Mizuiro-sakura commited on
Commit
753e139
1 Parent(s): 9ccce65

Upload 11 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ent2>": 32771,
3
+ "<ent>": 32770
4
+ }
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "studio-ousia/luke-japanese-base-lite",
3
+ "architectures": [
4
+ "LukeForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bert_model_name": "models/luke-japanese/hf_xlm_roberta",
8
+ "bos_token_id": 0,
9
+ "classifier_dropout": null,
10
+ "cls_entity_prediction": false,
11
+ "entity_emb_size": 256,
12
+ "entity_vocab_size": 4,
13
+ "eos_token_id": 2,
14
+ "hidden_act": "gelu",
15
+ "hidden_dropout_prob": 0.1,
16
+ "hidden_size": 768,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 3072,
19
+ "layer_norm_eps": 1e-05,
20
+ "max_position_embeddings": 514,
21
+ "model_type": "luke",
22
+ "num_attention_heads": 12,
23
+ "num_hidden_layers": 12,
24
+ "pad_token_id": 1,
25
+ "position_embedding_type": "absolute",
26
+ "torch_dtype": "float32",
27
+ "transformers_version": "4.26.0",
28
+ "type_vocab_size": 1,
29
+ "use_cache": true,
30
+ "use_entity_aware_attention": true,
31
+ "vocab_size": 32772
32
+ }
entity_vocab.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "[MASK2]": 3,
3
+ "[MASK]": 0,
4
+ "[PAD]": 2,
5
+ "[UNK]": 1
6
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03c2ba622c0868c956d294aa54d67dad91dede8bf01d3f9987977ee4b01115ab
3
+ size 530006753
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79eeb7b127f1c157168d9f5c5ab3f0d94396179378432f89a71f23d9c62121f
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad53b67dab70b9122dcc0624e6d45f081fdebb2466865fea90439d1349d73806
3
+ size 627
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8b73a5e054936c920cf5b7d1ec21ce9c281977078269963beb821c6c86fbff7
3
+ size 841889
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<ent>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "<ent2>",
12
+ "lstrip": false,
13
+ "normalized": false,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ }
17
+ ],
18
+ "bos_token": "<s>",
19
+ "cls_token": "<s>",
20
+ "eos_token": "</s>",
21
+ "mask_token": {
22
+ "content": "<mask>",
23
+ "lstrip": true,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ },
28
+ "pad_token": "<pad>",
29
+ "sep_token": "</s>",
30
+ "unk_token": "<unk>"
31
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "__type": "AddedToken",
5
+ "content": "<ent>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ {
12
+ "__type": "AddedToken",
13
+ "content": "<ent2>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ {
20
+ "__type": "AddedToken",
21
+ "content": "<ent>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ {
28
+ "__type": "AddedToken",
29
+ "content": "<ent2>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ ],
36
+ "bos_token": "<s>",
37
+ "cls_token": "<s>",
38
+ "entity_mask2_token": "[MASK2]",
39
+ "entity_mask_token": "[MASK]",
40
+ "entity_pad_token": "[PAD]",
41
+ "entity_token_1": {
42
+ "__type": "AddedToken",
43
+ "content": "<ent>",
44
+ "lstrip": false,
45
+ "normalized": true,
46
+ "rstrip": false,
47
+ "single_word": false
48
+ },
49
+ "entity_token_2": {
50
+ "__type": "AddedToken",
51
+ "content": "<ent2>",
52
+ "lstrip": false,
53
+ "normalized": true,
54
+ "rstrip": false,
55
+ "single_word": false
56
+ },
57
+ "entity_unk_token": "[UNK]",
58
+ "eos_token": "</s>",
59
+ "mask_token": {
60
+ "__type": "AddedToken",
61
+ "content": "<mask>",
62
+ "lstrip": true,
63
+ "normalized": true,
64
+ "rstrip": false,
65
+ "single_word": false
66
+ },
67
+ "max_entity_length": 32,
68
+ "max_mention_length": 30,
69
+ "model_max_length": 512,
70
+ "name_or_path": "studio-ousia/luke-japanese-base-lite",
71
+ "pad_token": "<pad>",
72
+ "sep_token": "</s>",
73
+ "sp_model_kwargs": {},
74
+ "special_tokens_map_file": "models/luke-japanese/hf_xlm_roberta/special_tokens_map.json",
75
+ "task": null,
76
+ "tokenizer_class": "MLukeTokenizer",
77
+ "tokenizer_file": "models/luke-japanese/hf_luke_japanese_lite_epoch20/tokenizer.json",
78
+ "unk_token": "<unk>"
79
+ }
trainer_state.json ADDED
@@ -0,0 +1,766 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.997184174103947,
5
+ "global_step": 62800,
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": 5e-05,
13
+ "loss": 3.2345,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 4.959909555958242e-05,
19
+ "loss": 1.994,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.07,
24
+ "learning_rate": 4.919819111916484e-05,
25
+ "loss": 1.6753,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.1,
30
+ "learning_rate": 4.8797286678747256e-05,
31
+ "loss": 1.5453,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.12,
36
+ "learning_rate": 4.8396382238329674e-05,
37
+ "loss": 1.4013,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.14,
42
+ "learning_rate": 4.799547779791209e-05,
43
+ "loss": 1.373,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "learning_rate": 4.759457335749451e-05,
49
+ "loss": 1.2924,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.19,
54
+ "learning_rate": 4.719366891707693e-05,
55
+ "loss": 1.2266,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.21,
60
+ "learning_rate": 4.6792764476659346e-05,
61
+ "loss": 1.1925,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.24,
66
+ "learning_rate": 4.6391860036241764e-05,
67
+ "loss": 1.22,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.26,
72
+ "learning_rate": 4.599095559582418e-05,
73
+ "loss": 1.0955,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.29,
78
+ "learning_rate": 4.55900511554066e-05,
79
+ "loss": 1.1189,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.31,
84
+ "learning_rate": 4.518914671498902e-05,
85
+ "loss": 1.0632,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.33,
90
+ "learning_rate": 4.4788242274571436e-05,
91
+ "loss": 1.0091,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.36,
96
+ "learning_rate": 4.4387337834153854e-05,
97
+ "loss": 1.0058,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.38,
102
+ "learning_rate": 4.398643339373627e-05,
103
+ "loss": 1.0111,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.41,
108
+ "learning_rate": 4.358552895331869e-05,
109
+ "loss": 0.8988,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.43,
114
+ "learning_rate": 4.318462451290111e-05,
115
+ "loss": 0.9305,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.45,
120
+ "learning_rate": 4.2783720072483526e-05,
121
+ "loss": 0.9372,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.48,
126
+ "learning_rate": 4.2382815632065944e-05,
127
+ "loss": 0.9176,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.5,
132
+ "learning_rate": 4.198191119164836e-05,
133
+ "loss": 0.8878,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.52,
138
+ "learning_rate": 4.158100675123078e-05,
139
+ "loss": 0.872,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.55,
144
+ "learning_rate": 4.11801023108132e-05,
145
+ "loss": 0.8525,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.57,
150
+ "learning_rate": 4.0779197870395616e-05,
151
+ "loss": 0.8511,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.6,
156
+ "learning_rate": 4.0378293429978034e-05,
157
+ "loss": 0.828,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.62,
162
+ "learning_rate": 3.997738898956045e-05,
163
+ "loss": 0.8198,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.64,
168
+ "learning_rate": 3.957648454914287e-05,
169
+ "loss": 0.8225,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.67,
174
+ "learning_rate": 3.917558010872529e-05,
175
+ "loss": 0.7818,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.69,
180
+ "learning_rate": 3.8774675668307706e-05,
181
+ "loss": 0.7939,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.72,
186
+ "learning_rate": 3.8373771227890124e-05,
187
+ "loss": 0.7799,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.74,
192
+ "learning_rate": 3.7972866787472535e-05,
193
+ "loss": 0.7588,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.76,
198
+ "learning_rate": 3.757196234705495e-05,
199
+ "loss": 0.8004,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.79,
204
+ "learning_rate": 3.717105790663737e-05,
205
+ "loss": 0.7468,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.81,
210
+ "learning_rate": 3.677015346621979e-05,
211
+ "loss": 0.7574,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.84,
216
+ "learning_rate": 3.636924902580221e-05,
217
+ "loss": 0.7193,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.86,
222
+ "learning_rate": 3.5968344585384625e-05,
223
+ "loss": 0.7122,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.88,
228
+ "learning_rate": 3.556744014496704e-05,
229
+ "loss": 0.7215,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.91,
234
+ "learning_rate": 3.516653570454946e-05,
235
+ "loss": 0.6629,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.93,
240
+ "learning_rate": 3.476563126413188e-05,
241
+ "loss": 0.7112,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.95,
246
+ "learning_rate": 3.43647268237143e-05,
247
+ "loss": 0.7116,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.98,
252
+ "learning_rate": 3.3963822383296715e-05,
253
+ "loss": 0.6947,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 1.0,
258
+ "learning_rate": 3.356291794287913e-05,
259
+ "loss": 0.6492,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 1.03,
264
+ "learning_rate": 3.316201350246155e-05,
265
+ "loss": 0.5083,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 1.05,
270
+ "learning_rate": 3.276110906204397e-05,
271
+ "loss": 0.536,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 1.07,
276
+ "learning_rate": 3.236020462162639e-05,
277
+ "loss": 0.5196,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 1.1,
282
+ "learning_rate": 3.1959300181208805e-05,
283
+ "loss": 0.5637,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 1.12,
288
+ "learning_rate": 3.155839574079122e-05,
289
+ "loss": 0.5338,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 1.15,
294
+ "learning_rate": 3.115749130037364e-05,
295
+ "loss": 0.5052,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 1.17,
300
+ "learning_rate": 3.075658685995606e-05,
301
+ "loss": 0.5186,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 1.19,
306
+ "learning_rate": 3.0355682419538477e-05,
307
+ "loss": 0.5168,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 1.22,
312
+ "learning_rate": 2.9954777979120895e-05,
313
+ "loss": 0.4931,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 1.24,
318
+ "learning_rate": 2.9553873538703313e-05,
319
+ "loss": 0.4993,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 1.26,
324
+ "learning_rate": 2.915296909828573e-05,
325
+ "loss": 0.5286,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 1.29,
330
+ "learning_rate": 2.875206465786815e-05,
331
+ "loss": 0.4865,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 1.31,
336
+ "learning_rate": 2.8351160217450567e-05,
337
+ "loss": 0.503,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 1.34,
342
+ "learning_rate": 2.7950255777032985e-05,
343
+ "loss": 0.5093,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 1.36,
348
+ "learning_rate": 2.7549351336615403e-05,
349
+ "loss": 0.4978,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 1.38,
354
+ "learning_rate": 2.714844689619782e-05,
355
+ "loss": 0.5042,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 1.41,
360
+ "learning_rate": 2.674754245578024e-05,
361
+ "loss": 0.4909,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 1.43,
366
+ "learning_rate": 2.6346638015362657e-05,
367
+ "loss": 0.4861,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 1.46,
372
+ "learning_rate": 2.5945733574945075e-05,
373
+ "loss": 0.472,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 1.48,
378
+ "learning_rate": 2.5544829134527493e-05,
379
+ "loss": 0.4673,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 1.5,
384
+ "learning_rate": 2.514392469410991e-05,
385
+ "loss": 0.4747,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 1.53,
390
+ "learning_rate": 2.4743020253692332e-05,
391
+ "loss": 0.4396,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 1.55,
396
+ "learning_rate": 2.434211581327475e-05,
397
+ "loss": 0.4675,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 1.57,
402
+ "learning_rate": 2.3941211372857168e-05,
403
+ "loss": 0.4421,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 1.6,
408
+ "learning_rate": 2.3540306932439586e-05,
409
+ "loss": 0.4189,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 1.62,
414
+ "learning_rate": 2.3139402492022004e-05,
415
+ "loss": 0.4305,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 1.65,
420
+ "learning_rate": 2.273849805160442e-05,
421
+ "loss": 0.4332,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 1.67,
426
+ "learning_rate": 2.2337593611186837e-05,
427
+ "loss": 0.4314,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 1.69,
432
+ "learning_rate": 2.1936689170769255e-05,
433
+ "loss": 0.4138,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 1.72,
438
+ "learning_rate": 2.1535784730351673e-05,
439
+ "loss": 0.4179,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 1.74,
444
+ "learning_rate": 2.113488028993409e-05,
445
+ "loss": 0.4104,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 1.77,
450
+ "learning_rate": 2.073397584951651e-05,
451
+ "loss": 0.4078,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 1.79,
456
+ "learning_rate": 2.0333071409098927e-05,
457
+ "loss": 0.4113,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 1.81,
462
+ "learning_rate": 1.9932166968681345e-05,
463
+ "loss": 0.3828,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 1.84,
468
+ "learning_rate": 1.9531262528263763e-05,
469
+ "loss": 0.4095,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 1.86,
474
+ "learning_rate": 1.913035808784618e-05,
475
+ "loss": 0.3782,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 1.89,
480
+ "learning_rate": 1.87294536474286e-05,
481
+ "loss": 0.3494,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 1.91,
486
+ "learning_rate": 1.8328549207011017e-05,
487
+ "loss": 0.3758,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 1.93,
492
+ "learning_rate": 1.7927644766593435e-05,
493
+ "loss": 0.3628,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 1.96,
498
+ "learning_rate": 1.7526740326175853e-05,
499
+ "loss": 0.3751,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 1.98,
504
+ "learning_rate": 1.712583588575827e-05,
505
+ "loss": 0.3474,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 2.0,
510
+ "learning_rate": 1.672493144534069e-05,
511
+ "loss": 0.3201,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 2.03,
516
+ "learning_rate": 1.6324027004923107e-05,
517
+ "loss": 0.237,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 2.05,
522
+ "learning_rate": 1.5923122564505525e-05,
523
+ "loss": 0.2274,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 2.08,
528
+ "learning_rate": 1.5522218124087943e-05,
529
+ "loss": 0.2453,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 2.1,
534
+ "learning_rate": 1.512131368367036e-05,
535
+ "loss": 0.2502,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 2.12,
540
+ "learning_rate": 1.4720409243252779e-05,
541
+ "loss": 0.2575,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 2.15,
546
+ "learning_rate": 1.4319504802835197e-05,
547
+ "loss": 0.2533,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 2.17,
552
+ "learning_rate": 1.3918600362417613e-05,
553
+ "loss": 0.2496,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 2.2,
558
+ "learning_rate": 1.3517695922000031e-05,
559
+ "loss": 0.2461,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 2.22,
564
+ "learning_rate": 1.3116791481582449e-05,
565
+ "loss": 0.2509,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 2.24,
570
+ "learning_rate": 1.2715887041164867e-05,
571
+ "loss": 0.2343,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 2.27,
576
+ "learning_rate": 1.2314982600747287e-05,
577
+ "loss": 0.2549,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 2.29,
582
+ "learning_rate": 1.1914078160329705e-05,
583
+ "loss": 0.2382,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 2.31,
588
+ "learning_rate": 1.1513173719912123e-05,
589
+ "loss": 0.2364,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 2.34,
594
+ "learning_rate": 1.111226927949454e-05,
595
+ "loss": 0.2334,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 2.36,
600
+ "learning_rate": 1.0711364839076959e-05,
601
+ "loss": 0.2291,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 2.39,
606
+ "learning_rate": 1.0310460398659377e-05,
607
+ "loss": 0.2187,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 2.41,
612
+ "learning_rate": 9.909555958241795e-06,
613
+ "loss": 0.1919,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 2.43,
618
+ "learning_rate": 9.50865151782421e-06,
619
+ "loss": 0.2364,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 2.46,
624
+ "learning_rate": 9.107747077406629e-06,
625
+ "loss": 0.2045,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 2.48,
630
+ "learning_rate": 8.706842636989047e-06,
631
+ "loss": 0.2117,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 2.51,
636
+ "learning_rate": 8.305938196571465e-06,
637
+ "loss": 0.2266,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 2.53,
642
+ "learning_rate": 7.905033756153883e-06,
643
+ "loss": 0.2023,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 2.55,
648
+ "learning_rate": 7.504129315736301e-06,
649
+ "loss": 0.2032,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 2.58,
654
+ "learning_rate": 7.103224875318719e-06,
655
+ "loss": 0.1957,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 2.6,
660
+ "learning_rate": 6.702320434901137e-06,
661
+ "loss": 0.1869,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 2.62,
666
+ "learning_rate": 6.301415994483555e-06,
667
+ "loss": 0.2016,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 2.65,
672
+ "learning_rate": 5.9005115540659736e-06,
673
+ "loss": 0.1831,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 2.67,
678
+ "learning_rate": 5.499607113648391e-06,
679
+ "loss": 0.1707,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 2.7,
684
+ "learning_rate": 5.098702673230809e-06,
685
+ "loss": 0.1902,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 2.72,
690
+ "learning_rate": 4.697798232813227e-06,
691
+ "loss": 0.1776,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 2.74,
696
+ "learning_rate": 4.296893792395645e-06,
697
+ "loss": 0.2065,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 2.77,
702
+ "learning_rate": 3.895989351978063e-06,
703
+ "loss": 0.1955,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 2.79,
708
+ "learning_rate": 3.4950849115604802e-06,
709
+ "loss": 0.1747,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 2.82,
714
+ "learning_rate": 3.0941804711428986e-06,
715
+ "loss": 0.198,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 2.84,
720
+ "learning_rate": 2.6932760307253166e-06,
721
+ "loss": 0.1885,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 2.86,
726
+ "learning_rate": 2.292371590307734e-06,
727
+ "loss": 0.1537,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 2.89,
732
+ "learning_rate": 1.8914671498901522e-06,
733
+ "loss": 0.1657,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 2.91,
738
+ "learning_rate": 1.4905627094725702e-06,
739
+ "loss": 0.1631,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 2.94,
744
+ "learning_rate": 1.0896582690549882e-06,
745
+ "loss": 0.1417,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 2.96,
750
+ "learning_rate": 6.88753828637406e-07,
751
+ "loss": 0.157,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 2.98,
756
+ "learning_rate": 2.878493882198239e-07,
757
+ "loss": 0.1536,
758
+ "step": 62500
759
+ }
760
+ ],
761
+ "max_steps": 62859,
762
+ "num_train_epochs": 3,
763
+ "total_flos": 4.536353409695568e+16,
764
+ "trial_name": null,
765
+ "trial_params": null
766
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ab4e40d5cb02f5fb5d0c022e8d4da2aa55f43a63828785fdca14114735e2e75
3
+ size 3515