yhshu commited on
Commit
36b3f69
1 Parent(s): 3362464

Upload 4 files

Browse files
Files changed (4) hide show
  1. config.json +60 -0
  2. pytorch_model.bin +3 -0
  3. trainer_state.json +992 -0
  4. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3072,
7
+ "d_kv": 64,
8
+ "d_model": 768,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "relu",
11
+ "dropout_rate": 0.1,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "relu",
14
+ "initializer_factor": 1.0,
15
+ "is_encoder_decoder": true,
16
+ "is_gated_act": false,
17
+ "layer_norm_epsilon": 1e-06,
18
+ "model_type": "t5",
19
+ "n_positions": 512,
20
+ "num_decoder_layers": 12,
21
+ "num_heads": 12,
22
+ "num_layers": 12,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "task_specific_params": {
28
+ "summarization": {
29
+ "early_stopping": true,
30
+ "length_penalty": 2.0,
31
+ "max_length": 200,
32
+ "min_length": 30,
33
+ "no_repeat_ngram_size": 3,
34
+ "num_beams": 4,
35
+ "prefix": "summarize: "
36
+ },
37
+ "translation_en_to_de": {
38
+ "early_stopping": true,
39
+ "max_length": 300,
40
+ "num_beams": 4,
41
+ "prefix": "translate English to German: "
42
+ },
43
+ "translation_en_to_fr": {
44
+ "early_stopping": true,
45
+ "max_length": 300,
46
+ "num_beams": 4,
47
+ "prefix": "translate English to French: "
48
+ },
49
+ "translation_en_to_ro": {
50
+ "early_stopping": true,
51
+ "max_length": 300,
52
+ "num_beams": 4,
53
+ "prefix": "translate English to Romanian: "
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.21.3",
58
+ "use_cache": true,
59
+ "vocab_size": 32128
60
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52ce4074bad5f274c6918148908f47026c9ef7835a2162fcd61d94e68a498af7
3
+ size 891700799
trainer_state.json ADDED
@@ -0,0 +1,992 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 8.0,
5
+ "global_step": 76416,
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
+ "learning_rate": 2.9803706030150755e-05,
13
+ "loss": 1.2294,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.1,
18
+ "learning_rate": 2.960741206030151e-05,
19
+ "loss": 0.9625,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.16,
24
+ "learning_rate": 2.9411118090452263e-05,
25
+ "loss": 0.9145,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.21,
30
+ "learning_rate": 2.9214824120603013e-05,
31
+ "loss": 0.8595,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.26,
36
+ "learning_rate": 2.901853015075377e-05,
37
+ "loss": 0.8622,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.31,
42
+ "learning_rate": 2.8822236180904525e-05,
43
+ "loss": 0.8504,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.37,
48
+ "learning_rate": 2.8625942211055275e-05,
49
+ "loss": 0.8308,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.42,
54
+ "learning_rate": 2.842964824120603e-05,
55
+ "loss": 0.8292,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.47,
60
+ "learning_rate": 2.8233354271356787e-05,
61
+ "loss": 0.8085,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.52,
66
+ "learning_rate": 2.8037060301507537e-05,
67
+ "loss": 0.7914,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.58,
72
+ "learning_rate": 2.784076633165829e-05,
73
+ "loss": 0.8014,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.63,
78
+ "learning_rate": 2.7644472361809045e-05,
79
+ "loss": 0.7876,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.68,
84
+ "learning_rate": 2.74481783919598e-05,
85
+ "loss": 0.7936,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.73,
90
+ "learning_rate": 2.7251884422110553e-05,
91
+ "loss": 0.7975,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.79,
96
+ "learning_rate": 2.7055590452261307e-05,
97
+ "loss": 0.7748,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.84,
102
+ "learning_rate": 2.685929648241206e-05,
103
+ "loss": 0.7804,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.89,
108
+ "learning_rate": 2.6663002512562815e-05,
109
+ "loss": 0.773,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.94,
114
+ "learning_rate": 2.646670854271357e-05,
115
+ "loss": 0.7652,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.99,
120
+ "learning_rate": 2.6270414572864323e-05,
121
+ "loss": 0.7727,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 1.0,
126
+ "eval_loss": 0.8350390195846558,
127
+ "eval_runtime": 8.4552,
128
+ "eval_samples_per_second": 1312.323,
129
+ "eval_steps_per_second": 41.04,
130
+ "step": 9552
131
+ },
132
+ {
133
+ "epoch": 1.05,
134
+ "learning_rate": 2.6074120603015074e-05,
135
+ "loss": 0.7397,
136
+ "step": 10000
137
+ },
138
+ {
139
+ "epoch": 1.1,
140
+ "learning_rate": 2.587782663316583e-05,
141
+ "loss": 0.7335,
142
+ "step": 10500
143
+ },
144
+ {
145
+ "epoch": 1.15,
146
+ "learning_rate": 2.5681532663316585e-05,
147
+ "loss": 0.7095,
148
+ "step": 11000
149
+ },
150
+ {
151
+ "epoch": 1.2,
152
+ "learning_rate": 2.5485238693467336e-05,
153
+ "loss": 0.7407,
154
+ "step": 11500
155
+ },
156
+ {
157
+ "epoch": 1.26,
158
+ "learning_rate": 2.528894472361809e-05,
159
+ "loss": 0.7235,
160
+ "step": 12000
161
+ },
162
+ {
163
+ "epoch": 1.31,
164
+ "learning_rate": 2.5092650753768844e-05,
165
+ "loss": 0.709,
166
+ "step": 12500
167
+ },
168
+ {
169
+ "epoch": 1.36,
170
+ "learning_rate": 2.4896356783919598e-05,
171
+ "loss": 0.7161,
172
+ "step": 13000
173
+ },
174
+ {
175
+ "epoch": 1.41,
176
+ "learning_rate": 2.4700062814070352e-05,
177
+ "loss": 0.7278,
178
+ "step": 13500
179
+ },
180
+ {
181
+ "epoch": 1.47,
182
+ "learning_rate": 2.4503768844221106e-05,
183
+ "loss": 0.74,
184
+ "step": 14000
185
+ },
186
+ {
187
+ "epoch": 1.52,
188
+ "learning_rate": 2.430747487437186e-05,
189
+ "loss": 0.7323,
190
+ "step": 14500
191
+ },
192
+ {
193
+ "epoch": 1.57,
194
+ "learning_rate": 2.4111180904522614e-05,
195
+ "loss": 0.7298,
196
+ "step": 15000
197
+ },
198
+ {
199
+ "epoch": 1.62,
200
+ "learning_rate": 2.3914886934673368e-05,
201
+ "loss": 0.7394,
202
+ "step": 15500
203
+ },
204
+ {
205
+ "epoch": 1.68,
206
+ "learning_rate": 2.3718592964824122e-05,
207
+ "loss": 0.7393,
208
+ "step": 16000
209
+ },
210
+ {
211
+ "epoch": 1.73,
212
+ "learning_rate": 2.3522298994974873e-05,
213
+ "loss": 0.715,
214
+ "step": 16500
215
+ },
216
+ {
217
+ "epoch": 1.78,
218
+ "learning_rate": 2.332600502512563e-05,
219
+ "loss": 0.7139,
220
+ "step": 17000
221
+ },
222
+ {
223
+ "epoch": 1.83,
224
+ "learning_rate": 2.3129711055276384e-05,
225
+ "loss": 0.7215,
226
+ "step": 17500
227
+ },
228
+ {
229
+ "epoch": 1.88,
230
+ "learning_rate": 2.2933417085427135e-05,
231
+ "loss": 0.7122,
232
+ "step": 18000
233
+ },
234
+ {
235
+ "epoch": 1.94,
236
+ "learning_rate": 2.273712311557789e-05,
237
+ "loss": 0.7119,
238
+ "step": 18500
239
+ },
240
+ {
241
+ "epoch": 1.99,
242
+ "learning_rate": 2.2540829145728646e-05,
243
+ "loss": 0.7113,
244
+ "step": 19000
245
+ },
246
+ {
247
+ "epoch": 2.0,
248
+ "eval_loss": 0.8123675584793091,
249
+ "eval_runtime": 8.2517,
250
+ "eval_samples_per_second": 1344.695,
251
+ "eval_steps_per_second": 42.052,
252
+ "step": 19104
253
+ },
254
+ {
255
+ "epoch": 2.04,
256
+ "learning_rate": 2.2344535175879397e-05,
257
+ "loss": 0.6898,
258
+ "step": 19500
259
+ },
260
+ {
261
+ "epoch": 2.09,
262
+ "learning_rate": 2.214824120603015e-05,
263
+ "loss": 0.6819,
264
+ "step": 20000
265
+ },
266
+ {
267
+ "epoch": 2.15,
268
+ "learning_rate": 2.1951947236180905e-05,
269
+ "loss": 0.6797,
270
+ "step": 20500
271
+ },
272
+ {
273
+ "epoch": 2.2,
274
+ "learning_rate": 2.175565326633166e-05,
275
+ "loss": 0.6808,
276
+ "step": 21000
277
+ },
278
+ {
279
+ "epoch": 2.25,
280
+ "learning_rate": 2.1559359296482413e-05,
281
+ "loss": 0.6949,
282
+ "step": 21500
283
+ },
284
+ {
285
+ "epoch": 2.3,
286
+ "learning_rate": 2.1363065326633167e-05,
287
+ "loss": 0.689,
288
+ "step": 22000
289
+ },
290
+ {
291
+ "epoch": 2.36,
292
+ "learning_rate": 2.116677135678392e-05,
293
+ "loss": 0.6993,
294
+ "step": 22500
295
+ },
296
+ {
297
+ "epoch": 2.41,
298
+ "learning_rate": 2.0970477386934675e-05,
299
+ "loss": 0.6795,
300
+ "step": 23000
301
+ },
302
+ {
303
+ "epoch": 2.46,
304
+ "learning_rate": 2.077418341708543e-05,
305
+ "loss": 0.6972,
306
+ "step": 23500
307
+ },
308
+ {
309
+ "epoch": 2.51,
310
+ "learning_rate": 2.0577889447236183e-05,
311
+ "loss": 0.6831,
312
+ "step": 24000
313
+ },
314
+ {
315
+ "epoch": 2.56,
316
+ "learning_rate": 2.0381595477386933e-05,
317
+ "loss": 0.678,
318
+ "step": 24500
319
+ },
320
+ {
321
+ "epoch": 2.62,
322
+ "learning_rate": 2.0185301507537687e-05,
323
+ "loss": 0.6871,
324
+ "step": 25000
325
+ },
326
+ {
327
+ "epoch": 2.67,
328
+ "learning_rate": 1.9989007537688445e-05,
329
+ "loss": 0.6793,
330
+ "step": 25500
331
+ },
332
+ {
333
+ "epoch": 2.72,
334
+ "learning_rate": 1.9792713567839195e-05,
335
+ "loss": 0.6808,
336
+ "step": 26000
337
+ },
338
+ {
339
+ "epoch": 2.77,
340
+ "learning_rate": 1.959641959798995e-05,
341
+ "loss": 0.6894,
342
+ "step": 26500
343
+ },
344
+ {
345
+ "epoch": 2.83,
346
+ "learning_rate": 1.9400125628140703e-05,
347
+ "loss": 0.6876,
348
+ "step": 27000
349
+ },
350
+ {
351
+ "epoch": 2.88,
352
+ "learning_rate": 1.9203831658291457e-05,
353
+ "loss": 0.6919,
354
+ "step": 27500
355
+ },
356
+ {
357
+ "epoch": 2.93,
358
+ "learning_rate": 1.900753768844221e-05,
359
+ "loss": 0.6856,
360
+ "step": 28000
361
+ },
362
+ {
363
+ "epoch": 2.98,
364
+ "learning_rate": 1.8811243718592965e-05,
365
+ "loss": 0.6708,
366
+ "step": 28500
367
+ },
368
+ {
369
+ "epoch": 3.0,
370
+ "eval_loss": 0.7998350262641907,
371
+ "eval_runtime": 8.2332,
372
+ "eval_samples_per_second": 1347.713,
373
+ "eval_steps_per_second": 42.146,
374
+ "step": 28656
375
+ },
376
+ {
377
+ "epoch": 3.04,
378
+ "learning_rate": 1.861494974874372e-05,
379
+ "loss": 0.6758,
380
+ "step": 29000
381
+ },
382
+ {
383
+ "epoch": 3.09,
384
+ "learning_rate": 1.8418655778894473e-05,
385
+ "loss": 0.6638,
386
+ "step": 29500
387
+ },
388
+ {
389
+ "epoch": 3.14,
390
+ "learning_rate": 1.8222361809045227e-05,
391
+ "loss": 0.654,
392
+ "step": 30000
393
+ },
394
+ {
395
+ "epoch": 3.19,
396
+ "learning_rate": 1.802606783919598e-05,
397
+ "loss": 0.6792,
398
+ "step": 30500
399
+ },
400
+ {
401
+ "epoch": 3.25,
402
+ "learning_rate": 1.7829773869346732e-05,
403
+ "loss": 0.6999,
404
+ "step": 31000
405
+ },
406
+ {
407
+ "epoch": 3.3,
408
+ "learning_rate": 1.763347989949749e-05,
409
+ "loss": 0.6834,
410
+ "step": 31500
411
+ },
412
+ {
413
+ "epoch": 3.35,
414
+ "learning_rate": 1.7437185929648243e-05,
415
+ "loss": 0.647,
416
+ "step": 32000
417
+ },
418
+ {
419
+ "epoch": 3.4,
420
+ "learning_rate": 1.7240891959798994e-05,
421
+ "loss": 0.662,
422
+ "step": 32500
423
+ },
424
+ {
425
+ "epoch": 3.45,
426
+ "learning_rate": 1.7044597989949748e-05,
427
+ "loss": 0.6674,
428
+ "step": 33000
429
+ },
430
+ {
431
+ "epoch": 3.51,
432
+ "learning_rate": 1.6848304020100505e-05,
433
+ "loss": 0.6409,
434
+ "step": 33500
435
+ },
436
+ {
437
+ "epoch": 3.56,
438
+ "learning_rate": 1.6652010050251256e-05,
439
+ "loss": 0.6766,
440
+ "step": 34000
441
+ },
442
+ {
443
+ "epoch": 3.61,
444
+ "learning_rate": 1.645571608040201e-05,
445
+ "loss": 0.6658,
446
+ "step": 34500
447
+ },
448
+ {
449
+ "epoch": 3.66,
450
+ "learning_rate": 1.6259422110552764e-05,
451
+ "loss": 0.6765,
452
+ "step": 35000
453
+ },
454
+ {
455
+ "epoch": 3.72,
456
+ "learning_rate": 1.6063128140703518e-05,
457
+ "loss": 0.6478,
458
+ "step": 35500
459
+ },
460
+ {
461
+ "epoch": 3.77,
462
+ "learning_rate": 1.5866834170854272e-05,
463
+ "loss": 0.6604,
464
+ "step": 36000
465
+ },
466
+ {
467
+ "epoch": 3.82,
468
+ "learning_rate": 1.5670540201005026e-05,
469
+ "loss": 0.6398,
470
+ "step": 36500
471
+ },
472
+ {
473
+ "epoch": 3.87,
474
+ "learning_rate": 1.547424623115578e-05,
475
+ "loss": 0.6554,
476
+ "step": 37000
477
+ },
478
+ {
479
+ "epoch": 3.93,
480
+ "learning_rate": 1.527795226130653e-05,
481
+ "loss": 0.6442,
482
+ "step": 37500
483
+ },
484
+ {
485
+ "epoch": 3.98,
486
+ "learning_rate": 1.5081658291457286e-05,
487
+ "loss": 0.6552,
488
+ "step": 38000
489
+ },
490
+ {
491
+ "epoch": 4.0,
492
+ "eval_loss": 0.793069064617157,
493
+ "eval_runtime": 8.0925,
494
+ "eval_samples_per_second": 1371.141,
495
+ "eval_steps_per_second": 42.879,
496
+ "step": 38208
497
+ },
498
+ {
499
+ "epoch": 4.03,
500
+ "learning_rate": 1.488536432160804e-05,
501
+ "loss": 0.6526,
502
+ "step": 38500
503
+ },
504
+ {
505
+ "epoch": 4.08,
506
+ "learning_rate": 1.4689070351758794e-05,
507
+ "loss": 0.656,
508
+ "step": 39000
509
+ },
510
+ {
511
+ "epoch": 4.14,
512
+ "learning_rate": 1.4492776381909548e-05,
513
+ "loss": 0.6499,
514
+ "step": 39500
515
+ },
516
+ {
517
+ "epoch": 4.19,
518
+ "learning_rate": 1.4296482412060302e-05,
519
+ "loss": 0.6319,
520
+ "step": 40000
521
+ },
522
+ {
523
+ "epoch": 4.24,
524
+ "learning_rate": 1.4100188442211055e-05,
525
+ "loss": 0.6511,
526
+ "step": 40500
527
+ },
528
+ {
529
+ "epoch": 4.29,
530
+ "learning_rate": 1.390389447236181e-05,
531
+ "loss": 0.6461,
532
+ "step": 41000
533
+ },
534
+ {
535
+ "epoch": 4.34,
536
+ "learning_rate": 1.3707600502512563e-05,
537
+ "loss": 0.6472,
538
+ "step": 41500
539
+ },
540
+ {
541
+ "epoch": 4.4,
542
+ "learning_rate": 1.3511306532663317e-05,
543
+ "loss": 0.6334,
544
+ "step": 42000
545
+ },
546
+ {
547
+ "epoch": 4.45,
548
+ "learning_rate": 1.331501256281407e-05,
549
+ "loss": 0.6302,
550
+ "step": 42500
551
+ },
552
+ {
553
+ "epoch": 4.5,
554
+ "learning_rate": 1.3118718592964825e-05,
555
+ "loss": 0.6584,
556
+ "step": 43000
557
+ },
558
+ {
559
+ "epoch": 4.55,
560
+ "learning_rate": 1.2922424623115579e-05,
561
+ "loss": 0.635,
562
+ "step": 43500
563
+ },
564
+ {
565
+ "epoch": 4.61,
566
+ "learning_rate": 1.2726130653266331e-05,
567
+ "loss": 0.6305,
568
+ "step": 44000
569
+ },
570
+ {
571
+ "epoch": 4.66,
572
+ "learning_rate": 1.2529836683417085e-05,
573
+ "loss": 0.6448,
574
+ "step": 44500
575
+ },
576
+ {
577
+ "epoch": 4.71,
578
+ "learning_rate": 1.233354271356784e-05,
579
+ "loss": 0.6546,
580
+ "step": 45000
581
+ },
582
+ {
583
+ "epoch": 4.76,
584
+ "learning_rate": 1.2137248743718593e-05,
585
+ "loss": 0.6426,
586
+ "step": 45500
587
+ },
588
+ {
589
+ "epoch": 4.82,
590
+ "learning_rate": 1.1940954773869347e-05,
591
+ "loss": 0.6405,
592
+ "step": 46000
593
+ },
594
+ {
595
+ "epoch": 4.87,
596
+ "learning_rate": 1.1744660804020101e-05,
597
+ "loss": 0.6207,
598
+ "step": 46500
599
+ },
600
+ {
601
+ "epoch": 4.92,
602
+ "learning_rate": 1.1548366834170855e-05,
603
+ "loss": 0.6319,
604
+ "step": 47000
605
+ },
606
+ {
607
+ "epoch": 4.97,
608
+ "learning_rate": 1.1352072864321609e-05,
609
+ "loss": 0.6419,
610
+ "step": 47500
611
+ },
612
+ {
613
+ "epoch": 5.0,
614
+ "eval_loss": 0.7893060445785522,
615
+ "eval_runtime": 8.4012,
616
+ "eval_samples_per_second": 1320.763,
617
+ "eval_steps_per_second": 41.304,
618
+ "step": 47760
619
+ },
620
+ {
621
+ "epoch": 5.03,
622
+ "learning_rate": 1.1155778894472361e-05,
623
+ "loss": 0.6382,
624
+ "step": 48000
625
+ },
626
+ {
627
+ "epoch": 5.08,
628
+ "learning_rate": 1.0959484924623115e-05,
629
+ "loss": 0.6264,
630
+ "step": 48500
631
+ },
632
+ {
633
+ "epoch": 5.13,
634
+ "learning_rate": 1.076319095477387e-05,
635
+ "loss": 0.627,
636
+ "step": 49000
637
+ },
638
+ {
639
+ "epoch": 5.18,
640
+ "learning_rate": 1.0566896984924623e-05,
641
+ "loss": 0.6284,
642
+ "step": 49500
643
+ },
644
+ {
645
+ "epoch": 5.23,
646
+ "learning_rate": 1.0370603015075377e-05,
647
+ "loss": 0.6256,
648
+ "step": 50000
649
+ },
650
+ {
651
+ "epoch": 5.29,
652
+ "learning_rate": 1.0174309045226131e-05,
653
+ "loss": 0.6423,
654
+ "step": 50500
655
+ },
656
+ {
657
+ "epoch": 5.34,
658
+ "learning_rate": 9.978015075376884e-06,
659
+ "loss": 0.634,
660
+ "step": 51000
661
+ },
662
+ {
663
+ "epoch": 5.39,
664
+ "learning_rate": 9.78172110552764e-06,
665
+ "loss": 0.6393,
666
+ "step": 51500
667
+ },
668
+ {
669
+ "epoch": 5.44,
670
+ "learning_rate": 9.585427135678392e-06,
671
+ "loss": 0.6259,
672
+ "step": 52000
673
+ },
674
+ {
675
+ "epoch": 5.5,
676
+ "learning_rate": 9.389133165829146e-06,
677
+ "loss": 0.628,
678
+ "step": 52500
679
+ },
680
+ {
681
+ "epoch": 5.55,
682
+ "learning_rate": 9.1928391959799e-06,
683
+ "loss": 0.6337,
684
+ "step": 53000
685
+ },
686
+ {
687
+ "epoch": 5.6,
688
+ "learning_rate": 8.996545226130654e-06,
689
+ "loss": 0.6128,
690
+ "step": 53500
691
+ },
692
+ {
693
+ "epoch": 5.65,
694
+ "learning_rate": 8.800251256281408e-06,
695
+ "loss": 0.6276,
696
+ "step": 54000
697
+ },
698
+ {
699
+ "epoch": 5.71,
700
+ "learning_rate": 8.60395728643216e-06,
701
+ "loss": 0.629,
702
+ "step": 54500
703
+ },
704
+ {
705
+ "epoch": 5.76,
706
+ "learning_rate": 8.407663316582914e-06,
707
+ "loss": 0.6261,
708
+ "step": 55000
709
+ },
710
+ {
711
+ "epoch": 5.81,
712
+ "learning_rate": 8.21136934673367e-06,
713
+ "loss": 0.6137,
714
+ "step": 55500
715
+ },
716
+ {
717
+ "epoch": 5.86,
718
+ "learning_rate": 8.015075376884422e-06,
719
+ "loss": 0.6353,
720
+ "step": 56000
721
+ },
722
+ {
723
+ "epoch": 5.91,
724
+ "learning_rate": 7.818781407035176e-06,
725
+ "loss": 0.6239,
726
+ "step": 56500
727
+ },
728
+ {
729
+ "epoch": 5.97,
730
+ "learning_rate": 7.622487437185929e-06,
731
+ "loss": 0.618,
732
+ "step": 57000
733
+ },
734
+ {
735
+ "epoch": 6.0,
736
+ "eval_loss": 0.7882150411605835,
737
+ "eval_runtime": 8.1734,
738
+ "eval_samples_per_second": 1357.582,
739
+ "eval_steps_per_second": 42.455,
740
+ "step": 57312
741
+ },
742
+ {
743
+ "epoch": 6.02,
744
+ "learning_rate": 7.426193467336683e-06,
745
+ "loss": 0.6427,
746
+ "step": 57500
747
+ },
748
+ {
749
+ "epoch": 6.07,
750
+ "learning_rate": 7.229899497487438e-06,
751
+ "loss": 0.6283,
752
+ "step": 58000
753
+ },
754
+ {
755
+ "epoch": 6.12,
756
+ "learning_rate": 7.033605527638191e-06,
757
+ "loss": 0.6277,
758
+ "step": 58500
759
+ },
760
+ {
761
+ "epoch": 6.18,
762
+ "learning_rate": 6.837311557788945e-06,
763
+ "loss": 0.6131,
764
+ "step": 59000
765
+ },
766
+ {
767
+ "epoch": 6.23,
768
+ "learning_rate": 6.641017587939698e-06,
769
+ "loss": 0.6086,
770
+ "step": 59500
771
+ },
772
+ {
773
+ "epoch": 6.28,
774
+ "learning_rate": 6.444723618090452e-06,
775
+ "loss": 0.6154,
776
+ "step": 60000
777
+ },
778
+ {
779
+ "epoch": 6.33,
780
+ "learning_rate": 6.248429648241206e-06,
781
+ "loss": 0.6184,
782
+ "step": 60500
783
+ },
784
+ {
785
+ "epoch": 6.39,
786
+ "learning_rate": 6.0521356783919595e-06,
787
+ "loss": 0.6139,
788
+ "step": 61000
789
+ },
790
+ {
791
+ "epoch": 6.44,
792
+ "learning_rate": 5.8558417085427135e-06,
793
+ "loss": 0.6148,
794
+ "step": 61500
795
+ },
796
+ {
797
+ "epoch": 6.49,
798
+ "learning_rate": 5.6595477386934675e-06,
799
+ "loss": 0.6256,
800
+ "step": 62000
801
+ },
802
+ {
803
+ "epoch": 6.54,
804
+ "learning_rate": 5.4632537688442215e-06,
805
+ "loss": 0.6135,
806
+ "step": 62500
807
+ },
808
+ {
809
+ "epoch": 6.6,
810
+ "learning_rate": 5.266959798994975e-06,
811
+ "loss": 0.6274,
812
+ "step": 63000
813
+ },
814
+ {
815
+ "epoch": 6.65,
816
+ "learning_rate": 5.070665829145729e-06,
817
+ "loss": 0.623,
818
+ "step": 63500
819
+ },
820
+ {
821
+ "epoch": 6.7,
822
+ "learning_rate": 4.874371859296483e-06,
823
+ "loss": 0.6254,
824
+ "step": 64000
825
+ },
826
+ {
827
+ "epoch": 6.75,
828
+ "learning_rate": 4.678077889447237e-06,
829
+ "loss": 0.6233,
830
+ "step": 64500
831
+ },
832
+ {
833
+ "epoch": 6.8,
834
+ "learning_rate": 4.48178391959799e-06,
835
+ "loss": 0.6124,
836
+ "step": 65000
837
+ },
838
+ {
839
+ "epoch": 6.86,
840
+ "learning_rate": 4.285489949748744e-06,
841
+ "loss": 0.6108,
842
+ "step": 65500
843
+ },
844
+ {
845
+ "epoch": 6.91,
846
+ "learning_rate": 4.089195979899497e-06,
847
+ "loss": 0.6182,
848
+ "step": 66000
849
+ },
850
+ {
851
+ "epoch": 6.96,
852
+ "learning_rate": 3.892902010050252e-06,
853
+ "loss": 0.5991,
854
+ "step": 66500
855
+ },
856
+ {
857
+ "epoch": 7.0,
858
+ "eval_loss": 0.7896326184272766,
859
+ "eval_runtime": 8.35,
860
+ "eval_samples_per_second": 1328.855,
861
+ "eval_steps_per_second": 41.557,
862
+ "step": 66864
863
+ },
864
+ {
865
+ "epoch": 7.01,
866
+ "learning_rate": 3.696608040201005e-06,
867
+ "loss": 0.6139,
868
+ "step": 67000
869
+ },
870
+ {
871
+ "epoch": 7.07,
872
+ "learning_rate": 3.5003140703517586e-06,
873
+ "loss": 0.6092,
874
+ "step": 67500
875
+ },
876
+ {
877
+ "epoch": 7.12,
878
+ "learning_rate": 3.3040201005025126e-06,
879
+ "loss": 0.6021,
880
+ "step": 68000
881
+ },
882
+ {
883
+ "epoch": 7.17,
884
+ "learning_rate": 3.107726130653266e-06,
885
+ "loss": 0.6138,
886
+ "step": 68500
887
+ },
888
+ {
889
+ "epoch": 7.22,
890
+ "learning_rate": 2.91143216080402e-06,
891
+ "loss": 0.6279,
892
+ "step": 69000
893
+ },
894
+ {
895
+ "epoch": 7.28,
896
+ "learning_rate": 2.7151381909547737e-06,
897
+ "loss": 0.6086,
898
+ "step": 69500
899
+ },
900
+ {
901
+ "epoch": 7.33,
902
+ "learning_rate": 2.5188442211055277e-06,
903
+ "loss": 0.6108,
904
+ "step": 70000
905
+ },
906
+ {
907
+ "epoch": 7.38,
908
+ "learning_rate": 2.3225502512562813e-06,
909
+ "loss": 0.5933,
910
+ "step": 70500
911
+ },
912
+ {
913
+ "epoch": 7.43,
914
+ "learning_rate": 2.1262562814070353e-06,
915
+ "loss": 0.6089,
916
+ "step": 71000
917
+ },
918
+ {
919
+ "epoch": 7.49,
920
+ "learning_rate": 1.929962311557789e-06,
921
+ "loss": 0.6086,
922
+ "step": 71500
923
+ },
924
+ {
925
+ "epoch": 7.54,
926
+ "learning_rate": 1.7336683417085427e-06,
927
+ "loss": 0.622,
928
+ "step": 72000
929
+ },
930
+ {
931
+ "epoch": 7.59,
932
+ "learning_rate": 1.5373743718592965e-06,
933
+ "loss": 0.6041,
934
+ "step": 72500
935
+ },
936
+ {
937
+ "epoch": 7.64,
938
+ "learning_rate": 1.3410804020100503e-06,
939
+ "loss": 0.6146,
940
+ "step": 73000
941
+ },
942
+ {
943
+ "epoch": 7.69,
944
+ "learning_rate": 1.144786432160804e-06,
945
+ "loss": 0.5999,
946
+ "step": 73500
947
+ },
948
+ {
949
+ "epoch": 7.75,
950
+ "learning_rate": 9.484924623115579e-07,
951
+ "loss": 0.6094,
952
+ "step": 74000
953
+ },
954
+ {
955
+ "epoch": 7.8,
956
+ "learning_rate": 7.521984924623115e-07,
957
+ "loss": 0.6137,
958
+ "step": 74500
959
+ },
960
+ {
961
+ "epoch": 7.85,
962
+ "learning_rate": 5.559045226130653e-07,
963
+ "loss": 0.6065,
964
+ "step": 75000
965
+ },
966
+ {
967
+ "epoch": 7.9,
968
+ "learning_rate": 3.5961055276381907e-07,
969
+ "loss": 0.6149,
970
+ "step": 75500
971
+ },
972
+ {
973
+ "epoch": 7.96,
974
+ "learning_rate": 1.6331658291457286e-07,
975
+ "loss": 0.6159,
976
+ "step": 76000
977
+ },
978
+ {
979
+ "epoch": 8.0,
980
+ "eval_loss": 0.7893399000167847,
981
+ "eval_runtime": 8.9446,
982
+ "eval_samples_per_second": 1240.522,
983
+ "eval_steps_per_second": 38.794,
984
+ "step": 76416
985
+ }
986
+ ],
987
+ "max_steps": 76416,
988
+ "num_train_epochs": 8,
989
+ "total_flos": 2.326706308251648e+16,
990
+ "trial_name": null,
991
+ "trial_params": null
992
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891dcffc906bdce9095831d147888a51075a4b151ccb2e97841da7fedaa48483
3
+ size 3439