
+ "loss": 0.3773,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 1.34,
402
+ "learning_rate": 2.7591264667535854e-05,
403
+ "loss": 0.3747,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 1.36,
408
+ "learning_rate": 2.7251738374619734e-05,
409
+ "loss": 0.3846,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 1.39,
414
+ "learning_rate": 2.6912212081703608e-05,
415
+ "loss": 0.3792,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 1.41,
420
+ "learning_rate": 2.6572685788787482e-05,
421
+ "loss": 0.3594,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 1.43,
426
+ "learning_rate": 2.6233159495871362e-05,
427
+ "loss": 0.3869,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 1.45,
432
+ "learning_rate": 2.5893633202955236e-05,
433
+ "loss": 0.3674,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 1.47,
438
+ "learning_rate": 2.5554106910039117e-05,
439
+ "loss": 0.3642,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 1.49,
444
+ "learning_rate": 2.521458061712299e-05,
445
+ "loss": 0.3608,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 1.51,
450
+ "learning_rate": 2.4875054324206867e-05,
451
+ "loss": 0.3739,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 1.53,
456
+ "learning_rate": 2.4535528031290745e-05,
457
+ "loss": 0.3662,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 1.55,
462
+ "learning_rate": 2.4196001738374622e-05,
463
+ "loss": 0.3646,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 1.57,
468
+ "learning_rate": 2.38564754454585e-05,
469
+ "loss": 0.3811,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 1.59,
474
+ "learning_rate": 2.3516949152542376e-05,
475
+ "loss": 0.3568,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 1.61,
480
+ "learning_rate": 2.317742285962625e-05,
481
+ "loss": 0.3663,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 1.63,
486
+ "learning_rate": 2.2837896566710127e-05,
487
+ "loss": 0.3716,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 1.65,
492
+ "learning_rate": 2.2498370273794004e-05,
493
+ "loss": 0.3562,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 1.67,
498
+ "learning_rate": 2.215884398087788e-05,
499
+ "loss": 0.3769,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 1.69,
504
+ "learning_rate": 2.181931768796176e-05,
505
+ "loss": 0.3712,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 1.71,
510
+ "learning_rate": 2.1479791395045636e-05,
511
+ "loss": 0.3691,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 1.73,
516
+ "learning_rate": 2.1140265102129513e-05,
517
+ "loss": 0.369,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 1.75,
522
+ "learning_rate": 2.0800738809213386e-05,
523
+ "loss": 0.3644,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 1.77,
528
+ "learning_rate": 2.0461212516297264e-05,
529
+ "loss": 0.3611,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 1.79,
534
+ "learning_rate": 2.012168622338114e-05,
535
+ "loss": 0.357,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 1.81,
540
+ "learning_rate": 1.9782159930465018e-05,
541
+ "loss": 0.377,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 1.83,
546
+ "learning_rate": 1.944263363754889e-05,
547
+ "loss": 0.3562,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 1.85,
552
+ "learning_rate": 1.910310734463277e-05,
553
+ "loss": 0.3466,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 1.87,
558
+ "learning_rate": 1.8763581051716646e-05,
559
+ "loss": 0.3519,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 1.89,
564
+ "learning_rate": 1.842405475880052e-05,
565
+ "loss": 0.3514,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 1.91,
570
+ "learning_rate": 1.8084528465884397e-05,
571
+ "loss": 0.3498,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 1.94,
576
+ "learning_rate": 1.7745002172968274e-05,
577
+ "loss": 0.3575,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 1.96,
582
+ "learning_rate": 1.740547588005215e-05,
583
+ "loss": 0.3398,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 1.98,
588
+ "learning_rate": 1.706594958713603e-05,
589
+ "loss": 0.3515,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 2.0,
594
+ "learning_rate": 1.6726423294219905e-05,
595
+ "loss": 0.3594,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 2.02,
600
+ "learning_rate": 1.6386897001303783e-05,
601
+ "loss": 0.2482,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 2.04,
606
+ "learning_rate": 1.6047370708387656e-05,
607
+ "loss": 0.2321,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 2.06,
612
+ "learning_rate": 1.5707844415471534e-05,
613
+ "loss": 0.2207,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 2.08,
618
+ "learning_rate": 1.536831812255541e-05,
619
+ "loss": 0.2141,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 2.1,
624
+ "learning_rate": 1.5028791829639288e-05,
625
+ "loss": 0.2253,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 2.12,
630
+ "learning_rate": 1.4689265536723165e-05,
631
+ "loss": 0.2415,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 2.14,
636
+ "learning_rate": 1.4349739243807042e-05,
637
+ "loss": 0.2079,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 2.16,
642
+ "learning_rate": 1.4010212950890916e-05,
643
+ "loss": 0.2156,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 2.18,
648
+ "learning_rate": 1.3670686657974793e-05,
649
+ "loss": 0.216,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 2.2,
654
+ "learning_rate": 1.333116036505867e-05,
655
+ "loss": 0.2245,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 2.22,
660
+ "learning_rate": 1.2991634072142547e-05,
661
+ "loss": 0.2209,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 2.24,
666
+ "learning_rate": 1.2652107779226425e-05,
667
+ "loss": 0.2121,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 2.26,
672
+ "learning_rate": 1.23125814863103e-05,
673
+ "loss": 0.2077,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 2.28,
678
+ "learning_rate": 1.1973055193394177e-05,
679
+ "loss": 0.2181,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 2.3,
684
+ "learning_rate": 1.1633528900478054e-05,
685
+ "loss": 0.221,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 2.32,
690
+ "learning_rate": 1.1294002607561931e-05,
691
+ "loss": 0.2275,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 2.34,
696
+ "learning_rate": 1.0954476314645807e-05,
697
+ "loss": 0.2258,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 2.36,
702
+ "learning_rate": 1.0614950021729684e-05,
703
+ "loss": 0.2246,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 2.38,
708
+ "learning_rate": 1.027542372881356e-05,
709
+ "loss": 0.2235,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 2.4,
714
+ "learning_rate": 9.935897435897435e-06,
715
+ "loss": 0.2123,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 2.42,
720
+ "learning_rate": 9.596371142981312e-06,
721
+ "loss": 0.2015,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 2.44,
726
+ "learning_rate": 9.25684485006519e-06,
727
+ "loss": 0.2204,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 2.46,
732
+ "learning_rate": 8.917318557149066e-06,
733
+ "loss": 0.2243,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 2.49,
738
+ "learning_rate": 8.577792264232942e-06,
739
+ "loss": 0.2134,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 2.51,
744
+ "learning_rate": 8.238265971316819e-06,
745
+ "loss": 0.2121,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 2.53,
750
+ "learning_rate": 7.898739678400696e-06,
751
+ "loss": 0.2147,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 2.55,
756
+ "learning_rate": 7.559213385484572e-06,
757
+ "loss": 0.2279,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 2.57,
762
+ "learning_rate": 7.219687092568449e-06,
763
+ "loss": 0.2051,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 2.59,
768
+ "learning_rate": 6.880160799652326e-06,
769
+ "loss": 0.2262,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 2.61,
774
+ "learning_rate": 6.540634506736203e-06,
775
+ "loss": 0.2125,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 2.63,
780
+ "learning_rate": 6.2011082138200785e-06,
781
+ "loss": 0.2235,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 2.65,
786
+ "learning_rate": 5.861581920903955e-06,
787
+ "loss": 0.2113,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 2.67,
792
+ "learning_rate": 5.522055627987831e-06,
793
+ "loss": 0.2281,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 2.69,
798
+ "learning_rate": 5.182529335071708e-06,
799
+ "loss": 0.203,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 2.71,
804
+ "learning_rate": 4.8430030421555846e-06,
805
+ "loss": 0.2096,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 2.73,
810
+ "learning_rate": 4.503476749239461e-06,
811
+ "loss": 0.2076,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 2.75,
816
+ "learning_rate": 4.163950456323338e-06,
817
+ "loss": 0.209,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 2.77,
822
+ "learning_rate": 3.824424163407214e-06,
823
+ "loss": 0.2124,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 2.79,
828
+ "learning_rate": 3.484897870491091e-06,
829
+ "loss": 0.2091,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 2.81,
834
+ "learning_rate": 3.1453715775749674e-06,
835
+ "loss": 0.2147,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 2.83,
840
+ "learning_rate": 2.805845284658844e-06,
841
+ "loss": 0.2153,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 2.85,
846
+ "learning_rate": 2.466318991742721e-06,
847
+ "loss": 0.2221,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 2.87,
852
+ "learning_rate": 2.126792698826597e-06,
853
+ "loss": 0.225,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 2.89,
858
+ "learning_rate": 1.7872664059104738e-06,
859
+ "loss": 0.2068,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 2.91,
864
+ "learning_rate": 1.4477401129943503e-06,
865
+ "loss": 0.2147,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 2.93,
870
+ "learning_rate": 1.108213820078227e-06,
871
+ "loss": 0.2133,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 2.95,
876
+ "learning_rate": 7.686875271621035e-07,
877
+ "loss": 0.209,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 2.97,
882
+ "learning_rate": 4.2916123424598005e-07,
883
+ "loss": 0.2092,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 2.99,
888
+ "learning_rate": 8.963494132985658e-08,
889
+ "loss": 0.2103,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 3.0,
894
+ "step": 73632,
895
+ "total_flos": 7.74938740264658e+16,
896
+ "train_loss": 0.3792748264984585,
897
+ "train_runtime": 5994.6559,
898
+ "train_samples_per_second": 196.526,
899
+ "train_steps_per_second": 12.283
900
+ }
901
+ ],
902
+ "max_steps": 73632,
903
+ "num_train_epochs": 3,
904
+ "total_flos": 7.74938740264658e+16,
905
+ "trial_name": null,
906
+ "trial_params": null
907
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8061bc1c8862f1af4386e72fe7478d40f46fcfc536525e6b6fb09ddfa71e9fc
3
+ size 2671
vocab.txt ADDED
The diff for this file is too large to render. See raw diff