shax commited on
Commit
f7e8ea3
1 Parent(s): 971e0fc

update model

Browse files

all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_loss": 1.5331978797912598,
4
+ "eval_runtime": 4.9685,
5
+ "eval_samples": 800,
6
+ "eval_samples_per_second": 161.014,
7
+ "eval_steps_per_second": 20.127,
8
+ "perplexity": 4.632968835566632,
9
+ "train_loss": 1.7776700918673223,
10
+ "train_runtime": 8899.7428,
11
+ "train_samples": 15300,
12
+ "train_samples_per_second": 8.596,
13
+ "train_steps_per_second": 1.075
14
+ }
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "xlm-roberta-large",
3
+ "architectures": [
4
+ "XLMRobertaForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 4096,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "xlm-roberta",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 24,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.12.5",
25
+ "type_vocab_size": 1,
26
+ "use_cache": true,
27
+ "vocab_size": 250002
28
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_loss": 1.5331978797912598,
4
+ "eval_runtime": 4.9685,
5
+ "eval_samples": 800,
6
+ "eval_samples_per_second": 161.014,
7
+ "eval_steps_per_second": 20.127,
8
+ "perplexity": 4.632968835566632
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54701c69c003b2c1701683b2feac870ff3740aa70e041a5c69cfb1044207d3b7
3
+ size 2240743698
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
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
+ {"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "xlm-roberta-large", "tokenizer_class": "XLMRobertaTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "train_loss": 1.7776700918673223,
4
+ "train_runtime": 8899.7428,
5
+ "train_samples": 15300,
6
+ "train_samples_per_second": 8.596,
7
+ "train_steps_per_second": 1.075
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,839 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.0,
5
+ "global_step": 9565,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.05,
12
+ "eval_loss": 1.9894485473632812,
13
+ "eval_runtime": 4.9418,
14
+ "eval_samples_per_second": 161.884,
15
+ "eval_steps_per_second": 20.236,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.1,
20
+ "eval_loss": 1.8623613119125366,
21
+ "eval_runtime": 4.9523,
22
+ "eval_samples_per_second": 161.541,
23
+ "eval_steps_per_second": 20.193,
24
+ "step": 200
25
+ },
26
+ {
27
+ "epoch": 0.16,
28
+ "eval_loss": 1.9076178073883057,
29
+ "eval_runtime": 4.9532,
30
+ "eval_samples_per_second": 161.512,
31
+ "eval_steps_per_second": 20.189,
32
+ "step": 300
33
+ },
34
+ {
35
+ "epoch": 0.21,
36
+ "eval_loss": 1.815214991569519,
37
+ "eval_runtime": 4.9711,
38
+ "eval_samples_per_second": 160.929,
39
+ "eval_steps_per_second": 20.116,
40
+ "step": 400
41
+ },
42
+ {
43
+ "epoch": 0.26,
44
+ "eval_loss": 1.7864209413528442,
45
+ "eval_runtime": 4.9618,
46
+ "eval_samples_per_second": 161.233,
47
+ "eval_steps_per_second": 20.154,
48
+ "step": 500
49
+ },
50
+ {
51
+ "epoch": 0.31,
52
+ "eval_loss": 1.8693923950195312,
53
+ "eval_runtime": 4.9535,
54
+ "eval_samples_per_second": 161.502,
55
+ "eval_steps_per_second": 20.188,
56
+ "step": 600
57
+ },
58
+ {
59
+ "epoch": 0.37,
60
+ "eval_loss": 1.8722848892211914,
61
+ "eval_runtime": 4.9564,
62
+ "eval_samples_per_second": 161.408,
63
+ "eval_steps_per_second": 20.176,
64
+ "step": 700
65
+ },
66
+ {
67
+ "epoch": 0.42,
68
+ "eval_loss": 1.7493282556533813,
69
+ "eval_runtime": 4.9483,
70
+ "eval_samples_per_second": 161.671,
71
+ "eval_steps_per_second": 20.209,
72
+ "step": 800
73
+ },
74
+ {
75
+ "epoch": 0.47,
76
+ "eval_loss": 1.9245165586471558,
77
+ "eval_runtime": 4.9768,
78
+ "eval_samples_per_second": 160.745,
79
+ "eval_steps_per_second": 20.093,
80
+ "step": 900
81
+ },
82
+ {
83
+ "epoch": 0.52,
84
+ "learning_rate": 8.954521693674857e-06,
85
+ "loss": 2.0644,
86
+ "step": 1000
87
+ },
88
+ {
89
+ "epoch": 0.52,
90
+ "eval_loss": 1.74406099319458,
91
+ "eval_runtime": 5.0299,
92
+ "eval_samples_per_second": 159.048,
93
+ "eval_steps_per_second": 19.881,
94
+ "step": 1000
95
+ },
96
+ {
97
+ "epoch": 0.58,
98
+ "eval_loss": 1.8684289455413818,
99
+ "eval_runtime": 4.9554,
100
+ "eval_samples_per_second": 161.439,
101
+ "eval_steps_per_second": 20.18,
102
+ "step": 1100
103
+ },
104
+ {
105
+ "epoch": 0.63,
106
+ "eval_loss": 1.8639626502990723,
107
+ "eval_runtime": 4.9751,
108
+ "eval_samples_per_second": 160.802,
109
+ "eval_steps_per_second": 20.1,
110
+ "step": 1200
111
+ },
112
+ {
113
+ "epoch": 0.68,
114
+ "eval_loss": 1.829074501991272,
115
+ "eval_runtime": 4.9541,
116
+ "eval_samples_per_second": 161.482,
117
+ "eval_steps_per_second": 20.185,
118
+ "step": 1300
119
+ },
120
+ {
121
+ "epoch": 0.73,
122
+ "eval_loss": 1.806042194366455,
123
+ "eval_runtime": 4.9361,
124
+ "eval_samples_per_second": 162.072,
125
+ "eval_steps_per_second": 20.259,
126
+ "step": 1400
127
+ },
128
+ {
129
+ "epoch": 0.78,
130
+ "eval_loss": 1.8194879293441772,
131
+ "eval_runtime": 4.953,
132
+ "eval_samples_per_second": 161.517,
133
+ "eval_steps_per_second": 20.19,
134
+ "step": 1500
135
+ },
136
+ {
137
+ "epoch": 0.84,
138
+ "eval_loss": 1.7241334915161133,
139
+ "eval_runtime": 4.943,
140
+ "eval_samples_per_second": 161.845,
141
+ "eval_steps_per_second": 20.231,
142
+ "step": 1600
143
+ },
144
+ {
145
+ "epoch": 0.89,
146
+ "eval_loss": 1.8226971626281738,
147
+ "eval_runtime": 4.9746,
148
+ "eval_samples_per_second": 160.818,
149
+ "eval_steps_per_second": 20.102,
150
+ "step": 1700
151
+ },
152
+ {
153
+ "epoch": 0.94,
154
+ "eval_loss": 1.6978307962417603,
155
+ "eval_runtime": 4.9672,
156
+ "eval_samples_per_second": 161.057,
157
+ "eval_steps_per_second": 20.132,
158
+ "step": 1800
159
+ },
160
+ {
161
+ "epoch": 0.99,
162
+ "eval_loss": 1.7191663980484009,
163
+ "eval_runtime": 4.9517,
164
+ "eval_samples_per_second": 161.559,
165
+ "eval_steps_per_second": 20.195,
166
+ "step": 1900
167
+ },
168
+ {
169
+ "epoch": 1.05,
170
+ "learning_rate": 7.909043387349713e-06,
171
+ "loss": 1.9856,
172
+ "step": 2000
173
+ },
174
+ {
175
+ "epoch": 1.05,
176
+ "eval_loss": 1.7717660665512085,
177
+ "eval_runtime": 4.9947,
178
+ "eval_samples_per_second": 160.17,
179
+ "eval_steps_per_second": 20.021,
180
+ "step": 2000
181
+ },
182
+ {
183
+ "epoch": 1.1,
184
+ "eval_loss": 1.7920727729797363,
185
+ "eval_runtime": 4.9637,
186
+ "eval_samples_per_second": 161.171,
187
+ "eval_steps_per_second": 20.146,
188
+ "step": 2100
189
+ },
190
+ {
191
+ "epoch": 1.15,
192
+ "eval_loss": 1.7202229499816895,
193
+ "eval_runtime": 4.9505,
194
+ "eval_samples_per_second": 161.6,
195
+ "eval_steps_per_second": 20.2,
196
+ "step": 2200
197
+ },
198
+ {
199
+ "epoch": 1.2,
200
+ "eval_loss": 1.7953866720199585,
201
+ "eval_runtime": 4.9571,
202
+ "eval_samples_per_second": 161.384,
203
+ "eval_steps_per_second": 20.173,
204
+ "step": 2300
205
+ },
206
+ {
207
+ "epoch": 1.25,
208
+ "eval_loss": 1.7172852754592896,
209
+ "eval_runtime": 4.9472,
210
+ "eval_samples_per_second": 161.708,
211
+ "eval_steps_per_second": 20.214,
212
+ "step": 2400
213
+ },
214
+ {
215
+ "epoch": 1.31,
216
+ "eval_loss": 1.7360247373580933,
217
+ "eval_runtime": 4.9571,
218
+ "eval_samples_per_second": 161.384,
219
+ "eval_steps_per_second": 20.173,
220
+ "step": 2500
221
+ },
222
+ {
223
+ "epoch": 1.36,
224
+ "eval_loss": 1.8252145051956177,
225
+ "eval_runtime": 4.9593,
226
+ "eval_samples_per_second": 161.312,
227
+ "eval_steps_per_second": 20.164,
228
+ "step": 2600
229
+ },
230
+ {
231
+ "epoch": 1.41,
232
+ "eval_loss": 1.6163873672485352,
233
+ "eval_runtime": 4.9617,
234
+ "eval_samples_per_second": 161.234,
235
+ "eval_steps_per_second": 20.154,
236
+ "step": 2700
237
+ },
238
+ {
239
+ "epoch": 1.46,
240
+ "eval_loss": 1.7062618732452393,
241
+ "eval_runtime": 4.9606,
242
+ "eval_samples_per_second": 161.272,
243
+ "eval_steps_per_second": 20.159,
244
+ "step": 2800
245
+ },
246
+ {
247
+ "epoch": 1.52,
248
+ "eval_loss": 1.63948392868042,
249
+ "eval_runtime": 4.9584,
250
+ "eval_samples_per_second": 161.343,
251
+ "eval_steps_per_second": 20.168,
252
+ "step": 2900
253
+ },
254
+ {
255
+ "epoch": 1.57,
256
+ "learning_rate": 6.863565081024569e-06,
257
+ "loss": 1.8559,
258
+ "step": 3000
259
+ },
260
+ {
261
+ "epoch": 1.57,
262
+ "eval_loss": 1.7427096366882324,
263
+ "eval_runtime": 5.0112,
264
+ "eval_samples_per_second": 159.642,
265
+ "eval_steps_per_second": 19.955,
266
+ "step": 3000
267
+ },
268
+ {
269
+ "epoch": 1.62,
270
+ "eval_loss": 1.6909571886062622,
271
+ "eval_runtime": 4.9514,
272
+ "eval_samples_per_second": 161.571,
273
+ "eval_steps_per_second": 20.196,
274
+ "step": 3100
275
+ },
276
+ {
277
+ "epoch": 1.67,
278
+ "eval_loss": 1.7176247835159302,
279
+ "eval_runtime": 4.949,
280
+ "eval_samples_per_second": 161.648,
281
+ "eval_steps_per_second": 20.206,
282
+ "step": 3200
283
+ },
284
+ {
285
+ "epoch": 1.73,
286
+ "eval_loss": 1.6864755153656006,
287
+ "eval_runtime": 4.954,
288
+ "eval_samples_per_second": 161.487,
289
+ "eval_steps_per_second": 20.186,
290
+ "step": 3300
291
+ },
292
+ {
293
+ "epoch": 1.78,
294
+ "eval_loss": 1.7220550775527954,
295
+ "eval_runtime": 4.9485,
296
+ "eval_samples_per_second": 161.665,
297
+ "eval_steps_per_second": 20.208,
298
+ "step": 3400
299
+ },
300
+ {
301
+ "epoch": 1.83,
302
+ "eval_loss": 1.7085040807724,
303
+ "eval_runtime": 4.9559,
304
+ "eval_samples_per_second": 161.424,
305
+ "eval_steps_per_second": 20.178,
306
+ "step": 3500
307
+ },
308
+ {
309
+ "epoch": 1.88,
310
+ "eval_loss": 1.6278865337371826,
311
+ "eval_runtime": 4.9576,
312
+ "eval_samples_per_second": 161.368,
313
+ "eval_steps_per_second": 20.171,
314
+ "step": 3600
315
+ },
316
+ {
317
+ "epoch": 1.93,
318
+ "eval_loss": 1.716471552848816,
319
+ "eval_runtime": 4.9658,
320
+ "eval_samples_per_second": 161.103,
321
+ "eval_steps_per_second": 20.138,
322
+ "step": 3700
323
+ },
324
+ {
325
+ "epoch": 1.99,
326
+ "eval_loss": 1.6102560758590698,
327
+ "eval_runtime": 4.9726,
328
+ "eval_samples_per_second": 160.883,
329
+ "eval_steps_per_second": 20.11,
330
+ "step": 3800
331
+ },
332
+ {
333
+ "epoch": 2.04,
334
+ "eval_loss": 1.6070889234542847,
335
+ "eval_runtime": 4.9783,
336
+ "eval_samples_per_second": 160.697,
337
+ "eval_steps_per_second": 20.087,
338
+ "step": 3900
339
+ },
340
+ {
341
+ "epoch": 2.09,
342
+ "learning_rate": 5.8180867746994255e-06,
343
+ "loss": 1.8098,
344
+ "step": 4000
345
+ },
346
+ {
347
+ "epoch": 2.09,
348
+ "eval_loss": 1.6875723600387573,
349
+ "eval_runtime": 4.9893,
350
+ "eval_samples_per_second": 160.344,
351
+ "eval_steps_per_second": 20.043,
352
+ "step": 4000
353
+ },
354
+ {
355
+ "epoch": 2.14,
356
+ "eval_loss": 1.5920379161834717,
357
+ "eval_runtime": 4.9822,
358
+ "eval_samples_per_second": 160.571,
359
+ "eval_steps_per_second": 20.071,
360
+ "step": 4100
361
+ },
362
+ {
363
+ "epoch": 2.2,
364
+ "eval_loss": 1.6611220836639404,
365
+ "eval_runtime": 4.9604,
366
+ "eval_samples_per_second": 161.279,
367
+ "eval_steps_per_second": 20.16,
368
+ "step": 4200
369
+ },
370
+ {
371
+ "epoch": 2.25,
372
+ "eval_loss": 1.6467419862747192,
373
+ "eval_runtime": 5.0216,
374
+ "eval_samples_per_second": 159.312,
375
+ "eval_steps_per_second": 19.914,
376
+ "step": 4300
377
+ },
378
+ {
379
+ "epoch": 2.3,
380
+ "eval_loss": 1.6913862228393555,
381
+ "eval_runtime": 4.9754,
382
+ "eval_samples_per_second": 160.791,
383
+ "eval_steps_per_second": 20.099,
384
+ "step": 4400
385
+ },
386
+ {
387
+ "epoch": 2.35,
388
+ "eval_loss": 1.7248626947402954,
389
+ "eval_runtime": 4.9681,
390
+ "eval_samples_per_second": 161.027,
391
+ "eval_steps_per_second": 20.128,
392
+ "step": 4500
393
+ },
394
+ {
395
+ "epoch": 2.4,
396
+ "eval_loss": 1.668603539466858,
397
+ "eval_runtime": 4.9751,
398
+ "eval_samples_per_second": 160.799,
399
+ "eval_steps_per_second": 20.1,
400
+ "step": 4600
401
+ },
402
+ {
403
+ "epoch": 2.46,
404
+ "eval_loss": 1.571094036102295,
405
+ "eval_runtime": 4.9862,
406
+ "eval_samples_per_second": 160.444,
407
+ "eval_steps_per_second": 20.055,
408
+ "step": 4700
409
+ },
410
+ {
411
+ "epoch": 2.51,
412
+ "eval_loss": 1.6627862453460693,
413
+ "eval_runtime": 4.9621,
414
+ "eval_samples_per_second": 161.224,
415
+ "eval_steps_per_second": 20.153,
416
+ "step": 4800
417
+ },
418
+ {
419
+ "epoch": 2.56,
420
+ "eval_loss": 1.5440031290054321,
421
+ "eval_runtime": 4.9714,
422
+ "eval_samples_per_second": 160.921,
423
+ "eval_steps_per_second": 20.115,
424
+ "step": 4900
425
+ },
426
+ {
427
+ "epoch": 2.61,
428
+ "learning_rate": 4.7726084683742815e-06,
429
+ "loss": 1.7335,
430
+ "step": 5000
431
+ },
432
+ {
433
+ "epoch": 2.61,
434
+ "eval_loss": 1.5720436573028564,
435
+ "eval_runtime": 5.0224,
436
+ "eval_samples_per_second": 159.285,
437
+ "eval_steps_per_second": 19.911,
438
+ "step": 5000
439
+ },
440
+ {
441
+ "epoch": 2.67,
442
+ "eval_loss": 1.6262255907058716,
443
+ "eval_runtime": 4.9733,
444
+ "eval_samples_per_second": 160.86,
445
+ "eval_steps_per_second": 20.108,
446
+ "step": 5100
447
+ },
448
+ {
449
+ "epoch": 2.72,
450
+ "eval_loss": 1.579461693763733,
451
+ "eval_runtime": 4.9655,
452
+ "eval_samples_per_second": 161.111,
453
+ "eval_steps_per_second": 20.139,
454
+ "step": 5200
455
+ },
456
+ {
457
+ "epoch": 2.77,
458
+ "eval_loss": 1.64106023311615,
459
+ "eval_runtime": 4.9697,
460
+ "eval_samples_per_second": 160.974,
461
+ "eval_steps_per_second": 20.122,
462
+ "step": 5300
463
+ },
464
+ {
465
+ "epoch": 2.82,
466
+ "eval_loss": 1.6266947984695435,
467
+ "eval_runtime": 4.9845,
468
+ "eval_samples_per_second": 160.496,
469
+ "eval_steps_per_second": 20.062,
470
+ "step": 5400
471
+ },
472
+ {
473
+ "epoch": 2.88,
474
+ "eval_loss": 1.5381076335906982,
475
+ "eval_runtime": 4.9925,
476
+ "eval_samples_per_second": 160.241,
477
+ "eval_steps_per_second": 20.03,
478
+ "step": 5500
479
+ },
480
+ {
481
+ "epoch": 2.93,
482
+ "eval_loss": 1.6071343421936035,
483
+ "eval_runtime": 4.9709,
484
+ "eval_samples_per_second": 160.937,
485
+ "eval_steps_per_second": 20.117,
486
+ "step": 5600
487
+ },
488
+ {
489
+ "epoch": 2.98,
490
+ "eval_loss": 1.57161283493042,
491
+ "eval_runtime": 4.984,
492
+ "eval_samples_per_second": 160.515,
493
+ "eval_steps_per_second": 20.064,
494
+ "step": 5700
495
+ },
496
+ {
497
+ "epoch": 3.03,
498
+ "eval_loss": 1.6507529020309448,
499
+ "eval_runtime": 4.9769,
500
+ "eval_samples_per_second": 160.744,
501
+ "eval_steps_per_second": 20.093,
502
+ "step": 5800
503
+ },
504
+ {
505
+ "epoch": 3.08,
506
+ "eval_loss": 1.6694835424423218,
507
+ "eval_runtime": 4.9622,
508
+ "eval_samples_per_second": 161.22,
509
+ "eval_steps_per_second": 20.153,
510
+ "step": 5900
511
+ },
512
+ {
513
+ "epoch": 3.14,
514
+ "learning_rate": 3.727130162049138e-06,
515
+ "loss": 1.7197,
516
+ "step": 6000
517
+ },
518
+ {
519
+ "epoch": 3.14,
520
+ "eval_loss": 1.5861258506774902,
521
+ "eval_runtime": 5.0223,
522
+ "eval_samples_per_second": 159.29,
523
+ "eval_steps_per_second": 19.911,
524
+ "step": 6000
525
+ },
526
+ {
527
+ "epoch": 3.19,
528
+ "eval_loss": 1.6594449281692505,
529
+ "eval_runtime": 4.9785,
530
+ "eval_samples_per_second": 160.692,
531
+ "eval_steps_per_second": 20.087,
532
+ "step": 6100
533
+ },
534
+ {
535
+ "epoch": 3.24,
536
+ "eval_loss": 1.5593453645706177,
537
+ "eval_runtime": 4.9715,
538
+ "eval_samples_per_second": 160.917,
539
+ "eval_steps_per_second": 20.115,
540
+ "step": 6200
541
+ },
542
+ {
543
+ "epoch": 3.29,
544
+ "eval_loss": 1.5538159608840942,
545
+ "eval_runtime": 4.9601,
546
+ "eval_samples_per_second": 161.285,
547
+ "eval_steps_per_second": 20.161,
548
+ "step": 6300
549
+ },
550
+ {
551
+ "epoch": 3.35,
552
+ "eval_loss": 1.626267671585083,
553
+ "eval_runtime": 4.9624,
554
+ "eval_samples_per_second": 161.213,
555
+ "eval_steps_per_second": 20.152,
556
+ "step": 6400
557
+ },
558
+ {
559
+ "epoch": 3.4,
560
+ "eval_loss": 1.5642911195755005,
561
+ "eval_runtime": 4.9736,
562
+ "eval_samples_per_second": 160.85,
563
+ "eval_steps_per_second": 20.106,
564
+ "step": 6500
565
+ },
566
+ {
567
+ "epoch": 3.45,
568
+ "eval_loss": 1.610222578048706,
569
+ "eval_runtime": 4.9769,
570
+ "eval_samples_per_second": 160.743,
571
+ "eval_steps_per_second": 20.093,
572
+ "step": 6600
573
+ },
574
+ {
575
+ "epoch": 3.5,
576
+ "eval_loss": 1.5607407093048096,
577
+ "eval_runtime": 4.9764,
578
+ "eval_samples_per_second": 160.76,
579
+ "eval_steps_per_second": 20.095,
580
+ "step": 6700
581
+ },
582
+ {
583
+ "epoch": 3.55,
584
+ "eval_loss": 1.4866421222686768,
585
+ "eval_runtime": 4.9715,
586
+ "eval_samples_per_second": 160.918,
587
+ "eval_steps_per_second": 20.115,
588
+ "step": 6800
589
+ },
590
+ {
591
+ "epoch": 3.61,
592
+ "eval_loss": 1.6732252836227417,
593
+ "eval_runtime": 4.9779,
594
+ "eval_samples_per_second": 160.711,
595
+ "eval_steps_per_second": 20.089,
596
+ "step": 6900
597
+ },
598
+ {
599
+ "epoch": 3.66,
600
+ "learning_rate": 2.681651855723994e-06,
601
+ "loss": 1.6777,
602
+ "step": 7000
603
+ },
604
+ {
605
+ "epoch": 3.66,
606
+ "eval_loss": 1.6289987564086914,
607
+ "eval_runtime": 5.0056,
608
+ "eval_samples_per_second": 159.819,
609
+ "eval_steps_per_second": 19.977,
610
+ "step": 7000
611
+ },
612
+ {
613
+ "epoch": 3.71,
614
+ "eval_loss": 1.62163507938385,
615
+ "eval_runtime": 4.9678,
616
+ "eval_samples_per_second": 161.037,
617
+ "eval_steps_per_second": 20.13,
618
+ "step": 7100
619
+ },
620
+ {
621
+ "epoch": 3.76,
622
+ "eval_loss": 1.5280898809432983,
623
+ "eval_runtime": 4.9766,
624
+ "eval_samples_per_second": 160.752,
625
+ "eval_steps_per_second": 20.094,
626
+ "step": 7200
627
+ },
628
+ {
629
+ "epoch": 3.82,
630
+ "eval_loss": 1.5369991064071655,
631
+ "eval_runtime": 4.9693,
632
+ "eval_samples_per_second": 160.988,
633
+ "eval_steps_per_second": 20.123,
634
+ "step": 7300
635
+ },
636
+ {
637
+ "epoch": 3.87,
638
+ "eval_loss": 1.4636890888214111,
639
+ "eval_runtime": 4.9847,
640
+ "eval_samples_per_second": 160.491,
641
+ "eval_steps_per_second": 20.061,
642
+ "step": 7400
643
+ },
644
+ {
645
+ "epoch": 3.92,
646
+ "eval_loss": 1.5771527290344238,
647
+ "eval_runtime": 4.969,
648
+ "eval_samples_per_second": 160.999,
649
+ "eval_steps_per_second": 20.125,
650
+ "step": 7500
651
+ },
652
+ {
653
+ "epoch": 3.97,
654
+ "eval_loss": 1.5966529846191406,
655
+ "eval_runtime": 4.9711,
656
+ "eval_samples_per_second": 160.931,
657
+ "eval_steps_per_second": 20.116,
658
+ "step": 7600
659
+ },
660
+ {
661
+ "epoch": 4.03,
662
+ "eval_loss": 1.5244237184524536,
663
+ "eval_runtime": 4.9746,
664
+ "eval_samples_per_second": 160.816,
665
+ "eval_steps_per_second": 20.102,
666
+ "step": 7700
667
+ },
668
+ {
669
+ "epoch": 4.08,
670
+ "eval_loss": 1.5681614875793457,
671
+ "eval_runtime": 4.9878,
672
+ "eval_samples_per_second": 160.392,
673
+ "eval_steps_per_second": 20.049,
674
+ "step": 7800
675
+ },
676
+ {
677
+ "epoch": 4.13,
678
+ "eval_loss": 1.565970778465271,
679
+ "eval_runtime": 4.9728,
680
+ "eval_samples_per_second": 160.874,
681
+ "eval_steps_per_second": 20.109,
682
+ "step": 7900
683
+ },
684
+ {
685
+ "epoch": 4.18,
686
+ "learning_rate": 1.63617354939885e-06,
687
+ "loss": 1.6225,
688
+ "step": 8000
689
+ },
690
+ {
691
+ "epoch": 4.18,
692
+ "eval_loss": 1.539146065711975,
693
+ "eval_runtime": 5.023,
694
+ "eval_samples_per_second": 159.266,
695
+ "eval_steps_per_second": 19.908,
696
+ "step": 8000
697
+ },
698
+ {
699
+ "epoch": 4.23,
700
+ "eval_loss": 1.6028637886047363,
701
+ "eval_runtime": 4.9743,
702
+ "eval_samples_per_second": 160.826,
703
+ "eval_steps_per_second": 20.103,
704
+ "step": 8100
705
+ },
706
+ {
707
+ "epoch": 4.29,
708
+ "eval_loss": 1.5656570196151733,
709
+ "eval_runtime": 4.9679,
710
+ "eval_samples_per_second": 161.033,
711
+ "eval_steps_per_second": 20.129,
712
+ "step": 8200
713
+ },
714
+ {
715
+ "epoch": 4.34,
716
+ "eval_loss": 1.5960673093795776,
717
+ "eval_runtime": 5.0005,
718
+ "eval_samples_per_second": 159.985,
719
+ "eval_steps_per_second": 19.998,
720
+ "step": 8300
721
+ },
722
+ {
723
+ "epoch": 4.39,
724
+ "eval_loss": 1.6787633895874023,
725
+ "eval_runtime": 4.9828,
726
+ "eval_samples_per_second": 160.552,
727
+ "eval_steps_per_second": 20.069,
728
+ "step": 8400
729
+ },
730
+ {
731
+ "epoch": 4.44,
732
+ "eval_loss": 1.6510525941848755,
733
+ "eval_runtime": 4.9935,
734
+ "eval_samples_per_second": 160.209,
735
+ "eval_steps_per_second": 20.026,
736
+ "step": 8500
737
+ },
738
+ {
739
+ "epoch": 4.5,
740
+ "eval_loss": 1.5883086919784546,
741
+ "eval_runtime": 4.9809,
742
+ "eval_samples_per_second": 160.614,
743
+ "eval_steps_per_second": 20.077,
744
+ "step": 8600
745
+ },
746
+ {
747
+ "epoch": 4.55,
748
+ "eval_loss": 1.502740502357483,
749
+ "eval_runtime": 4.9768,
750
+ "eval_samples_per_second": 160.746,
751
+ "eval_steps_per_second": 20.093,
752
+ "step": 8700
753
+ },
754
+ {
755
+ "epoch": 4.6,
756
+ "eval_loss": 1.6155657768249512,
757
+ "eval_runtime": 4.9737,
758
+ "eval_samples_per_second": 160.848,
759
+ "eval_steps_per_second": 20.106,
760
+ "step": 8800
761
+ },
762
+ {
763
+ "epoch": 4.65,
764
+ "eval_loss": 1.5786443948745728,
765
+ "eval_runtime": 4.9685,
766
+ "eval_samples_per_second": 161.015,
767
+ "eval_steps_per_second": 20.127,
768
+ "step": 8900
769
+ },
770
+ {
771
+ "epoch": 4.7,
772
+ "learning_rate": 5.906952430737063e-07,
773
+ "loss": 1.6198,
774
+ "step": 9000
775
+ },
776
+ {
777
+ "epoch": 4.7,
778
+ "eval_loss": 1.5171962976455688,
779
+ "eval_runtime": 5.0043,
780
+ "eval_samples_per_second": 159.861,
781
+ "eval_steps_per_second": 19.983,
782
+ "step": 9000
783
+ },
784
+ {
785
+ "epoch": 4.76,
786
+ "eval_loss": 1.5533108711242676,
787
+ "eval_runtime": 4.976,
788
+ "eval_samples_per_second": 160.773,
789
+ "eval_steps_per_second": 20.097,
790
+ "step": 9100
791
+ },
792
+ {
793
+ "epoch": 4.81,
794
+ "eval_loss": 1.4648842811584473,
795
+ "eval_runtime": 4.9677,
796
+ "eval_samples_per_second": 161.04,
797
+ "eval_steps_per_second": 20.13,
798
+ "step": 9200
799
+ },
800
+ {
801
+ "epoch": 4.86,
802
+ "eval_loss": 1.5831387042999268,
803
+ "eval_runtime": 4.9721,
804
+ "eval_samples_per_second": 160.898,
805
+ "eval_steps_per_second": 20.112,
806
+ "step": 9300
807
+ },
808
+ {
809
+ "epoch": 4.91,
810
+ "eval_loss": 1.524021029472351,
811
+ "eval_runtime": 4.9684,
812
+ "eval_samples_per_second": 161.017,
813
+ "eval_steps_per_second": 20.127,
814
+ "step": 9400
815
+ },
816
+ {
817
+ "epoch": 4.97,
818
+ "eval_loss": 1.6083089113235474,
819
+ "eval_runtime": 4.9658,
820
+ "eval_samples_per_second": 161.103,
821
+ "eval_steps_per_second": 20.138,
822
+ "step": 9500
823
+ },
824
+ {
825
+ "epoch": 5.0,
826
+ "step": 9565,
827
+ "total_flos": 6408621045709344.0,
828
+ "train_loss": 1.7776700918673223,
829
+ "train_runtime": 8899.7428,
830
+ "train_samples_per_second": 8.596,
831
+ "train_steps_per_second": 1.075
832
+ }
833
+ ],
834
+ "max_steps": 9565,
835
+ "num_train_epochs": 5,
836
+ "total_flos": 6408621045709344.0,
837
+ "trial_name": null,
838
+ "trial_params": null
839
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd20efea017d0d9aa32366c11536782342004bb9ad9ca8ecebec65d43372a3bd
3
+ size 2927