Hugging Face's logo

ModelTC
/
bert-base-squad2

Question Answering
Transformers
PyTorch
bert
Inference Endpoints
Model card Files Community
1
ZhangYunchenY commited on
Commit
df24e7f
路
1 Parent(s): fdf7dcc

[Model] bert-base-squad2

Browse files
Files changed (11) hide show
  1. config.json +26 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scheduler.pt +3 -0
  6. special_tokens_map.json +1 -0
  7. tokenizer.json +0 -0
  8. tokenizer_config.json +1 -0
  9. trainer_state.json +591 -0
  10. training_args.bin +3 -0
  11. vocab.txt +0 -0
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/mnt/lustre/zhangyunchen/transformers/bert-base-uncased",
3
+ "architectures": [
4
+ "BertForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "gradient_checkpointing": false,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 12,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "torch_dtype": "float32",
22
+ "transformers_version": "4.16.2",
23
+ "type_vocab_size": 2,
24
+ "use_cache": true,
25
+ "vocab_size": 30522
26
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:248ea22bdf5b21a2e3e54150e444541fa6007537c354ecf228a941e9162a0cee
3
+ size 871271163
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:611d6f1bbce79a2115d5116790a78ed439001158d73224a3f48fb7d70a44889f
3
+ size 435661695
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e66a480a22a6740defdd69506fe116fe20206d31dbc7fe1e92b428bb77564b90
3
+ size 14659
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67302e3b96ddf900bdb46c7d677371afe566d2b582154e0b906760b678af5a9d
3
+ size 623
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, "use_fase": true, "special_tokens_map_file": null, "name_or_path": "/mnt/lustre/zhangyunchen/transformers/bert-base-uncased", "tokenizer_class": "BertTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,591 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0358227079538556,
5
+ "global_step": 25000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.12,
12
+ "learning_rate": 2.0234722784297853e-05,
13
+ "loss": 3.0626,
14
+ "step": 1000
15
+ },
16
+ {
17
+ "epoch": 0.12,
18
+ "eval_HasAns_exact": 49.527665317139004,
19
+ "eval_HasAns_f1": 57.18183857248376,
20
+ "eval_HasAns_total": 5928,
21
+ "eval_NoAns_exact": 49.201009251471824,
22
+ "eval_NoAns_f1": 49.201009251471824,
23
+ "eval_NoAns_total": 5945,
24
+ "eval_best_exact": 51.04017518739998,
25
+ "eval_best_exact_thresh": 0.0,
26
+ "eval_best_f1": 53.577488916632994,
27
+ "eval_best_f1_thresh": 0.0,
28
+ "eval_exact": 49.36410342794576,
29
+ "eval_f1": 53.18571035607559,
30
+ "eval_total": 11873,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 0.24,
35
+ "learning_rate": 4.0469445568595707e-05,
36
+ "loss": 1.5868,
37
+ "step": 2000
38
+ },
39
+ {
40
+ "epoch": 0.24,
41
+ "eval_HasAns_exact": 63.41093117408907,
42
+ "eval_HasAns_f1": 71.54881846166067,
43
+ "eval_HasAns_total": 5928,
44
+ "eval_NoAns_exact": 49.991589571068126,
45
+ "eval_NoAns_f1": 49.991589571068126,
46
+ "eval_NoAns_total": 5945,
47
+ "eval_best_exact": 56.73376568685252,
48
+ "eval_best_exact_thresh": 0.0,
49
+ "eval_best_f1": 60.78846086420696,
50
+ "eval_best_f1_thresh": 0.0,
51
+ "eval_exact": 56.69165333108734,
52
+ "eval_f1": 60.75477097959467,
53
+ "eval_total": 11873,
54
+ "step": 2000
55
+ },
56
+ {
57
+ "epoch": 0.36,
58
+ "learning_rate": 4.931660810252171e-05,
59
+ "loss": 1.4132,
60
+ "step": 3000
61
+ },
62
+ {
63
+ "epoch": 0.36,
64
+ "eval_HasAns_exact": 62.7361673414305,
65
+ "eval_HasAns_f1": 70.43865494687375,
66
+ "eval_HasAns_total": 5928,
67
+ "eval_NoAns_exact": 60.26913372582002,
68
+ "eval_NoAns_f1": 60.26913372582002,
69
+ "eval_NoAns_total": 5945,
70
+ "eval_best_exact": 61.509306830624105,
71
+ "eval_best_exact_thresh": 0.0,
72
+ "eval_best_f1": 65.35503634507485,
73
+ "eval_best_f1_thresh": 0.0,
74
+ "eval_exact": 61.500884359471065,
75
+ "eval_f1": 65.34661387392151,
76
+ "eval_total": 11873,
77
+ "step": 3000
78
+ },
79
+ {
80
+ "epoch": 0.49,
81
+ "learning_rate": 4.8024751963621334e-05,
82
+ "loss": 1.2797,
83
+ "step": 4000
84
+ },
85
+ {
86
+ "epoch": 0.49,
87
+ "eval_HasAns_exact": 64.1531713900135,
88
+ "eval_HasAns_f1": 70.62491669164416,
89
+ "eval_HasAns_total": 5928,
90
+ "eval_NoAns_exact": 69.7560975609756,
91
+ "eval_NoAns_f1": 69.7560975609756,
92
+ "eval_NoAns_total": 5945,
93
+ "eval_best_exact": 66.95864566663859,
94
+ "eval_best_exact_thresh": 0.0,
95
+ "eval_best_f1": 70.18988512996461,
96
+ "eval_best_f1_thresh": 0.0,
97
+ "eval_exact": 66.95864566663859,
98
+ "eval_f1": 70.18988512996451,
99
+ "eval_total": 11873,
100
+ "step": 4000
101
+ },
102
+ {
103
+ "epoch": 0.61,
104
+ "learning_rate": 4.673289582472096e-05,
105
+ "loss": 1.2075,
106
+ "step": 5000
107
+ },
108
+ {
109
+ "epoch": 0.61,
110
+ "eval_HasAns_exact": 67.25708502024291,
111
+ "eval_HasAns_f1": 75.0100721049388,
112
+ "eval_HasAns_total": 5928,
113
+ "eval_NoAns_exact": 63.65012615643398,
114
+ "eval_NoAns_f1": 63.65012615643398,
115
+ "eval_NoAns_total": 5945,
116
+ "eval_best_exact": 65.45102333024509,
117
+ "eval_best_exact_thresh": 0.0,
118
+ "eval_best_f1": 69.32196643123734,
119
+ "eval_best_f1_thresh": 0.0,
120
+ "eval_exact": 65.45102333024509,
121
+ "eval_f1": 69.32196643123719,
122
+ "eval_total": 11873,
123
+ "step": 5000
124
+ },
125
+ {
126
+ "epoch": 0.73,
127
+ "learning_rate": 4.544103968582059e-05,
128
+ "loss": 1.125,
129
+ "step": 6000
130
+ },
131
+ {
132
+ "epoch": 0.73,
133
+ "eval_HasAns_exact": 68.47165991902834,
134
+ "eval_HasAns_f1": 75.39561673194235,
135
+ "eval_HasAns_total": 5928,
136
+ "eval_NoAns_exact": 67.48528174936922,
137
+ "eval_NoAns_f1": 67.48528174936922,
138
+ "eval_NoAns_total": 5945,
139
+ "eval_best_exact": 67.98618714730902,
140
+ "eval_best_exact_thresh": 0.0,
141
+ "eval_best_f1": 71.44320862351178,
142
+ "eval_best_f1_thresh": 0.0,
143
+ "eval_exact": 67.97776467615599,
144
+ "eval_f1": 71.43478615235871,
145
+ "eval_total": 11873,
146
+ "step": 6000
147
+ },
148
+ {
149
+ "epoch": 0.85,
150
+ "learning_rate": 4.414918354692022e-05,
151
+ "loss": 1.1063,
152
+ "step": 7000
153
+ },
154
+ {
155
+ "epoch": 0.85,
156
+ "eval_HasAns_exact": 66.07624831309042,
157
+ "eval_HasAns_f1": 72.59462670438361,
158
+ "eval_HasAns_total": 5928,
159
+ "eval_NoAns_exact": 72.95206055508831,
160
+ "eval_NoAns_f1": 72.95206055508831,
161
+ "eval_NoAns_total": 5945,
162
+ "eval_best_exact": 69.51065442600859,
163
+ "eval_best_exact_thresh": 0.0,
164
+ "eval_best_f1": 72.76517704906819,
165
+ "eval_best_f1_thresh": 0.0,
166
+ "eval_exact": 69.51907689716163,
167
+ "eval_f1": 72.77359952022128,
168
+ "eval_total": 11873,
169
+ "step": 7000
170
+ },
171
+ {
172
+ "epoch": 0.97,
173
+ "learning_rate": 4.2857327408019844e-05,
174
+ "loss": 1.0754,
175
+ "step": 8000
176
+ },
177
+ {
178
+ "epoch": 0.97,
179
+ "eval_HasAns_exact": 68.55600539811066,
180
+ "eval_HasAns_f1": 75.38705540287036,
181
+ "eval_HasAns_total": 5928,
182
+ "eval_NoAns_exact": 71.8587047939445,
183
+ "eval_NoAns_f1": 71.8587047939445,
184
+ "eval_NoAns_total": 5945,
185
+ "eval_best_exact": 70.2097195317106,
186
+ "eval_best_exact_thresh": 0.0,
187
+ "eval_best_f1": 73.62035411675363,
188
+ "eval_best_f1_thresh": 0.0,
189
+ "eval_exact": 70.2097195317106,
190
+ "eval_f1": 73.62035411675373,
191
+ "eval_total": 11873,
192
+ "step": 8000
193
+ },
194
+ {
195
+ "epoch": 1.09,
196
+ "learning_rate": 4.156547126911947e-05,
197
+ "loss": 0.8572,
198
+ "step": 9000
199
+ },
200
+ {
201
+ "epoch": 1.09,
202
+ "eval_HasAns_exact": 67.66194331983806,
203
+ "eval_HasAns_f1": 75.07304484460943,
204
+ "eval_HasAns_total": 5928,
205
+ "eval_NoAns_exact": 72.44743481917578,
206
+ "eval_NoAns_f1": 72.44743481917578,
207
+ "eval_NoAns_total": 5945,
208
+ "eval_best_exact": 70.04969257980291,
209
+ "eval_best_exact_thresh": 0.0,
210
+ "eval_best_f1": 73.74993766014028,
211
+ "eval_best_f1_thresh": 0.0,
212
+ "eval_exact": 70.05811505095595,
213
+ "eval_f1": 73.75836013129344,
214
+ "eval_total": 11873,
215
+ "step": 9000
216
+ },
217
+ {
218
+ "epoch": 1.21,
219
+ "learning_rate": 4.02736151302191e-05,
220
+ "loss": 0.7951,
221
+ "step": 10000
222
+ },
223
+ {
224
+ "epoch": 1.21,
225
+ "eval_HasAns_exact": 66.75101214574899,
226
+ "eval_HasAns_f1": 72.62656614039999,
227
+ "eval_HasAns_total": 5928,
228
+ "eval_NoAns_exact": 77.49369217830109,
229
+ "eval_NoAns_f1": 77.49369217830109,
230
+ "eval_NoAns_total": 5945,
231
+ "eval_best_exact": 72.13004295460289,
232
+ "eval_best_exact_thresh": 0.0,
233
+ "eval_best_f1": 75.063613583786,
234
+ "eval_best_f1_thresh": 0.0,
235
+ "eval_exact": 72.13004295460289,
236
+ "eval_f1": 75.0636135837861,
237
+ "eval_total": 11873,
238
+ "step": 10000
239
+ },
240
+ {
241
+ "epoch": 1.34,
242
+ "learning_rate": 3.898175899131873e-05,
243
+ "loss": 0.7932,
244
+ "step": 11000
245
+ },
246
+ {
247
+ "epoch": 1.34,
248
+ "eval_HasAns_exact": 69.29824561403508,
249
+ "eval_HasAns_f1": 75.49290590630278,
250
+ "eval_HasAns_total": 5928,
251
+ "eval_NoAns_exact": 74.68460891505467,
252
+ "eval_NoAns_f1": 74.68460891505467,
253
+ "eval_NoAns_total": 5945,
254
+ "eval_best_exact": 71.9952834161543,
255
+ "eval_best_exact_thresh": 0.0,
256
+ "eval_best_f1": 75.0881787427409,
257
+ "eval_best_f1_thresh": 0.0,
258
+ "eval_exact": 71.9952834161543,
259
+ "eval_f1": 75.08817874274101,
260
+ "eval_total": 11873,
261
+ "step": 11000
262
+ },
263
+ {
264
+ "epoch": 1.46,
265
+ "learning_rate": 3.7689902852418354e-05,
266
+ "loss": 0.7958,
267
+ "step": 12000
268
+ },
269
+ {
270
+ "epoch": 1.46,
271
+ "eval_HasAns_exact": 70.78272604588395,
272
+ "eval_HasAns_f1": 77.6996709947113,
273
+ "eval_HasAns_total": 5928,
274
+ "eval_NoAns_exact": 72.17830109335576,
275
+ "eval_NoAns_f1": 72.17830109335576,
276
+ "eval_NoAns_total": 5945,
277
+ "eval_best_exact": 71.48151267581909,
278
+ "eval_best_exact_thresh": 0.0,
279
+ "eval_best_f1": 74.93503323984241,
280
+ "eval_best_f1_thresh": 0.0,
281
+ "eval_exact": 71.48151267581909,
282
+ "eval_f1": 74.93503323984244,
283
+ "eval_total": 11873,
284
+ "step": 12000
285
+ },
286
+ {
287
+ "epoch": 1.58,
288
+ "learning_rate": 3.639804671351799e-05,
289
+ "loss": 0.7954,
290
+ "step": 13000
291
+ },
292
+ {
293
+ "epoch": 1.58,
294
+ "eval_HasAns_exact": 68.08367071524967,
295
+ "eval_HasAns_f1": 74.95631750408432,
296
+ "eval_HasAns_total": 5928,
297
+ "eval_NoAns_exact": 74.90328006728343,
298
+ "eval_NoAns_f1": 74.90328006728343,
299
+ "eval_NoAns_total": 5945,
300
+ "eval_best_exact": 71.49835761812515,
301
+ "eval_best_exact_thresh": 0.0,
302
+ "eval_best_f1": 74.92976081565001,
303
+ "eval_best_f1_thresh": 0.0,
304
+ "eval_exact": 71.49835761812515,
305
+ "eval_f1": 74.92976081565006,
306
+ "eval_total": 11873,
307
+ "step": 13000
308
+ },
309
+ {
310
+ "epoch": 1.7,
311
+ "learning_rate": 3.5106190574617616e-05,
312
+ "loss": 0.8068,
313
+ "step": 14000
314
+ },
315
+ {
316
+ "epoch": 1.7,
317
+ "eval_HasAns_exact": 69.41632928475033,
318
+ "eval_HasAns_f1": 76.45120101125195,
319
+ "eval_HasAns_total": 5928,
320
+ "eval_NoAns_exact": 73.75946173254836,
321
+ "eval_NoAns_f1": 73.75946173254836,
322
+ "eval_NoAns_total": 5945,
323
+ "eval_best_exact": 71.59100480080856,
324
+ "eval_best_exact_thresh": 0.0,
325
+ "eval_best_f1": 75.10340432870397,
326
+ "eval_best_f1_thresh": 0.0,
327
+ "eval_exact": 71.59100480080856,
328
+ "eval_f1": 75.10340432870404,
329
+ "eval_total": 11873,
330
+ "step": 14000
331
+ },
332
+ {
333
+ "epoch": 1.82,
334
+ "learning_rate": 3.381433443571724e-05,
335
+ "loss": 0.7992,
336
+ "step": 15000
337
+ },
338
+ {
339
+ "epoch": 1.82,
340
+ "eval_HasAns_exact": 70.6646423751687,
341
+ "eval_HasAns_f1": 77.88066795671843,
342
+ "eval_HasAns_total": 5928,
343
+ "eval_NoAns_exact": 72.6829268292683,
344
+ "eval_NoAns_f1": 72.6829268292683,
345
+ "eval_NoAns_total": 5945,
346
+ "eval_best_exact": 71.67522951233892,
347
+ "eval_best_exact_thresh": 0.0,
348
+ "eval_best_f1": 75.27807627789335,
349
+ "eval_best_f1_thresh": 0.0,
350
+ "eval_exact": 71.67522951233892,
351
+ "eval_f1": 75.2780762778934,
352
+ "eval_total": 11873,
353
+ "step": 15000
354
+ },
355
+ {
356
+ "epoch": 1.94,
357
+ "learning_rate": 3.2522478296816864e-05,
358
+ "loss": 0.7675,
359
+ "step": 16000
360
+ },
361
+ {
362
+ "epoch": 1.94,
363
+ "eval_HasAns_exact": 73.29622132253711,
364
+ "eval_HasAns_f1": 80.6750111110637,
365
+ "eval_HasAns_total": 5928,
366
+ "eval_NoAns_exact": 68.29268292682927,
367
+ "eval_NoAns_f1": 68.29268292682927,
368
+ "eval_NoAns_total": 5945,
369
+ "eval_best_exact": 70.79087004127011,
370
+ "eval_best_exact_thresh": 0.0,
371
+ "eval_best_f1": 74.47498238578171,
372
+ "eval_best_f1_thresh": 0.0,
373
+ "eval_exact": 70.79087004127011,
374
+ "eval_f1": 74.47498238578176,
375
+ "eval_total": 11873,
376
+ "step": 16000
377
+ },
378
+ {
379
+ "epoch": 2.06,
380
+ "learning_rate": 3.123062215791649e-05,
381
+ "loss": 0.6262,
382
+ "step": 17000
383
+ },
384
+ {
385
+ "epoch": 2.06,
386
+ "eval_HasAns_exact": 74.07219973009447,
387
+ "eval_HasAns_f1": 81.81573652565241,
388
+ "eval_HasAns_total": 5928,
389
+ "eval_NoAns_exact": 65.36585365853658,
390
+ "eval_NoAns_f1": 65.36585365853658,
391
+ "eval_NoAns_total": 5945,
392
+ "eval_best_exact": 69.71279373368147,
393
+ "eval_best_exact_thresh": 0.0,
394
+ "eval_best_f1": 73.57901845566141,
395
+ "eval_best_f1_thresh": 0.0,
396
+ "eval_exact": 69.71279373368147,
397
+ "eval_f1": 73.57901845566144,
398
+ "eval_total": 11873,
399
+ "step": 17000
400
+ },
401
+ {
402
+ "epoch": 2.19,
403
+ "learning_rate": 2.9938766019016122e-05,
404
+ "loss": 0.4898,
405
+ "step": 18000
406
+ },
407
+ {
408
+ "epoch": 2.19,
409
+ "eval_HasAns_exact": 70.17543859649123,
410
+ "eval_HasAns_f1": 77.1654297749236,
411
+ "eval_HasAns_total": 5928,
412
+ "eval_NoAns_exact": 73.4735071488646,
413
+ "eval_NoAns_f1": 73.4735071488646,
414
+ "eval_NoAns_total": 5945,
415
+ "eval_best_exact": 71.82683399309357,
416
+ "eval_best_exact_thresh": 0.0,
417
+ "eval_best_f1": 75.31682537738969,
418
+ "eval_best_f1_thresh": 0.0,
419
+ "eval_exact": 71.82683399309357,
420
+ "eval_f1": 75.31682537738969,
421
+ "eval_total": 11873,
422
+ "step": 18000
423
+ },
424
+ {
425
+ "epoch": 2.31,
426
+ "learning_rate": 2.8646909880115753e-05,
427
+ "loss": 0.4974,
428
+ "step": 19000
429
+ },
430
+ {
431
+ "epoch": 2.31,
432
+ "eval_HasAns_exact": 71.22132253711202,
433
+ "eval_HasAns_f1": 78.71079807483557,
434
+ "eval_HasAns_total": 5928,
435
+ "eval_NoAns_exact": 71.50546677880573,
436
+ "eval_NoAns_f1": 71.50546677880573,
437
+ "eval_NoAns_total": 5945,
438
+ "eval_best_exact": 71.36359807967658,
439
+ "eval_best_exact_thresh": 0.0,
440
+ "eval_best_f1": 75.10297405774665,
441
+ "eval_best_f1_thresh": 0.0,
442
+ "eval_exact": 71.36359807967658,
443
+ "eval_f1": 75.10297405774665,
444
+ "eval_total": 11873,
445
+ "step": 19000
446
+ },
447
+ {
448
+ "epoch": 2.43,
449
+ "learning_rate": 2.735505374121538e-05,
450
+ "loss": 0.4995,
451
+ "step": 20000
452
+ },
453
+ {
454
+ "epoch": 2.43,
455
+ "eval_HasAns_exact": 71.54183535762483,
456
+ "eval_HasAns_f1": 79.38541316849339,
457
+ "eval_HasAns_total": 5928,
458
+ "eval_NoAns_exact": 69.33557611438184,
459
+ "eval_NoAns_f1": 69.33557611438184,
460
+ "eval_NoAns_total": 5945,
461
+ "eval_best_exact": 70.43712625284259,
462
+ "eval_best_exact_thresh": 0.0,
463
+ "eval_best_f1": 74.3532998621098,
464
+ "eval_best_f1_thresh": 0.0,
465
+ "eval_exact": 70.43712625284259,
466
+ "eval_f1": 74.3532998621098,
467
+ "eval_total": 11873,
468
+ "step": 20000
469
+ },
470
+ {
471
+ "epoch": 2.55,
472
+ "learning_rate": 2.606319760231501e-05,
473
+ "loss": 0.4876,
474
+ "step": 21000
475
+ },
476
+ {
477
+ "epoch": 2.55,
478
+ "eval_HasAns_exact": 73.0600539811066,
479
+ "eval_HasAns_f1": 80.20732398479694,
480
+ "eval_HasAns_total": 5928,
481
+ "eval_NoAns_exact": 69.30193439865432,
482
+ "eval_NoAns_f1": 69.30193439865432,
483
+ "eval_NoAns_total": 5945,
484
+ "eval_best_exact": 71.17830371430978,
485
+ "eval_best_exact_thresh": 0.0,
486
+ "eval_best_f1": 74.7468219137435,
487
+ "eval_best_f1_thresh": 0.0,
488
+ "eval_exact": 71.17830371430978,
489
+ "eval_f1": 74.7468219137435,
490
+ "eval_total": 11873,
491
+ "step": 21000
492
+ },
493
+ {
494
+ "epoch": 2.67,
495
+ "learning_rate": 2.4771341463414636e-05,
496
+ "loss": 0.4949,
497
+ "step": 22000
498
+ },
499
+ {
500
+ "epoch": 2.67,
501
+ "eval_HasAns_exact": 70.95141700404858,
502
+ "eval_HasAns_f1": 77.97020701423389,
503
+ "eval_HasAns_total": 5928,
504
+ "eval_NoAns_exact": 73.62489486963835,
505
+ "eval_NoAns_f1": 73.62489486963835,
506
+ "eval_NoAns_total": 5945,
507
+ "eval_best_exact": 72.29006990651057,
508
+ "eval_best_exact_thresh": 0.0,
509
+ "eval_best_f1": 75.79444008930999,
510
+ "eval_best_f1_thresh": 0.0,
511
+ "eval_exact": 72.29006990651057,
512
+ "eval_f1": 75.79444008930999,
513
+ "eval_total": 11873,
514
+ "step": 22000
515
+ },
516
+ {
517
+ "epoch": 2.79,
518
+ "learning_rate": 2.3479485324514263e-05,
519
+ "loss": 0.492,
520
+ "step": 23000
521
+ },
522
+ {
523
+ "epoch": 2.79,
524
+ "eval_HasAns_exact": 69.98987854251013,
525
+ "eval_HasAns_f1": 76.7639044657501,
526
+ "eval_HasAns_total": 5928,
527
+ "eval_NoAns_exact": 76.56854499579478,
528
+ "eval_NoAns_f1": 76.56854499579478,
529
+ "eval_NoAns_total": 5945,
530
+ "eval_best_exact": 73.28392150256886,
531
+ "eval_best_exact_thresh": 0.0,
532
+ "eval_best_f1": 76.66608487096481,
533
+ "eval_best_f1_thresh": 0.0,
534
+ "eval_exact": 73.28392150256886,
535
+ "eval_f1": 76.66608487096487,
536
+ "eval_total": 11873,
537
+ "step": 23000
538
+ },
539
+ {
540
+ "epoch": 2.91,
541
+ "learning_rate": 2.218762918561389e-05,
542
+ "loss": 0.4969,
543
+ "step": 24000
544
+ },
545
+ {
546
+ "epoch": 2.91,
547
+ "eval_HasAns_exact": 72.36842105263158,
548
+ "eval_HasAns_f1": 79.80006191408404,
549
+ "eval_HasAns_total": 5928,
550
+ "eval_NoAns_exact": 71.62321278385198,
551
+ "eval_NoAns_f1": 71.62321278385198,
552
+ "eval_NoAns_total": 5945,
553
+ "eval_best_exact": 71.9952834161543,
554
+ "eval_best_exact_thresh": 0.0,
555
+ "eval_best_f1": 75.70578346051465,
556
+ "eval_best_f1_thresh": 0.0,
557
+ "eval_exact": 71.9952834161543,
558
+ "eval_f1": 75.70578346051462,
559
+ "eval_total": 11873,
560
+ "step": 24000
561
+ },
562
+ {
563
+ "epoch": 3.04,
564
+ "learning_rate": 2.0895773046713518e-05,
565
+ "loss": 0.4351,
566
+ "step": 25000
567
+ },
568
+ {
569
+ "epoch": 3.04,
570
+ "eval_HasAns_exact": 70.41160593792172,
571
+ "eval_HasAns_f1": 77.89997827092157,
572
+ "eval_HasAns_total": 5928,
573
+ "eval_NoAns_exact": 76.78721614802355,
574
+ "eval_NoAns_f1": 76.78721614802355,
575
+ "eval_NoAns_total": 5945,
576
+ "eval_best_exact": 73.60397540638424,
577
+ "eval_best_exact_thresh": 0.0,
578
+ "eval_best_f1": 77.34280057188766,
579
+ "eval_best_f1_thresh": 0.0,
580
+ "eval_exact": 73.60397540638424,
581
+ "eval_f1": 77.34280057188766,
582
+ "eval_total": 11873,
583
+ "step": 25000
584
+ }
585
+ ],
586
+ "max_steps": 41175,
587
+ "num_train_epochs": 5,
588
+ "total_flos": 7.838549951458406e+16,
589
+ "trial_name": null,
590
+ "trial_params": null
591
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6876c600d758a6201e003081023b4b7311d17bd1887e875d7315cdd39d1a50e3
3
+ size 3119
vocab.txt ADDED
The diff for this file is too large to render. See raw diff