Hugging Face's logo

ModelTC
/
bart-base-squad2

Question Answering
Transformers
PyTorch
bart
Inference Endpoints
Model card Files Community
1
ZhangYunchenY commited on
Commit
7bd44c5
•
1 Parent(s): bf7445c

[Model] bart-base-squad2

Browse files
Files changed (12) hide show
  1. config.json +72 -0
  2. merges.txt +0 -0
  3. optimizer.pt +3 -0
  4. pytorch_model.bin +3 -0
  5. rng_state.pth +3 -0
  6. scheduler.pt +3 -0
  7. special_tokens_map.json +1 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +1 -0
  10. trainer_state.json +775 -0
  11. training_args.bin +3 -0
  12. vocab.json +0 -0
config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/mnt/lustre/zhangyunchen/transformers/bart-base",
3
+ "activation_dropout": 0.1,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BartForQuestionAnswering"
9
+ ],
10
+ "attention_dropout": 0.1,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.1,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 768,
15
+ "decoder_attention_heads": 12,
16
+ "decoder_ffn_dim": 3072,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
+ "early_stopping": true,
22
+ "encoder_attention_heads": 12,
23
+ "encoder_ffn_dim": 3072,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 6,
26
+ "eos_token_id": 2,
27
+ "forced_eos_token_id": 2,
28
+ "gradient_checkpointing": false,
29
+ "id2label": {
30
+ "0": "LABEL_0",
31
+ "1": "LABEL_1"
32
+ },
33
+ "init_std": 0.02,
34
+ "is_encoder_decoder": true,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1
38
+ },
39
+ "max_position_embeddings": 1024,
40
+ "model_type": "bart",
41
+ "no_repeat_ngram_size": 3,
42
+ "normalize_before": false,
43
+ "normalize_embedding": true,
44
+ "num_beams": 4,
45
+ "num_hidden_layers": 6,
46
+ "pad_token_id": 1,
47
+ "scale_embedding": false,
48
+ "task_specific_params": {
49
+ "summarization": {
50
+ "length_penalty": 1.0,
51
+ "max_length": 128,
52
+ "min_length": 12,
53
+ "num_beams": 4
54
+ },
55
+ "summarization_cnn": {
56
+ "length_penalty": 2.0,
57
+ "max_length": 142,
58
+ "min_length": 56,
59
+ "num_beams": 4
60
+ },
61
+ "summarization_xsum": {
62
+ "length_penalty": 1.0,
63
+ "max_length": 62,
64
+ "min_length": 11,
65
+ "num_beams": 6
66
+ }
67
+ },
68
+ "torch_dtype": "float32",
69
+ "transformers_version": "4.16.2",
70
+ "use_cache": true,
71
+ "vocab_size": 50265
72
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ece5333e6bf0883afd48d2c77a43bce7c381048f0a320e3d0eb52d47ff7dea49
3
+ size 1115540499
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abf5fd790f02334655e29425ea5c7b28f3de0b9ab250932a9319b9b0357ecf01
3
+ size 557791521
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c234e96deaad84ae791ebdbf1594e2d6fc6dfa987ac3c2200fe3fddfd4f382f6
3
+ size 14595
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f267d6879c225ff45c28ccd1c46fafe7d0ec95f4a19ed1cccd7e35439eae0c28
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "use_fase": true, "special_tokens_map_file": null, "name_or_path": "/mnt/lustre/zhangyunchen/transformers/bart-base", "tokenizer_class": "BartTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,775 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.005340453938585,
5
+ "global_step": 33000,
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": 1.2135922330097088e-05,
13
+ "loss": 2.9902,
14
+ "step": 1000
15
+ },
16
+ {
17
+ "epoch": 0.12,
18
+ "eval_HasAns_exact": 58.41767881241565,
19
+ "eval_HasAns_f1": 67.0415484792148,
20
+ "eval_HasAns_total": 5928,
21
+ "eval_NoAns_exact": 32.38015138772077,
22
+ "eval_NoAns_f1": 32.38015138772077,
23
+ "eval_NoAns_total": 5945,
24
+ "eval_best_exact": 50.11370336056599,
25
+ "eval_best_exact_thresh": 0.0,
26
+ "eval_best_f1": 51.236780700905136,
27
+ "eval_best_f1_thresh": 0.0,
28
+ "eval_exact": 45.380274572559586,
29
+ "eval_f1": 49.686035491012206,
30
+ "eval_total": 11873,
31
+ "step": 1000
32
+ },
33
+ {
34
+ "epoch": 0.24,
35
+ "learning_rate": 2.4271844660194176e-05,
36
+ "loss": 1.5894,
37
+ "step": 2000
38
+ },
39
+ {
40
+ "epoch": 0.24,
41
+ "eval_HasAns_exact": 69.28137651821862,
42
+ "eval_HasAns_f1": 77.89405016908765,
43
+ "eval_HasAns_total": 5928,
44
+ "eval_NoAns_exact": 46.913372582001685,
45
+ "eval_NoAns_f1": 46.913372582001685,
46
+ "eval_NoAns_total": 5945,
47
+ "eval_best_exact": 58.08978354249137,
48
+ "eval_best_exact_thresh": 0.0,
49
+ "eval_best_f1": 62.389954468319395,
50
+ "eval_best_f1_thresh": 0.0,
51
+ "eval_exact": 58.08136107133833,
52
+ "eval_f1": 62.38153199716619,
53
+ "eval_total": 11873,
54
+ "step": 2000
55
+ },
56
+ {
57
+ "epoch": 0.36,
58
+ "learning_rate": 2.959094078454665e-05,
59
+ "loss": 1.3761,
60
+ "step": 3000
61
+ },
62
+ {
63
+ "epoch": 0.36,
64
+ "eval_HasAns_exact": 71.87921727395411,
65
+ "eval_HasAns_f1": 80.58982724903632,
66
+ "eval_HasAns_total": 5928,
67
+ "eval_NoAns_exact": 51.42136248948697,
68
+ "eval_NoAns_f1": 51.42136248948697,
69
+ "eval_NoAns_total": 5945,
70
+ "eval_best_exact": 61.65248884022572,
71
+ "eval_best_exact_thresh": 0.0,
72
+ "eval_best_f1": 66.00155781456188,
73
+ "eval_best_f1_thresh": 0.0,
74
+ "eval_exact": 61.63564389791965,
75
+ "eval_f1": 65.98471287225554,
76
+ "eval_total": 11873,
77
+ "step": 3000
78
+ },
79
+ {
80
+ "epoch": 0.49,
81
+ "learning_rate": 2.8816207421945613e-05,
82
+ "loss": 1.249,
83
+ "step": 4000
84
+ },
85
+ {
86
+ "epoch": 0.49,
87
+ "eval_HasAns_exact": 68.47165991902834,
88
+ "eval_HasAns_f1": 75.13575702386423,
89
+ "eval_HasAns_total": 5928,
90
+ "eval_NoAns_exact": 74.3313708999159,
91
+ "eval_NoAns_f1": 74.3313708999159,
92
+ "eval_NoAns_total": 5945,
93
+ "eval_best_exact": 71.40571043544176,
94
+ "eval_best_exact_thresh": 0.0,
95
+ "eval_best_f1": 74.7329880937815,
96
+ "eval_best_f1_thresh": 0.0,
97
+ "eval_exact": 71.40571043544176,
98
+ "eval_f1": 74.73298809378153,
99
+ "eval_total": 11873,
100
+ "step": 4000
101
+ },
102
+ {
103
+ "epoch": 0.61,
104
+ "learning_rate": 2.8041474059344576e-05,
105
+ "loss": 1.1796,
106
+ "step": 5000
107
+ },
108
+ {
109
+ "epoch": 0.61,
110
+ "eval_HasAns_exact": 67.62820512820512,
111
+ "eval_HasAns_f1": 73.71252998774504,
112
+ "eval_HasAns_total": 5928,
113
+ "eval_NoAns_exact": 78.26745164003364,
114
+ "eval_NoAns_f1": 78.26745164003364,
115
+ "eval_NoAns_total": 5945,
116
+ "eval_best_exact": 72.95544512760044,
117
+ "eval_best_exact_thresh": 0.0,
118
+ "eval_best_f1": 75.99325172806816,
119
+ "eval_best_f1_thresh": 0.0,
120
+ "eval_exact": 72.95544512760044,
121
+ "eval_f1": 75.99325172806819,
122
+ "eval_total": 11873,
123
+ "step": 5000
124
+ },
125
+ {
126
+ "epoch": 0.73,
127
+ "learning_rate": 2.726674069674354e-05,
128
+ "loss": 1.1097,
129
+ "step": 6000
130
+ },
131
+ {
132
+ "epoch": 0.73,
133
+ "eval_HasAns_exact": 73.80229419703105,
134
+ "eval_HasAns_f1": 80.14574039018072,
135
+ "eval_HasAns_total": 5928,
136
+ "eval_NoAns_exact": 70.21026072329688,
137
+ "eval_NoAns_f1": 70.21026072329688,
138
+ "eval_NoAns_total": 5945,
139
+ "eval_best_exact": 72.01212835846037,
140
+ "eval_best_exact_thresh": 0.0,
141
+ "eval_best_f1": 75.17931011816671,
142
+ "eval_best_f1_thresh": 0.0,
143
+ "eval_exact": 72.00370588730733,
144
+ "eval_f1": 75.17088764701359,
145
+ "eval_total": 11873,
146
+ "step": 6000
147
+ },
148
+ {
149
+ "epoch": 0.85,
150
+ "learning_rate": 2.64920073341425e-05,
151
+ "loss": 1.0804,
152
+ "step": 7000
153
+ },
154
+ {
155
+ "epoch": 0.85,
156
+ "eval_HasAns_exact": 75.60728744939271,
157
+ "eval_HasAns_f1": 82.66989971492067,
158
+ "eval_HasAns_total": 5928,
159
+ "eval_NoAns_exact": 68.56181665264928,
160
+ "eval_NoAns_f1": 68.56181665264928,
161
+ "eval_NoAns_total": 5945,
162
+ "eval_best_exact": 72.0879305988377,
163
+ "eval_best_exact_thresh": 0.0,
164
+ "eval_best_f1": 75.61418053651563,
165
+ "eval_best_f1_thresh": 0.0,
166
+ "eval_exact": 72.07950812768466,
167
+ "eval_f1": 75.60575806536264,
168
+ "eval_total": 11873,
169
+ "step": 7000
170
+ },
171
+ {
172
+ "epoch": 0.97,
173
+ "learning_rate": 2.5717273971541462e-05,
174
+ "loss": 1.0272,
175
+ "step": 8000
176
+ },
177
+ {
178
+ "epoch": 0.97,
179
+ "eval_HasAns_exact": 72.85762483130904,
180
+ "eval_HasAns_f1": 79.23824809473913,
181
+ "eval_HasAns_total": 5928,
182
+ "eval_NoAns_exact": 76.01345668629101,
183
+ "eval_NoAns_f1": 76.01345668629101,
184
+ "eval_NoAns_total": 5945,
185
+ "eval_best_exact": 74.43780005053483,
186
+ "eval_best_exact_thresh": 0.0,
187
+ "eval_best_f1": 77.62354372994311,
188
+ "eval_best_f1_thresh": 0.0,
189
+ "eval_exact": 74.43780005053483,
190
+ "eval_f1": 77.62354372994308,
191
+ "eval_total": 11873,
192
+ "step": 8000
193
+ },
194
+ {
195
+ "epoch": 1.09,
196
+ "learning_rate": 2.4942540608940426e-05,
197
+ "loss": 0.8905,
198
+ "step": 9000
199
+ },
200
+ {
201
+ "epoch": 1.09,
202
+ "eval_HasAns_exact": 76.417004048583,
203
+ "eval_HasAns_f1": 83.82699462168138,
204
+ "eval_HasAns_total": 5928,
205
+ "eval_NoAns_exact": 66.27417998317914,
206
+ "eval_NoAns_f1": 66.27417998317914,
207
+ "eval_NoAns_total": 5945,
208
+ "eval_best_exact": 71.3467531373705,
209
+ "eval_best_exact_thresh": 0.0,
210
+ "eval_best_f1": 75.04644353721282,
211
+ "eval_best_f1_thresh": 0.0,
212
+ "eval_exact": 71.33833066621747,
213
+ "eval_f1": 75.03802106605983,
214
+ "eval_total": 11873,
215
+ "step": 9000
216
+ },
217
+ {
218
+ "epoch": 1.21,
219
+ "learning_rate": 2.4167807246339385e-05,
220
+ "loss": 0.8348,
221
+ "step": 10000
222
+ },
223
+ {
224
+ "epoch": 1.21,
225
+ "eval_HasAns_exact": 74.00472334682861,
226
+ "eval_HasAns_f1": 80.47538325931156,
227
+ "eval_HasAns_total": 5928,
228
+ "eval_NoAns_exact": 75.55929352396973,
229
+ "eval_NoAns_f1": 75.55929352396973,
230
+ "eval_NoAns_total": 5945,
231
+ "eval_best_exact": 74.78312136780931,
232
+ "eval_best_exact_thresh": 0.0,
233
+ "eval_best_f1": 78.01381891360221,
234
+ "eval_best_f1_thresh": 0.0,
235
+ "eval_exact": 74.78312136780931,
236
+ "eval_f1": 78.01381891360226,
237
+ "eval_total": 11873,
238
+ "step": 10000
239
+ },
240
+ {
241
+ "epoch": 1.34,
242
+ "learning_rate": 2.3393073883738345e-05,
243
+ "loss": 0.8345,
244
+ "step": 11000
245
+ },
246
+ {
247
+ "epoch": 1.34,
248
+ "eval_HasAns_exact": 73.21187584345479,
249
+ "eval_HasAns_f1": 79.61538989647931,
250
+ "eval_HasAns_total": 5928,
251
+ "eval_NoAns_exact": 77.91421362489487,
252
+ "eval_NoAns_f1": 77.91421362489487,
253
+ "eval_NoAns_total": 5945,
254
+ "eval_best_exact": 75.56641118504169,
255
+ "eval_best_exact_thresh": 0.0,
256
+ "eval_best_f1": 78.76358387150074,
257
+ "eval_best_f1_thresh": 0.0,
258
+ "eval_exact": 75.56641118504169,
259
+ "eval_f1": 78.76358387150083,
260
+ "eval_total": 11873,
261
+ "step": 11000
262
+ },
263
+ {
264
+ "epoch": 1.46,
265
+ "learning_rate": 2.2618340521137308e-05,
266
+ "loss": 0.8372,
267
+ "step": 12000
268
+ },
269
+ {
270
+ "epoch": 1.46,
271
+ "eval_HasAns_exact": 74.12280701754386,
272
+ "eval_HasAns_f1": 80.2215790589714,
273
+ "eval_HasAns_total": 5928,
274
+ "eval_NoAns_exact": 77.6955424726661,
275
+ "eval_NoAns_f1": 77.6955424726661,
276
+ "eval_NoAns_total": 5945,
277
+ "eval_best_exact": 75.91173250231618,
278
+ "eval_best_exact_thresh": 0.0,
279
+ "eval_best_f1": 78.95675235084504,
280
+ "eval_best_f1_thresh": 0.0,
281
+ "eval_exact": 75.91173250231618,
282
+ "eval_f1": 78.95675235084498,
283
+ "eval_total": 11873,
284
+ "step": 12000
285
+ },
286
+ {
287
+ "epoch": 1.58,
288
+ "learning_rate": 2.184360715853627e-05,
289
+ "loss": 0.8396,
290
+ "step": 13000
291
+ },
292
+ {
293
+ "epoch": 1.58,
294
+ "eval_HasAns_exact": 77.74966261808368,
295
+ "eval_HasAns_f1": 84.34526051529392,
296
+ "eval_HasAns_total": 5928,
297
+ "eval_NoAns_exact": 70.68124474348191,
298
+ "eval_NoAns_f1": 70.68124474348191,
299
+ "eval_NoAns_total": 5945,
300
+ "eval_best_exact": 74.21881580055589,
301
+ "eval_best_exact_thresh": 0.0,
302
+ "eval_best_f1": 77.51189289435379,
303
+ "eval_best_f1_thresh": 0.0,
304
+ "eval_exact": 74.21039332940285,
305
+ "eval_f1": 77.50347042320072,
306
+ "eval_total": 11873,
307
+ "step": 13000
308
+ },
309
+ {
310
+ "epoch": 1.7,
311
+ "learning_rate": 2.1068873795935234e-05,
312
+ "loss": 0.8164,
313
+ "step": 14000
314
+ },
315
+ {
316
+ "epoch": 1.7,
317
+ "eval_HasAns_exact": 75.65789473684211,
318
+ "eval_HasAns_f1": 82.25660585284308,
319
+ "eval_HasAns_total": 5928,
320
+ "eval_NoAns_exact": 73.59125315391086,
321
+ "eval_NoAns_f1": 73.59125315391086,
322
+ "eval_NoAns_total": 5945,
323
+ "eval_best_exact": 74.62309441590162,
324
+ "eval_best_exact_thresh": 0.0,
325
+ "eval_best_f1": 77.9177258903103,
326
+ "eval_best_f1_thresh": 0.0,
327
+ "eval_exact": 74.62309441590162,
328
+ "eval_f1": 77.91772589031022,
329
+ "eval_total": 11873,
330
+ "step": 14000
331
+ },
332
+ {
333
+ "epoch": 1.82,
334
+ "learning_rate": 2.0294140433334194e-05,
335
+ "loss": 0.8052,
336
+ "step": 15000
337
+ },
338
+ {
339
+ "epoch": 1.82,
340
+ "eval_HasAns_exact": 78.71120107962213,
341
+ "eval_HasAns_f1": 85.01212257884671,
342
+ "eval_HasAns_total": 5928,
343
+ "eval_NoAns_exact": 68.78048780487805,
344
+ "eval_NoAns_f1": 68.78048780487805,
345
+ "eval_NoAns_total": 5945,
346
+ "eval_best_exact": 73.74715741598585,
347
+ "eval_best_exact_thresh": 0.0,
348
+ "eval_best_f1": 76.89310727258507,
349
+ "eval_best_f1_thresh": 0.0,
350
+ "eval_exact": 73.73873494483281,
351
+ "eval_f1": 76.88468480143192,
352
+ "eval_total": 11873,
353
+ "step": 15000
354
+ },
355
+ {
356
+ "epoch": 1.94,
357
+ "learning_rate": 1.9519407070733157e-05,
358
+ "loss": 0.8066,
359
+ "step": 16000
360
+ },
361
+ {
362
+ "epoch": 1.94,
363
+ "eval_HasAns_exact": 74.69635627530364,
364
+ "eval_HasAns_f1": 80.71107846446266,
365
+ "eval_HasAns_total": 5928,
366
+ "eval_NoAns_exact": 77.89739276703112,
367
+ "eval_NoAns_f1": 77.89739276703112,
368
+ "eval_NoAns_total": 5945,
369
+ "eval_best_exact": 76.29916617535585,
370
+ "eval_best_exact_thresh": 0.0,
371
+ "eval_best_f1": 79.30222126988423,
372
+ "eval_best_f1_thresh": 0.0,
373
+ "eval_exact": 76.29916617535585,
374
+ "eval_f1": 79.30222126988414,
375
+ "eval_total": 11873,
376
+ "step": 16000
377
+ },
378
+ {
379
+ "epoch": 2.06,
380
+ "learning_rate": 1.874467370813212e-05,
381
+ "loss": 0.7057,
382
+ "step": 17000
383
+ },
384
+ {
385
+ "epoch": 2.06,
386
+ "eval_HasAns_exact": 77.47975708502024,
387
+ "eval_HasAns_f1": 84.97990004810745,
388
+ "eval_HasAns_total": 5928,
389
+ "eval_NoAns_exact": 68.71320437342304,
390
+ "eval_NoAns_f1": 68.71320437342304,
391
+ "eval_NoAns_total": 5945,
392
+ "eval_best_exact": 73.09020466604902,
393
+ "eval_best_exact_thresh": 0.0,
394
+ "eval_best_f1": 76.8349067198839,
395
+ "eval_best_f1_thresh": 0.0,
396
+ "eval_exact": 73.09020466604902,
397
+ "eval_f1": 76.83490671988388,
398
+ "eval_total": 11873,
399
+ "step": 17000
400
+ },
401
+ {
402
+ "epoch": 2.18,
403
+ "learning_rate": 1.796994034553108e-05,
404
+ "loss": 0.6376,
405
+ "step": 18000
406
+ },
407
+ {
408
+ "epoch": 2.18,
409
+ "eval_HasAns_exact": 74.25775978407557,
410
+ "eval_HasAns_f1": 81.15876336598458,
411
+ "eval_HasAns_total": 5928,
412
+ "eval_NoAns_exact": 77.37594617325483,
413
+ "eval_NoAns_f1": 77.37594617325483,
414
+ "eval_NoAns_total": 5945,
415
+ "eval_best_exact": 75.81908531963278,
416
+ "eval_best_exact_thresh": 0.0,
417
+ "eval_best_f1": 79.26464661278166,
418
+ "eval_best_f1_thresh": 0.0,
419
+ "eval_exact": 75.81908531963278,
420
+ "eval_f1": 79.26464661278159,
421
+ "eval_total": 11873,
422
+ "step": 18000
423
+ },
424
+ {
425
+ "epoch": 2.31,
426
+ "learning_rate": 1.719520698293004e-05,
427
+ "loss": 0.6315,
428
+ "step": 19000
429
+ },
430
+ {
431
+ "epoch": 2.31,
432
+ "eval_HasAns_exact": 73.3468286099865,
433
+ "eval_HasAns_f1": 79.65372011100962,
434
+ "eval_HasAns_total": 5928,
435
+ "eval_NoAns_exact": 80.43734230445753,
436
+ "eval_NoAns_f1": 80.43734230445753,
437
+ "eval_NoAns_total": 5945,
438
+ "eval_best_exact": 76.89716162722142,
439
+ "eval_best_exact_thresh": 0.0,
440
+ "eval_best_f1": 80.04609221073564,
441
+ "eval_best_f1_thresh": 0.0,
442
+ "eval_exact": 76.89716162722142,
443
+ "eval_f1": 80.0460922107357,
444
+ "eval_total": 11873,
445
+ "step": 19000
446
+ },
447
+ {
448
+ "epoch": 2.43,
449
+ "learning_rate": 1.6420473620329003e-05,
450
+ "loss": 0.6323,
451
+ "step": 20000
452
+ },
453
+ {
454
+ "epoch": 2.43,
455
+ "eval_HasAns_exact": 75.21929824561404,
456
+ "eval_HasAns_f1": 81.79839130722176,
457
+ "eval_HasAns_total": 5928,
458
+ "eval_NoAns_exact": 76.85449957947856,
459
+ "eval_NoAns_f1": 76.85449957947856,
460
+ "eval_NoAns_total": 5945,
461
+ "eval_best_exact": 76.03806956961172,
462
+ "eval_best_exact_thresh": 0.0,
463
+ "eval_best_f1": 79.32290606158605,
464
+ "eval_best_f1_thresh": 0.0,
465
+ "eval_exact": 76.03806956961172,
466
+ "eval_f1": 79.322906061586,
467
+ "eval_total": 11873,
468
+ "step": 20000
469
+ },
470
+ {
471
+ "epoch": 2.55,
472
+ "learning_rate": 1.5645740257727966e-05,
473
+ "loss": 0.6125,
474
+ "step": 21000
475
+ },
476
+ {
477
+ "epoch": 2.55,
478
+ "eval_HasAns_exact": 75.82658569500674,
479
+ "eval_HasAns_f1": 82.54925380201965,
480
+ "eval_HasAns_total": 5928,
481
+ "eval_NoAns_exact": 76.36669470142978,
482
+ "eval_NoAns_f1": 76.36669470142978,
483
+ "eval_NoAns_total": 5945,
484
+ "eval_best_exact": 76.09702686768298,
485
+ "eval_best_exact_thresh": 0.0,
486
+ "eval_best_f1": 79.45354809554222,
487
+ "eval_best_f1_thresh": 0.0,
488
+ "eval_exact": 76.09702686768298,
489
+ "eval_f1": 79.45354809554216,
490
+ "eval_total": 11873,
491
+ "step": 21000
492
+ },
493
+ {
494
+ "epoch": 2.67,
495
+ "learning_rate": 1.4871006895126928e-05,
496
+ "loss": 0.6095,
497
+ "step": 22000
498
+ },
499
+ {
500
+ "epoch": 2.67,
501
+ "eval_HasAns_exact": 77.36167341430499,
502
+ "eval_HasAns_f1": 84.06251850374095,
503
+ "eval_HasAns_total": 5928,
504
+ "eval_NoAns_exact": 73.92767031118586,
505
+ "eval_NoAns_f1": 73.92767031118586,
506
+ "eval_NoAns_total": 5945,
507
+ "eval_best_exact": 75.65063589657206,
508
+ "eval_best_exact_thresh": 0.0,
509
+ "eval_best_f1": 78.99626123896034,
510
+ "eval_best_f1_thresh": 0.0,
511
+ "eval_exact": 75.64221342541902,
512
+ "eval_f1": 78.98783876780718,
513
+ "eval_total": 11873,
514
+ "step": 22000
515
+ },
516
+ {
517
+ "epoch": 2.79,
518
+ "learning_rate": 1.4096273532525889e-05,
519
+ "loss": 0.5988,
520
+ "step": 23000
521
+ },
522
+ {
523
+ "epoch": 2.79,
524
+ "eval_HasAns_exact": 77.17611336032388,
525
+ "eval_HasAns_f1": 83.48725009021588,
526
+ "eval_HasAns_total": 5928,
527
+ "eval_NoAns_exact": 74.95374264087468,
528
+ "eval_NoAns_f1": 74.95374264087468,
529
+ "eval_NoAns_total": 5945,
530
+ "eval_best_exact": 76.07175945422387,
531
+ "eval_best_exact_thresh": 0.0,
532
+ "eval_best_f1": 79.22280961297066,
533
+ "eval_best_f1_thresh": 0.0,
534
+ "eval_exact": 76.06333698307083,
535
+ "eval_f1": 79.21438714181748,
536
+ "eval_total": 11873,
537
+ "step": 23000
538
+ },
539
+ {
540
+ "epoch": 2.91,
541
+ "learning_rate": 1.3321540169924852e-05,
542
+ "loss": 0.5968,
543
+ "step": 24000
544
+ },
545
+ {
546
+ "epoch": 2.91,
547
+ "eval_HasAns_exact": 74.13967611336032,
548
+ "eval_HasAns_f1": 80.3794525038334,
549
+ "eval_HasAns_total": 5928,
550
+ "eval_NoAns_exact": 81.22792262405383,
551
+ "eval_NoAns_f1": 81.22792262405383,
552
+ "eval_NoAns_total": 5945,
553
+ "eval_best_exact": 77.68887391560683,
554
+ "eval_best_exact_thresh": 0.0,
555
+ "eval_best_f1": 80.80429499222815,
556
+ "eval_best_f1_thresh": 0.0,
557
+ "eval_exact": 77.68887391560683,
558
+ "eval_f1": 80.80429499222812,
559
+ "eval_total": 11873,
560
+ "step": 24000
561
+ },
562
+ {
563
+ "epoch": 3.03,
564
+ "learning_rate": 1.2546806807323812e-05,
565
+ "loss": 0.5722,
566
+ "step": 25000
567
+ },
568
+ {
569
+ "epoch": 3.03,
570
+ "eval_HasAns_exact": 75.2867746288799,
571
+ "eval_HasAns_f1": 81.508728103177,
572
+ "eval_HasAns_total": 5928,
573
+ "eval_NoAns_exact": 77.30866274179984,
574
+ "eval_NoAns_f1": 77.30866274179984,
575
+ "eval_NoAns_total": 5945,
576
+ "eval_best_exact": 76.29916617535585,
577
+ "eval_best_exact_thresh": 0.0,
578
+ "eval_best_f1": 79.40568855349397,
579
+ "eval_best_f1_thresh": 0.0,
580
+ "eval_exact": 76.29916617535585,
581
+ "eval_f1": 79.4056885534939,
582
+ "eval_total": 11873,
583
+ "step": 25000
584
+ },
585
+ {
586
+ "epoch": 3.16,
587
+ "learning_rate": 1.1772073444722775e-05,
588
+ "loss": 0.4844,
589
+ "step": 26000
590
+ },
591
+ {
592
+ "epoch": 3.16,
593
+ "eval_HasAns_exact": 76.78812415654521,
594
+ "eval_HasAns_f1": 84.07498836042032,
595
+ "eval_HasAns_total": 5928,
596
+ "eval_NoAns_exact": 73.10344827586206,
597
+ "eval_NoAns_f1": 73.10344827586206,
598
+ "eval_NoAns_total": 5945,
599
+ "eval_best_exact": 74.95157079087004,
600
+ "eval_best_exact_thresh": 0.0,
601
+ "eval_best_f1": 78.58978615350551,
602
+ "eval_best_f1_thresh": 0.0,
603
+ "eval_exact": 74.94314831971701,
604
+ "eval_f1": 78.58136368235249,
605
+ "eval_total": 11873,
606
+ "step": 26000
607
+ },
608
+ {
609
+ "epoch": 3.28,
610
+ "learning_rate": 1.0997340082121736e-05,
611
+ "loss": 0.4664,
612
+ "step": 27000
613
+ },
614
+ {
615
+ "epoch": 3.28,
616
+ "eval_HasAns_exact": 76.40013495276654,
617
+ "eval_HasAns_f1": 83.06591024545988,
618
+ "eval_HasAns_total": 5928,
619
+ "eval_NoAns_exact": 78.04878048780488,
620
+ "eval_NoAns_f1": 78.04878048780488,
621
+ "eval_NoAns_total": 5945,
622
+ "eval_best_exact": 77.23406047334288,
623
+ "eval_best_exact_thresh": 0.0,
624
+ "eval_best_f1": 80.56217602417964,
625
+ "eval_best_f1_thresh": 0.0,
626
+ "eval_exact": 77.22563800218984,
627
+ "eval_f1": 80.55375355302654,
628
+ "eval_total": 11873,
629
+ "step": 27000
630
+ },
631
+ {
632
+ "epoch": 3.4,
633
+ "learning_rate": 1.02226067195207e-05,
634
+ "loss": 0.4914,
635
+ "step": 28000
636
+ },
637
+ {
638
+ "epoch": 3.4,
639
+ "eval_HasAns_exact": 76.38326585695006,
640
+ "eval_HasAns_f1": 83.19767154572087,
641
+ "eval_HasAns_total": 5928,
642
+ "eval_NoAns_exact": 76.01345668629101,
643
+ "eval_NoAns_f1": 76.01345668629101,
644
+ "eval_NoAns_total": 5945,
645
+ "eval_best_exact": 76.20651899267246,
646
+ "eval_best_exact_thresh": 0.0,
647
+ "eval_best_f1": 79.60884333555407,
648
+ "eval_best_f1_thresh": 0.0,
649
+ "eval_exact": 76.19809652151942,
650
+ "eval_f1": 79.60042086440092,
651
+ "eval_total": 11873,
652
+ "step": 28000
653
+ },
654
+ {
655
+ "epoch": 3.52,
656
+ "learning_rate": 9.44787335691966e-06,
657
+ "loss": 0.4696,
658
+ "step": 29000
659
+ },
660
+ {
661
+ "epoch": 3.52,
662
+ "eval_HasAns_exact": 74.91565452091768,
663
+ "eval_HasAns_f1": 81.5931788001518,
664
+ "eval_HasAns_total": 5928,
665
+ "eval_NoAns_exact": 79.37762825904122,
666
+ "eval_NoAns_f1": 79.37762825904122,
667
+ "eval_NoAns_total": 5945,
668
+ "eval_best_exact": 77.14983576181251,
669
+ "eval_best_exact_thresh": 0.0,
670
+ "eval_best_f1": 80.48381739470226,
671
+ "eval_best_f1_thresh": 0.0,
672
+ "eval_exact": 77.14983576181251,
673
+ "eval_f1": 80.48381739470223,
674
+ "eval_total": 11873,
675
+ "step": 29000
676
+ },
677
+ {
678
+ "epoch": 3.64,
679
+ "learning_rate": 8.673139994318622e-06,
680
+ "loss": 0.483,
681
+ "step": 30000
682
+ },
683
+ {
684
+ "epoch": 3.64,
685
+ "eval_HasAns_exact": 76.6025641025641,
686
+ "eval_HasAns_f1": 83.549417980174,
687
+ "eval_HasAns_total": 5928,
688
+ "eval_NoAns_exact": 76.13120269133726,
689
+ "eval_NoAns_f1": 76.13120269133726,
690
+ "eval_NoAns_total": 5945,
691
+ "eval_best_exact": 76.36654594458014,
692
+ "eval_best_exact_thresh": 0.0,
693
+ "eval_best_f1": 79.83499956089204,
694
+ "eval_best_f1_thresh": 0.0,
695
+ "eval_exact": 76.36654594458014,
696
+ "eval_f1": 79.83499956089204,
697
+ "eval_total": 11873,
698
+ "step": 30000
699
+ },
700
+ {
701
+ "epoch": 3.76,
702
+ "learning_rate": 7.898406631717584e-06,
703
+ "loss": 0.4602,
704
+ "step": 31000
705
+ },
706
+ {
707
+ "epoch": 3.76,
708
+ "eval_HasAns_exact": 73.61673414304994,
709
+ "eval_HasAns_f1": 79.59590506841275,
710
+ "eval_HasAns_total": 5928,
711
+ "eval_NoAns_exact": 81.69890664423886,
712
+ "eval_NoAns_f1": 81.69890664423886,
713
+ "eval_NoAns_total": 5945,
714
+ "eval_best_exact": 77.66360650214773,
715
+ "eval_best_exact_thresh": 0.0,
716
+ "eval_best_f1": 80.64891141628488,
717
+ "eval_best_f1_thresh": 0.0,
718
+ "eval_exact": 77.66360650214773,
719
+ "eval_f1": 80.64891141628489,
720
+ "eval_total": 11873,
721
+ "step": 31000
722
+ },
723
+ {
724
+ "epoch": 3.88,
725
+ "learning_rate": 7.123673269116546e-06,
726
+ "loss": 0.4706,
727
+ "step": 32000
728
+ },
729
+ {
730
+ "epoch": 3.88,
731
+ "eval_HasAns_exact": 76.1302294197031,
732
+ "eval_HasAns_f1": 83.02595617673859,
733
+ "eval_HasAns_total": 5928,
734
+ "eval_NoAns_exact": 77.14045416316232,
735
+ "eval_NoAns_f1": 77.14045416316232,
736
+ "eval_NoAns_total": 5945,
737
+ "eval_best_exact": 76.63606502147731,
738
+ "eval_best_exact_thresh": 0.0,
739
+ "eval_best_f1": 80.07899167992126,
740
+ "eval_best_f1_thresh": 0.0,
741
+ "eval_exact": 76.63606502147731,
742
+ "eval_f1": 80.07899167992143,
743
+ "eval_total": 11873,
744
+ "step": 32000
745
+ },
746
+ {
747
+ "epoch": 4.01,
748
+ "learning_rate": 6.348939906515508e-06,
749
+ "loss": 0.4633,
750
+ "step": 33000
751
+ },
752
+ {
753
+ "epoch": 4.01,
754
+ "eval_HasAns_exact": 75.38798920377867,
755
+ "eval_HasAns_f1": 82.20772908314846,
756
+ "eval_HasAns_total": 5928,
757
+ "eval_NoAns_exact": 79.42809083263246,
758
+ "eval_NoAns_f1": 79.42809083263246,
759
+ "eval_NoAns_total": 5945,
760
+ "eval_best_exact": 77.41935483870968,
761
+ "eval_best_exact_thresh": 0.0,
762
+ "eval_best_f1": 80.82434245809002,
763
+ "eval_best_f1_thresh": 0.0,
764
+ "eval_exact": 77.41093236755664,
765
+ "eval_f1": 80.81591998693705,
766
+ "eval_total": 11873,
767
+ "step": 33000
768
+ }
769
+ ],
770
+ "max_steps": 41195,
771
+ "num_train_epochs": 5,
772
+ "total_flos": 1.2072877510811443e+17,
773
+ "trial_name": null,
774
+ "trial_params": null
775
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:205cb8a71d5356a56600b223b3f6517a8a18f228d32259990e1cda7ec3357fcf
3
+ size 3055
vocab.json ADDED
The diff for this file is too large to render. See raw diff