Tianyu Gao commited on
Commit
0eed4bd
1 Parent(s): 17a49de
config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "result/bert-large-uncased-cls_before_pooler-sym_mlp-mlp_bert-bs64-gpu8-gs1-lr5e-5-m=stsb-norm0.05-l32-wiki-id-1epoch",
3
+ "architectures": [
4
+ "BertForCL"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 1024,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 4096,
13
+ "layer_norm_eps": 1e-12,
14
+ "max_position_embeddings": 512,
15
+ "model_type": "bert",
16
+ "num_attention_heads": 16,
17
+ "num_hidden_layers": 24,
18
+ "pad_token_id": 0,
19
+ "position_embedding_type": "absolute",
20
+ "transformers_version": "4.2.1",
21
+ "type_vocab_size": 2,
22
+ "use_cache": true,
23
+ "vocab_size": 30522
24
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56469ce2bc958e078c00c62ffc7a855897f7c00a878b385f8f6c161b40f841d2
3
+ size 2681374517
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0d75eb5e1c7699fe7095919c542b0f3e1664be3e4962d0dfb8a48e9141fe738
3
+ size 1340700103
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c677d46c0d626f3fec1579bd9577f5c4f4ab164833476209eb04e4f8c325e50
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_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, "name_or_path": "bert-large-uncased"}
train_results.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ epoch = 1.0
2
+ train_runtime = 3451.5882
3
+ train_samples_per_second = 0.566
trainer_state.json ADDED
@@ -0,0 +1,1297 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8233758432892795,
3
+ "best_model_checkpoint": "result/bert-large-uncased-cls_before_pooler-sym_mlp-mlp_bert-bs64-gpu8-gs1-lr5e-5-m=stsb-norm0.05-l32-wiki-id-1epoch",
4
+ "epoch": 1.0,
5
+ "global_step": 1954,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.9744114636642786e-05,
13
+ "loss": 0.7987,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.948822927328557e-05,
19
+ "loss": 0.0132,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.923234390992835e-05,
25
+ "loss": 0.0083,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.02,
30
+ "learning_rate": 4.897645854657114e-05,
31
+ "loss": 0.0044,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.03,
36
+ "learning_rate": 4.8720573183213926e-05,
37
+ "loss": 0.0041,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.03,
42
+ "learning_rate": 4.846468781985671e-05,
43
+ "loss": 0.0032,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.04,
48
+ "learning_rate": 4.820880245649949e-05,
49
+ "loss": 0.0022,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.04,
54
+ "learning_rate": 4.795291709314228e-05,
55
+ "loss": 0.0014,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.05,
60
+ "learning_rate": 4.769703172978506e-05,
61
+ "loss": 0.0023,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.05,
66
+ "learning_rate": 4.7441146366427843e-05,
67
+ "loss": 0.0012,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.06,
72
+ "learning_rate": 4.718526100307063e-05,
73
+ "loss": 0.0013,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.06,
78
+ "learning_rate": 4.692937563971341e-05,
79
+ "loss": 0.0017,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "eval_avg_sts": 0.7527941583986333,
85
+ "eval_sickr_spearman": 0.7351572662264496,
86
+ "eval_stsb_spearman": 0.770431050570817,
87
+ "step": 125
88
+ },
89
+ {
90
+ "epoch": 0.07,
91
+ "learning_rate": 4.6673490276356194e-05,
92
+ "loss": 0.0021,
93
+ "step": 130
94
+ },
95
+ {
96
+ "epoch": 0.07,
97
+ "learning_rate": 4.641760491299898e-05,
98
+ "loss": 0.0012,
99
+ "step": 140
100
+ },
101
+ {
102
+ "epoch": 0.08,
103
+ "learning_rate": 4.616171954964176e-05,
104
+ "loss": 0.0017,
105
+ "step": 150
106
+ },
107
+ {
108
+ "epoch": 0.08,
109
+ "learning_rate": 4.5905834186284544e-05,
110
+ "loss": 0.0011,
111
+ "step": 160
112
+ },
113
+ {
114
+ "epoch": 0.09,
115
+ "learning_rate": 4.564994882292733e-05,
116
+ "loss": 0.0004,
117
+ "step": 170
118
+ },
119
+ {
120
+ "epoch": 0.09,
121
+ "learning_rate": 4.539406345957011e-05,
122
+ "loss": 0.001,
123
+ "step": 180
124
+ },
125
+ {
126
+ "epoch": 0.1,
127
+ "learning_rate": 4.5138178096212894e-05,
128
+ "loss": 0.0013,
129
+ "step": 190
130
+ },
131
+ {
132
+ "epoch": 0.1,
133
+ "learning_rate": 4.4882292732855685e-05,
134
+ "loss": 0.0011,
135
+ "step": 200
136
+ },
137
+ {
138
+ "epoch": 0.11,
139
+ "learning_rate": 4.462640736949847e-05,
140
+ "loss": 0.0016,
141
+ "step": 210
142
+ },
143
+ {
144
+ "epoch": 0.11,
145
+ "learning_rate": 4.437052200614125e-05,
146
+ "loss": 0.0008,
147
+ "step": 220
148
+ },
149
+ {
150
+ "epoch": 0.12,
151
+ "learning_rate": 4.4114636642784035e-05,
152
+ "loss": 0.0006,
153
+ "step": 230
154
+ },
155
+ {
156
+ "epoch": 0.12,
157
+ "learning_rate": 4.385875127942682e-05,
158
+ "loss": 0.0012,
159
+ "step": 240
160
+ },
161
+ {
162
+ "epoch": 0.13,
163
+ "learning_rate": 4.36028659160696e-05,
164
+ "loss": 0.0012,
165
+ "step": 250
166
+ },
167
+ {
168
+ "epoch": 0.13,
169
+ "eval_avg_sts": 0.7786723498526811,
170
+ "eval_sickr_spearman": 0.7543037120683187,
171
+ "eval_stsb_spearman": 0.8030409876370435,
172
+ "step": 250
173
+ },
174
+ {
175
+ "epoch": 0.13,
176
+ "learning_rate": 4.3346980552712385e-05,
177
+ "loss": 0.001,
178
+ "step": 260
179
+ },
180
+ {
181
+ "epoch": 0.14,
182
+ "learning_rate": 4.309109518935517e-05,
183
+ "loss": 0.0007,
184
+ "step": 270
185
+ },
186
+ {
187
+ "epoch": 0.14,
188
+ "learning_rate": 4.283520982599796e-05,
189
+ "loss": 0.0009,
190
+ "step": 280
191
+ },
192
+ {
193
+ "epoch": 0.15,
194
+ "learning_rate": 4.257932446264074e-05,
195
+ "loss": 0.0012,
196
+ "step": 290
197
+ },
198
+ {
199
+ "epoch": 0.15,
200
+ "learning_rate": 4.2323439099283526e-05,
201
+ "loss": 0.0015,
202
+ "step": 300
203
+ },
204
+ {
205
+ "epoch": 0.16,
206
+ "learning_rate": 4.206755373592631e-05,
207
+ "loss": 0.0013,
208
+ "step": 310
209
+ },
210
+ {
211
+ "epoch": 0.16,
212
+ "learning_rate": 4.181166837256909e-05,
213
+ "loss": 0.0019,
214
+ "step": 320
215
+ },
216
+ {
217
+ "epoch": 0.17,
218
+ "learning_rate": 4.1555783009211876e-05,
219
+ "loss": 0.0013,
220
+ "step": 330
221
+ },
222
+ {
223
+ "epoch": 0.17,
224
+ "learning_rate": 4.129989764585466e-05,
225
+ "loss": 0.0007,
226
+ "step": 340
227
+ },
228
+ {
229
+ "epoch": 0.18,
230
+ "learning_rate": 4.104401228249744e-05,
231
+ "loss": 0.0004,
232
+ "step": 350
233
+ },
234
+ {
235
+ "epoch": 0.18,
236
+ "learning_rate": 4.0788126919140226e-05,
237
+ "loss": 0.0015,
238
+ "step": 360
239
+ },
240
+ {
241
+ "epoch": 0.19,
242
+ "learning_rate": 4.053224155578301e-05,
243
+ "loss": 0.0004,
244
+ "step": 370
245
+ },
246
+ {
247
+ "epoch": 0.19,
248
+ "eval_avg_sts": 0.7913077528900125,
249
+ "eval_sickr_spearman": 0.7655320387137092,
250
+ "eval_stsb_spearman": 0.8170834670663158,
251
+ "step": 375
252
+ },
253
+ {
254
+ "epoch": 0.19,
255
+ "learning_rate": 4.027635619242579e-05,
256
+ "loss": 0.0014,
257
+ "step": 380
258
+ },
259
+ {
260
+ "epoch": 0.2,
261
+ "learning_rate": 4.0020470829068577e-05,
262
+ "loss": 0.0013,
263
+ "step": 390
264
+ },
265
+ {
266
+ "epoch": 0.2,
267
+ "learning_rate": 3.976458546571136e-05,
268
+ "loss": 0.0015,
269
+ "step": 400
270
+ },
271
+ {
272
+ "epoch": 0.21,
273
+ "learning_rate": 3.9508700102354143e-05,
274
+ "loss": 0.0006,
275
+ "step": 410
276
+ },
277
+ {
278
+ "epoch": 0.21,
279
+ "learning_rate": 3.925281473899693e-05,
280
+ "loss": 0.0005,
281
+ "step": 420
282
+ },
283
+ {
284
+ "epoch": 0.22,
285
+ "learning_rate": 3.899692937563972e-05,
286
+ "loss": 0.0006,
287
+ "step": 430
288
+ },
289
+ {
290
+ "epoch": 0.23,
291
+ "learning_rate": 3.87410440122825e-05,
292
+ "loss": 0.0004,
293
+ "step": 440
294
+ },
295
+ {
296
+ "epoch": 0.23,
297
+ "learning_rate": 3.8485158648925284e-05,
298
+ "loss": 0.001,
299
+ "step": 450
300
+ },
301
+ {
302
+ "epoch": 0.24,
303
+ "learning_rate": 3.822927328556807e-05,
304
+ "loss": 0.0012,
305
+ "step": 460
306
+ },
307
+ {
308
+ "epoch": 0.24,
309
+ "learning_rate": 3.797338792221085e-05,
310
+ "loss": 0.0014,
311
+ "step": 470
312
+ },
313
+ {
314
+ "epoch": 0.25,
315
+ "learning_rate": 3.7717502558853634e-05,
316
+ "loss": 0.0007,
317
+ "step": 480
318
+ },
319
+ {
320
+ "epoch": 0.25,
321
+ "learning_rate": 3.746161719549642e-05,
322
+ "loss": 0.0011,
323
+ "step": 490
324
+ },
325
+ {
326
+ "epoch": 0.26,
327
+ "learning_rate": 3.72057318321392e-05,
328
+ "loss": 0.0011,
329
+ "step": 500
330
+ },
331
+ {
332
+ "epoch": 0.26,
333
+ "eval_avg_sts": 0.7880331799657297,
334
+ "eval_sickr_spearman": 0.7571696318289279,
335
+ "eval_stsb_spearman": 0.8188967281025316,
336
+ "step": 500
337
+ },
338
+ {
339
+ "epoch": 0.26,
340
+ "learning_rate": 3.694984646878199e-05,
341
+ "loss": 0.0003,
342
+ "step": 510
343
+ },
344
+ {
345
+ "epoch": 0.27,
346
+ "learning_rate": 3.6693961105424775e-05,
347
+ "loss": 0.0007,
348
+ "step": 520
349
+ },
350
+ {
351
+ "epoch": 0.27,
352
+ "learning_rate": 3.643807574206756e-05,
353
+ "loss": 0.0007,
354
+ "step": 530
355
+ },
356
+ {
357
+ "epoch": 0.28,
358
+ "learning_rate": 3.618219037871034e-05,
359
+ "loss": 0.0021,
360
+ "step": 540
361
+ },
362
+ {
363
+ "epoch": 0.28,
364
+ "learning_rate": 3.5926305015353125e-05,
365
+ "loss": 0.0019,
366
+ "step": 550
367
+ },
368
+ {
369
+ "epoch": 0.29,
370
+ "learning_rate": 3.567041965199591e-05,
371
+ "loss": 0.0015,
372
+ "step": 560
373
+ },
374
+ {
375
+ "epoch": 0.29,
376
+ "learning_rate": 3.541453428863869e-05,
377
+ "loss": 0.0002,
378
+ "step": 570
379
+ },
380
+ {
381
+ "epoch": 0.3,
382
+ "learning_rate": 3.5158648925281475e-05,
383
+ "loss": 0.0017,
384
+ "step": 580
385
+ },
386
+ {
387
+ "epoch": 0.3,
388
+ "learning_rate": 3.4902763561924265e-05,
389
+ "loss": 0.0018,
390
+ "step": 590
391
+ },
392
+ {
393
+ "epoch": 0.31,
394
+ "learning_rate": 3.464687819856704e-05,
395
+ "loss": 0.0013,
396
+ "step": 600
397
+ },
398
+ {
399
+ "epoch": 0.31,
400
+ "learning_rate": 3.4390992835209826e-05,
401
+ "loss": 0.0008,
402
+ "step": 610
403
+ },
404
+ {
405
+ "epoch": 0.32,
406
+ "learning_rate": 3.413510747185261e-05,
407
+ "loss": 0.0022,
408
+ "step": 620
409
+ },
410
+ {
411
+ "epoch": 0.32,
412
+ "eval_avg_sts": 0.7732885189334773,
413
+ "eval_sickr_spearman": 0.7409421801147573,
414
+ "eval_stsb_spearman": 0.8056348577521973,
415
+ "step": 625
416
+ },
417
+ {
418
+ "epoch": 0.32,
419
+ "learning_rate": 3.387922210849539e-05,
420
+ "loss": 0.0022,
421
+ "step": 630
422
+ },
423
+ {
424
+ "epoch": 0.33,
425
+ "learning_rate": 3.3623336745138176e-05,
426
+ "loss": 0.0006,
427
+ "step": 640
428
+ },
429
+ {
430
+ "epoch": 0.33,
431
+ "learning_rate": 3.336745138178096e-05,
432
+ "loss": 0.0011,
433
+ "step": 650
434
+ },
435
+ {
436
+ "epoch": 0.34,
437
+ "learning_rate": 3.311156601842374e-05,
438
+ "loss": 0.0097,
439
+ "step": 660
440
+ },
441
+ {
442
+ "epoch": 0.34,
443
+ "learning_rate": 3.285568065506653e-05,
444
+ "loss": 0.0065,
445
+ "step": 670
446
+ },
447
+ {
448
+ "epoch": 0.35,
449
+ "learning_rate": 3.2599795291709316e-05,
450
+ "loss": 0.0026,
451
+ "step": 680
452
+ },
453
+ {
454
+ "epoch": 0.35,
455
+ "learning_rate": 3.23439099283521e-05,
456
+ "loss": 0.002,
457
+ "step": 690
458
+ },
459
+ {
460
+ "epoch": 0.36,
461
+ "learning_rate": 3.208802456499488e-05,
462
+ "loss": 0.0029,
463
+ "step": 700
464
+ },
465
+ {
466
+ "epoch": 0.36,
467
+ "learning_rate": 3.183213920163767e-05,
468
+ "loss": 0.0025,
469
+ "step": 710
470
+ },
471
+ {
472
+ "epoch": 0.37,
473
+ "learning_rate": 3.157625383828045e-05,
474
+ "loss": 0.0024,
475
+ "step": 720
476
+ },
477
+ {
478
+ "epoch": 0.37,
479
+ "learning_rate": 3.1320368474923234e-05,
480
+ "loss": 0.001,
481
+ "step": 730
482
+ },
483
+ {
484
+ "epoch": 0.38,
485
+ "learning_rate": 3.106448311156602e-05,
486
+ "loss": 0.0016,
487
+ "step": 740
488
+ },
489
+ {
490
+ "epoch": 0.38,
491
+ "learning_rate": 3.080859774820881e-05,
492
+ "loss": 0.001,
493
+ "step": 750
494
+ },
495
+ {
496
+ "epoch": 0.38,
497
+ "eval_avg_sts": 0.7885603958950645,
498
+ "eval_sickr_spearman": 0.7648866928347867,
499
+ "eval_stsb_spearman": 0.8122340989553424,
500
+ "step": 750
501
+ },
502
+ {
503
+ "epoch": 0.39,
504
+ "learning_rate": 3.055271238485159e-05,
505
+ "loss": 0.0016,
506
+ "step": 760
507
+ },
508
+ {
509
+ "epoch": 0.39,
510
+ "learning_rate": 3.0296827021494374e-05,
511
+ "loss": 0.0007,
512
+ "step": 770
513
+ },
514
+ {
515
+ "epoch": 0.4,
516
+ "learning_rate": 3.0040941658137157e-05,
517
+ "loss": 0.0015,
518
+ "step": 780
519
+ },
520
+ {
521
+ "epoch": 0.4,
522
+ "learning_rate": 2.978505629477994e-05,
523
+ "loss": 0.0017,
524
+ "step": 790
525
+ },
526
+ {
527
+ "epoch": 0.41,
528
+ "learning_rate": 2.9529170931422724e-05,
529
+ "loss": 0.001,
530
+ "step": 800
531
+ },
532
+ {
533
+ "epoch": 0.41,
534
+ "learning_rate": 2.9273285568065504e-05,
535
+ "loss": 0.0005,
536
+ "step": 810
537
+ },
538
+ {
539
+ "epoch": 0.42,
540
+ "learning_rate": 2.9017400204708288e-05,
541
+ "loss": 0.001,
542
+ "step": 820
543
+ },
544
+ {
545
+ "epoch": 0.42,
546
+ "learning_rate": 2.8761514841351078e-05,
547
+ "loss": 0.0004,
548
+ "step": 830
549
+ },
550
+ {
551
+ "epoch": 0.43,
552
+ "learning_rate": 2.850562947799386e-05,
553
+ "loss": 0.0008,
554
+ "step": 840
555
+ },
556
+ {
557
+ "epoch": 0.44,
558
+ "learning_rate": 2.8249744114636645e-05,
559
+ "loss": 0.0008,
560
+ "step": 850
561
+ },
562
+ {
563
+ "epoch": 0.44,
564
+ "learning_rate": 2.7993858751279428e-05,
565
+ "loss": 0.0018,
566
+ "step": 860
567
+ },
568
+ {
569
+ "epoch": 0.45,
570
+ "learning_rate": 2.7737973387922212e-05,
571
+ "loss": 0.0004,
572
+ "step": 870
573
+ },
574
+ {
575
+ "epoch": 0.45,
576
+ "eval_avg_sts": 0.7926162200321307,
577
+ "eval_sickr_spearman": 0.7705847664916943,
578
+ "eval_stsb_spearman": 0.8146476735725671,
579
+ "step": 875
580
+ },
581
+ {
582
+ "epoch": 0.45,
583
+ "learning_rate": 2.7482088024564995e-05,
584
+ "loss": 0.0009,
585
+ "step": 880
586
+ },
587
+ {
588
+ "epoch": 0.46,
589
+ "learning_rate": 2.722620266120778e-05,
590
+ "loss": 0.0008,
591
+ "step": 890
592
+ },
593
+ {
594
+ "epoch": 0.46,
595
+ "learning_rate": 2.6970317297850565e-05,
596
+ "loss": 0.0011,
597
+ "step": 900
598
+ },
599
+ {
600
+ "epoch": 0.47,
601
+ "learning_rate": 2.671443193449335e-05,
602
+ "loss": 0.0009,
603
+ "step": 910
604
+ },
605
+ {
606
+ "epoch": 0.47,
607
+ "learning_rate": 2.6458546571136132e-05,
608
+ "loss": 0.0012,
609
+ "step": 920
610
+ },
611
+ {
612
+ "epoch": 0.48,
613
+ "learning_rate": 2.6202661207778916e-05,
614
+ "loss": 0.0006,
615
+ "step": 930
616
+ },
617
+ {
618
+ "epoch": 0.48,
619
+ "learning_rate": 2.59467758444217e-05,
620
+ "loss": 0.0006,
621
+ "step": 940
622
+ },
623
+ {
624
+ "epoch": 0.49,
625
+ "learning_rate": 2.5690890481064483e-05,
626
+ "loss": 0.0019,
627
+ "step": 950
628
+ },
629
+ {
630
+ "epoch": 0.49,
631
+ "learning_rate": 2.5435005117707266e-05,
632
+ "loss": 0.0003,
633
+ "step": 960
634
+ },
635
+ {
636
+ "epoch": 0.5,
637
+ "learning_rate": 2.517911975435005e-05,
638
+ "loss": 0.0015,
639
+ "step": 970
640
+ },
641
+ {
642
+ "epoch": 0.5,
643
+ "learning_rate": 2.4923234390992836e-05,
644
+ "loss": 0.0007,
645
+ "step": 980
646
+ },
647
+ {
648
+ "epoch": 0.51,
649
+ "learning_rate": 2.466734902763562e-05,
650
+ "loss": 0.0009,
651
+ "step": 990
652
+ },
653
+ {
654
+ "epoch": 0.51,
655
+ "learning_rate": 2.4411463664278406e-05,
656
+ "loss": 0.0007,
657
+ "step": 1000
658
+ },
659
+ {
660
+ "epoch": 0.51,
661
+ "eval_avg_sts": 0.7959315624689012,
662
+ "eval_sickr_spearman": 0.774320769654855,
663
+ "eval_stsb_spearman": 0.8175423552829473,
664
+ "step": 1000
665
+ },
666
+ {
667
+ "epoch": 0.52,
668
+ "learning_rate": 2.415557830092119e-05,
669
+ "loss": 0.0008,
670
+ "step": 1010
671
+ },
672
+ {
673
+ "epoch": 0.52,
674
+ "learning_rate": 2.3899692937563973e-05,
675
+ "loss": 0.0007,
676
+ "step": 1020
677
+ },
678
+ {
679
+ "epoch": 0.53,
680
+ "learning_rate": 2.3643807574206757e-05,
681
+ "loss": 0.0008,
682
+ "step": 1030
683
+ },
684
+ {
685
+ "epoch": 0.53,
686
+ "learning_rate": 2.338792221084954e-05,
687
+ "loss": 0.0017,
688
+ "step": 1040
689
+ },
690
+ {
691
+ "epoch": 0.54,
692
+ "learning_rate": 2.3132036847492324e-05,
693
+ "loss": 0.0008,
694
+ "step": 1050
695
+ },
696
+ {
697
+ "epoch": 0.54,
698
+ "learning_rate": 2.2876151484135107e-05,
699
+ "loss": 0.0013,
700
+ "step": 1060
701
+ },
702
+ {
703
+ "epoch": 0.55,
704
+ "learning_rate": 2.262026612077789e-05,
705
+ "loss": 0.0003,
706
+ "step": 1070
707
+ },
708
+ {
709
+ "epoch": 0.55,
710
+ "learning_rate": 2.2364380757420677e-05,
711
+ "loss": 0.0007,
712
+ "step": 1080
713
+ },
714
+ {
715
+ "epoch": 0.56,
716
+ "learning_rate": 2.210849539406346e-05,
717
+ "loss": 0.0014,
718
+ "step": 1090
719
+ },
720
+ {
721
+ "epoch": 0.56,
722
+ "learning_rate": 2.1852610030706244e-05,
723
+ "loss": 0.001,
724
+ "step": 1100
725
+ },
726
+ {
727
+ "epoch": 0.57,
728
+ "learning_rate": 2.1596724667349028e-05,
729
+ "loss": 0.0005,
730
+ "step": 1110
731
+ },
732
+ {
733
+ "epoch": 0.57,
734
+ "learning_rate": 2.1340839303991814e-05,
735
+ "loss": 0.0005,
736
+ "step": 1120
737
+ },
738
+ {
739
+ "epoch": 0.58,
740
+ "eval_avg_sts": 0.795896362310046,
741
+ "eval_sickr_spearman": 0.7711842426687433,
742
+ "eval_stsb_spearman": 0.8206084819513487,
743
+ "step": 1125
744
+ },
745
+ {
746
+ "epoch": 0.58,
747
+ "learning_rate": 2.1084953940634598e-05,
748
+ "loss": 0.0009,
749
+ "step": 1130
750
+ },
751
+ {
752
+ "epoch": 0.58,
753
+ "learning_rate": 2.082906857727738e-05,
754
+ "loss": 0.0015,
755
+ "step": 1140
756
+ },
757
+ {
758
+ "epoch": 0.59,
759
+ "learning_rate": 2.0573183213920165e-05,
760
+ "loss": 0.0009,
761
+ "step": 1150
762
+ },
763
+ {
764
+ "epoch": 0.59,
765
+ "learning_rate": 2.0317297850562948e-05,
766
+ "loss": 0.0006,
767
+ "step": 1160
768
+ },
769
+ {
770
+ "epoch": 0.6,
771
+ "learning_rate": 2.006141248720573e-05,
772
+ "loss": 0.0004,
773
+ "step": 1170
774
+ },
775
+ {
776
+ "epoch": 0.6,
777
+ "learning_rate": 1.9805527123848515e-05,
778
+ "loss": 0.0008,
779
+ "step": 1180
780
+ },
781
+ {
782
+ "epoch": 0.61,
783
+ "learning_rate": 1.95496417604913e-05,
784
+ "loss": 0.0005,
785
+ "step": 1190
786
+ },
787
+ {
788
+ "epoch": 0.61,
789
+ "learning_rate": 1.9293756397134085e-05,
790
+ "loss": 0.0014,
791
+ "step": 1200
792
+ },
793
+ {
794
+ "epoch": 0.62,
795
+ "learning_rate": 1.903787103377687e-05,
796
+ "loss": 0.0008,
797
+ "step": 1210
798
+ },
799
+ {
800
+ "epoch": 0.62,
801
+ "learning_rate": 1.8781985670419652e-05,
802
+ "loss": 0.0014,
803
+ "step": 1220
804
+ },
805
+ {
806
+ "epoch": 0.63,
807
+ "learning_rate": 1.8526100307062436e-05,
808
+ "loss": 0.0006,
809
+ "step": 1230
810
+ },
811
+ {
812
+ "epoch": 0.63,
813
+ "learning_rate": 1.8270214943705222e-05,
814
+ "loss": 0.0009,
815
+ "step": 1240
816
+ },
817
+ {
818
+ "epoch": 0.64,
819
+ "learning_rate": 1.8014329580348006e-05,
820
+ "loss": 0.0006,
821
+ "step": 1250
822
+ },
823
+ {
824
+ "epoch": 0.64,
825
+ "eval_avg_sts": 0.7963228217070192,
826
+ "eval_sickr_spearman": 0.772572149375899,
827
+ "eval_stsb_spearman": 0.8200734940381393,
828
+ "step": 1250
829
+ },
830
+ {
831
+ "epoch": 0.64,
832
+ "learning_rate": 1.775844421699079e-05,
833
+ "loss": 0.0009,
834
+ "step": 1260
835
+ },
836
+ {
837
+ "epoch": 0.65,
838
+ "learning_rate": 1.7502558853633573e-05,
839
+ "loss": 0.0002,
840
+ "step": 1270
841
+ },
842
+ {
843
+ "epoch": 0.66,
844
+ "learning_rate": 1.7246673490276356e-05,
845
+ "loss": 0.0007,
846
+ "step": 1280
847
+ },
848
+ {
849
+ "epoch": 0.66,
850
+ "learning_rate": 1.699078812691914e-05,
851
+ "loss": 0.0007,
852
+ "step": 1290
853
+ },
854
+ {
855
+ "epoch": 0.67,
856
+ "learning_rate": 1.6734902763561923e-05,
857
+ "loss": 0.0014,
858
+ "step": 1300
859
+ },
860
+ {
861
+ "epoch": 0.67,
862
+ "learning_rate": 1.647901740020471e-05,
863
+ "loss": 0.0011,
864
+ "step": 1310
865
+ },
866
+ {
867
+ "epoch": 0.68,
868
+ "learning_rate": 1.6223132036847493e-05,
869
+ "loss": 0.0011,
870
+ "step": 1320
871
+ },
872
+ {
873
+ "epoch": 0.68,
874
+ "learning_rate": 1.5967246673490277e-05,
875
+ "loss": 0.0008,
876
+ "step": 1330
877
+ },
878
+ {
879
+ "epoch": 0.69,
880
+ "learning_rate": 1.571136131013306e-05,
881
+ "loss": 0.001,
882
+ "step": 1340
883
+ },
884
+ {
885
+ "epoch": 0.69,
886
+ "learning_rate": 1.5455475946775847e-05,
887
+ "loss": 0.0017,
888
+ "step": 1350
889
+ },
890
+ {
891
+ "epoch": 0.7,
892
+ "learning_rate": 1.519959058341863e-05,
893
+ "loss": 0.0004,
894
+ "step": 1360
895
+ },
896
+ {
897
+ "epoch": 0.7,
898
+ "learning_rate": 1.4943705220061414e-05,
899
+ "loss": 0.0009,
900
+ "step": 1370
901
+ },
902
+ {
903
+ "epoch": 0.7,
904
+ "eval_avg_sts": 0.794838346221856,
905
+ "eval_sickr_spearman": 0.7716415467855355,
906
+ "eval_stsb_spearman": 0.8180351456581764,
907
+ "step": 1375
908
+ },
909
+ {
910
+ "epoch": 0.71,
911
+ "learning_rate": 1.4687819856704197e-05,
912
+ "loss": 0.0007,
913
+ "step": 1380
914
+ },
915
+ {
916
+ "epoch": 0.71,
917
+ "learning_rate": 1.4431934493346982e-05,
918
+ "loss": 0.0014,
919
+ "step": 1390
920
+ },
921
+ {
922
+ "epoch": 0.72,
923
+ "learning_rate": 1.4176049129989766e-05,
924
+ "loss": 0.0007,
925
+ "step": 1400
926
+ },
927
+ {
928
+ "epoch": 0.72,
929
+ "learning_rate": 1.392016376663255e-05,
930
+ "loss": 0.0007,
931
+ "step": 1410
932
+ },
933
+ {
934
+ "epoch": 0.73,
935
+ "learning_rate": 1.3664278403275333e-05,
936
+ "loss": 0.0009,
937
+ "step": 1420
938
+ },
939
+ {
940
+ "epoch": 0.73,
941
+ "learning_rate": 1.3408393039918118e-05,
942
+ "loss": 0.0002,
943
+ "step": 1430
944
+ },
945
+ {
946
+ "epoch": 0.74,
947
+ "learning_rate": 1.3152507676560901e-05,
948
+ "loss": 0.0024,
949
+ "step": 1440
950
+ },
951
+ {
952
+ "epoch": 0.74,
953
+ "learning_rate": 1.2896622313203685e-05,
954
+ "loss": 0.0005,
955
+ "step": 1450
956
+ },
957
+ {
958
+ "epoch": 0.75,
959
+ "learning_rate": 1.2640736949846468e-05,
960
+ "loss": 0.0016,
961
+ "step": 1460
962
+ },
963
+ {
964
+ "epoch": 0.75,
965
+ "learning_rate": 1.2384851586489253e-05,
966
+ "loss": 0.0007,
967
+ "step": 1470
968
+ },
969
+ {
970
+ "epoch": 0.76,
971
+ "learning_rate": 1.2128966223132038e-05,
972
+ "loss": 0.0002,
973
+ "step": 1480
974
+ },
975
+ {
976
+ "epoch": 0.76,
977
+ "learning_rate": 1.1873080859774822e-05,
978
+ "loss": 0.0004,
979
+ "step": 1490
980
+ },
981
+ {
982
+ "epoch": 0.77,
983
+ "learning_rate": 1.1617195496417605e-05,
984
+ "loss": 0.0005,
985
+ "step": 1500
986
+ },
987
+ {
988
+ "epoch": 0.77,
989
+ "eval_avg_sts": 0.7997389471860341,
990
+ "eval_sickr_spearman": 0.7761020510827888,
991
+ "eval_stsb_spearman": 0.8233758432892795,
992
+ "step": 1500
993
+ },
994
+ {
995
+ "epoch": 0.77,
996
+ "learning_rate": 1.1361310133060389e-05,
997
+ "loss": 0.0005,
998
+ "step": 1510
999
+ },
1000
+ {
1001
+ "epoch": 0.78,
1002
+ "learning_rate": 1.1105424769703174e-05,
1003
+ "loss": 0.0006,
1004
+ "step": 1520
1005
+ },
1006
+ {
1007
+ "epoch": 0.78,
1008
+ "learning_rate": 1.0849539406345957e-05,
1009
+ "loss": 0.0011,
1010
+ "step": 1530
1011
+ },
1012
+ {
1013
+ "epoch": 0.79,
1014
+ "learning_rate": 1.0593654042988742e-05,
1015
+ "loss": 0.0006,
1016
+ "step": 1540
1017
+ },
1018
+ {
1019
+ "epoch": 0.79,
1020
+ "learning_rate": 1.0337768679631526e-05,
1021
+ "loss": 0.001,
1022
+ "step": 1550
1023
+ },
1024
+ {
1025
+ "epoch": 0.8,
1026
+ "learning_rate": 1.0081883316274309e-05,
1027
+ "loss": 0.0011,
1028
+ "step": 1560
1029
+ },
1030
+ {
1031
+ "epoch": 0.8,
1032
+ "learning_rate": 9.825997952917093e-06,
1033
+ "loss": 0.001,
1034
+ "step": 1570
1035
+ },
1036
+ {
1037
+ "epoch": 0.81,
1038
+ "learning_rate": 9.570112589559878e-06,
1039
+ "loss": 0.0012,
1040
+ "step": 1580
1041
+ },
1042
+ {
1043
+ "epoch": 0.81,
1044
+ "learning_rate": 9.314227226202661e-06,
1045
+ "loss": 0.0007,
1046
+ "step": 1590
1047
+ },
1048
+ {
1049
+ "epoch": 0.82,
1050
+ "learning_rate": 9.058341862845446e-06,
1051
+ "loss": 0.001,
1052
+ "step": 1600
1053
+ },
1054
+ {
1055
+ "epoch": 0.82,
1056
+ "learning_rate": 8.80245649948823e-06,
1057
+ "loss": 0.0005,
1058
+ "step": 1610
1059
+ },
1060
+ {
1061
+ "epoch": 0.83,
1062
+ "learning_rate": 8.546571136131013e-06,
1063
+ "loss": 0.0012,
1064
+ "step": 1620
1065
+ },
1066
+ {
1067
+ "epoch": 0.83,
1068
+ "eval_avg_sts": 0.7990492556729138,
1069
+ "eval_sickr_spearman": 0.7761147312935684,
1070
+ "eval_stsb_spearman": 0.8219837800522592,
1071
+ "step": 1625
1072
+ },
1073
+ {
1074
+ "epoch": 0.83,
1075
+ "learning_rate": 8.290685772773797e-06,
1076
+ "loss": 0.0015,
1077
+ "step": 1630
1078
+ },
1079
+ {
1080
+ "epoch": 0.84,
1081
+ "learning_rate": 8.034800409416582e-06,
1082
+ "loss": 0.0015,
1083
+ "step": 1640
1084
+ },
1085
+ {
1086
+ "epoch": 0.84,
1087
+ "learning_rate": 7.778915046059365e-06,
1088
+ "loss": 0.0005,
1089
+ "step": 1650
1090
+ },
1091
+ {
1092
+ "epoch": 0.85,
1093
+ "learning_rate": 7.52302968270215e-06,
1094
+ "loss": 0.0006,
1095
+ "step": 1660
1096
+ },
1097
+ {
1098
+ "epoch": 0.85,
1099
+ "learning_rate": 7.267144319344933e-06,
1100
+ "loss": 0.0006,
1101
+ "step": 1670
1102
+ },
1103
+ {
1104
+ "epoch": 0.86,
1105
+ "learning_rate": 7.011258955987718e-06,
1106
+ "loss": 0.0016,
1107
+ "step": 1680
1108
+ },
1109
+ {
1110
+ "epoch": 0.86,
1111
+ "learning_rate": 6.755373592630501e-06,
1112
+ "loss": 0.0016,
1113
+ "step": 1690
1114
+ },
1115
+ {
1116
+ "epoch": 0.87,
1117
+ "learning_rate": 6.499488229273286e-06,
1118
+ "loss": 0.0017,
1119
+ "step": 1700
1120
+ },
1121
+ {
1122
+ "epoch": 0.88,
1123
+ "learning_rate": 6.24360286591607e-06,
1124
+ "loss": 0.0013,
1125
+ "step": 1710
1126
+ },
1127
+ {
1128
+ "epoch": 0.88,
1129
+ "learning_rate": 5.987717502558854e-06,
1130
+ "loss": 0.0009,
1131
+ "step": 1720
1132
+ },
1133
+ {
1134
+ "epoch": 0.89,
1135
+ "learning_rate": 5.731832139201638e-06,
1136
+ "loss": 0.0007,
1137
+ "step": 1730
1138
+ },
1139
+ {
1140
+ "epoch": 0.89,
1141
+ "learning_rate": 5.475946775844422e-06,
1142
+ "loss": 0.0002,
1143
+ "step": 1740
1144
+ },
1145
+ {
1146
+ "epoch": 0.9,
1147
+ "learning_rate": 5.220061412487206e-06,
1148
+ "loss": 0.0004,
1149
+ "step": 1750
1150
+ },
1151
+ {
1152
+ "epoch": 0.9,
1153
+ "eval_avg_sts": 0.7986109641804153,
1154
+ "eval_sickr_spearman": 0.7760801969316344,
1155
+ "eval_stsb_spearman": 0.8211417314291962,
1156
+ "step": 1750
1157
+ },
1158
+ {
1159
+ "epoch": 0.9,
1160
+ "learning_rate": 4.96417604912999e-06,
1161
+ "loss": 0.0008,
1162
+ "step": 1760
1163
+ },
1164
+ {
1165
+ "epoch": 0.91,
1166
+ "learning_rate": 4.708290685772774e-06,
1167
+ "loss": 0.0014,
1168
+ "step": 1770
1169
+ },
1170
+ {
1171
+ "epoch": 0.91,
1172
+ "learning_rate": 4.452405322415558e-06,
1173
+ "loss": 0.0007,
1174
+ "step": 1780
1175
+ },
1176
+ {
1177
+ "epoch": 0.92,
1178
+ "learning_rate": 4.1965199590583416e-06,
1179
+ "loss": 0.0002,
1180
+ "step": 1790
1181
+ },
1182
+ {
1183
+ "epoch": 0.92,
1184
+ "learning_rate": 3.940634595701126e-06,
1185
+ "loss": 0.001,
1186
+ "step": 1800
1187
+ },
1188
+ {
1189
+ "epoch": 0.93,
1190
+ "learning_rate": 3.68474923234391e-06,
1191
+ "loss": 0.0008,
1192
+ "step": 1810
1193
+ },
1194
+ {
1195
+ "epoch": 0.93,
1196
+ "learning_rate": 3.428863868986694e-06,
1197
+ "loss": 0.0015,
1198
+ "step": 1820
1199
+ },
1200
+ {
1201
+ "epoch": 0.94,
1202
+ "learning_rate": 3.172978505629478e-06,
1203
+ "loss": 0.0002,
1204
+ "step": 1830
1205
+ },
1206
+ {
1207
+ "epoch": 0.94,
1208
+ "learning_rate": 2.917093142272262e-06,
1209
+ "loss": 0.0011,
1210
+ "step": 1840
1211
+ },
1212
+ {
1213
+ "epoch": 0.95,
1214
+ "learning_rate": 2.6612077789150464e-06,
1215
+ "loss": 0.0005,
1216
+ "step": 1850
1217
+ },
1218
+ {
1219
+ "epoch": 0.95,
1220
+ "learning_rate": 2.4053224155578302e-06,
1221
+ "loss": 0.0006,
1222
+ "step": 1860
1223
+ },
1224
+ {
1225
+ "epoch": 0.96,
1226
+ "learning_rate": 2.149437052200614e-06,
1227
+ "loss": 0.0012,
1228
+ "step": 1870
1229
+ },
1230
+ {
1231
+ "epoch": 0.96,
1232
+ "eval_avg_sts": 0.7985301863887153,
1233
+ "eval_sickr_spearman": 0.7759786591831943,
1234
+ "eval_stsb_spearman": 0.8210817135942363,
1235
+ "step": 1875
1236
+ },
1237
+ {
1238
+ "epoch": 0.96,
1239
+ "learning_rate": 1.8935516888433984e-06,
1240
+ "loss": 0.0004,
1241
+ "step": 1880
1242
+ },
1243
+ {
1244
+ "epoch": 0.97,
1245
+ "learning_rate": 1.6376663254861822e-06,
1246
+ "loss": 0.0009,
1247
+ "step": 1890
1248
+ },
1249
+ {
1250
+ "epoch": 0.97,
1251
+ "learning_rate": 1.3817809621289663e-06,
1252
+ "loss": 0.0008,
1253
+ "step": 1900
1254
+ },
1255
+ {
1256
+ "epoch": 0.98,
1257
+ "learning_rate": 1.1258955987717504e-06,
1258
+ "loss": 0.001,
1259
+ "step": 1910
1260
+ },
1261
+ {
1262
+ "epoch": 0.98,
1263
+ "learning_rate": 8.700102354145343e-07,
1264
+ "loss": 0.0005,
1265
+ "step": 1920
1266
+ },
1267
+ {
1268
+ "epoch": 0.99,
1269
+ "learning_rate": 6.141248720573183e-07,
1270
+ "loss": 0.0002,
1271
+ "step": 1930
1272
+ },
1273
+ {
1274
+ "epoch": 0.99,
1275
+ "learning_rate": 3.5823950870010235e-07,
1276
+ "loss": 0.0005,
1277
+ "step": 1940
1278
+ },
1279
+ {
1280
+ "epoch": 1.0,
1281
+ "learning_rate": 1.0235414534288639e-07,
1282
+ "loss": 0.0004,
1283
+ "step": 1950
1284
+ },
1285
+ {
1286
+ "epoch": 1.0,
1287
+ "step": 1954,
1288
+ "train_runtime": 3451.5882,
1289
+ "train_samples_per_second": 0.566
1290
+ }
1291
+ ],
1292
+ "max_steps": 1954,
1293
+ "num_train_epochs": 1,
1294
+ "total_flos": 129098564172251136,
1295
+ "trial_name": null,
1296
+ "trial_params": null
1297
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efabd960e72b358593c2aab050363e5e82b7098586f2274261ca2d54e2dfdc6a
3
+ size 2287
vocab.txt ADDED
The diff for this file is too large to render. See raw diff