hiroyukikaneko commited on
Commit
82ea53c
·
verified ·
1 Parent(s): 0045ec8

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +427 -0
trainer_state.json ADDED
@@ -0,0 +1,427 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 556,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.017985611510791366,
13
+ "grad_norm": 8.904035568237305,
14
+ "learning_rate": 3.571428571428572e-05,
15
+ "loss": 1.2754,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.03597122302158273,
20
+ "grad_norm": 3.598201274871826,
21
+ "learning_rate": 7.142857142857143e-05,
22
+ "loss": 0.8598,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.0539568345323741,
27
+ "grad_norm": 4.27315616607666,
28
+ "learning_rate": 9.999645980833454e-05,
29
+ "loss": 0.5009,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.07194244604316546,
34
+ "grad_norm": 1.9983025789260864,
35
+ "learning_rate": 9.987260573051269e-05,
36
+ "loss": 0.3516,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.08992805755395683,
41
+ "grad_norm": 3.3641581535339355,
42
+ "learning_rate": 9.957224306869053e-05,
43
+ "loss": 0.2792,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.1079136690647482,
48
+ "grad_norm": 1.5614864826202393,
49
+ "learning_rate": 9.909643486313533e-05,
50
+ "loss": 0.233,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.12589928057553956,
55
+ "grad_norm": 1.1084058284759521,
56
+ "learning_rate": 9.844686508907537e-05,
57
+ "loss": 0.1771,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.14388489208633093,
62
+ "grad_norm": 0.6237399578094482,
63
+ "learning_rate": 9.762583269679303e-05,
64
+ "loss": 0.1681,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.1618705035971223,
69
+ "grad_norm": 0.8216731548309326,
70
+ "learning_rate": 9.663624347520505e-05,
71
+ "loss": 0.1467,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.17985611510791366,
76
+ "grad_norm": 1.565852165222168,
77
+ "learning_rate": 9.548159976772592e-05,
78
+ "loss": 0.145,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.19784172661870503,
83
+ "grad_norm": 0.950880765914917,
84
+ "learning_rate": 9.41659880768122e-05,
85
+ "loss": 0.1361,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.2158273381294964,
90
+ "grad_norm": 1.6979942321777344,
91
+ "learning_rate": 9.26940646010574e-05,
92
+ "loss": 0.1257,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.23381294964028776,
97
+ "grad_norm": 1.192500114440918,
98
+ "learning_rate": 9.107103875602459e-05,
99
+ "loss": 0.1144,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.2517985611510791,
104
+ "grad_norm": 0.7761622667312622,
105
+ "learning_rate": 8.930265473713938e-05,
106
+ "loss": 0.1192,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.2697841726618705,
111
+ "grad_norm": 0.8879750370979309,
112
+ "learning_rate": 8.739517118989605e-05,
113
+ "loss": 0.1098,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.28776978417266186,
118
+ "grad_norm": 1.1123480796813965,
119
+ "learning_rate": 8.535533905932738e-05,
120
+ "loss": 0.0984,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.3057553956834532,
125
+ "grad_norm": 0.9935937523841858,
126
+ "learning_rate": 8.319037769713338e-05,
127
+ "loss": 0.0999,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.3237410071942446,
132
+ "grad_norm": 0.8555904030799866,
133
+ "learning_rate": 8.090794931103026e-05,
134
+ "loss": 0.097,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.34172661870503596,
139
+ "grad_norm": 0.4569714069366455,
140
+ "learning_rate": 7.85161318467482e-05,
141
+ "loss": 0.0909,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.3597122302158273,
146
+ "grad_norm": 0.5646055936813354,
147
+ "learning_rate": 7.602339039865362e-05,
148
+ "loss": 0.0915,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.3776978417266187,
153
+ "grad_norm": 0.6606709957122803,
154
+ "learning_rate": 7.343854725017918e-05,
155
+ "loss": 0.0965,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.39568345323741005,
160
+ "grad_norm": 0.6112151145935059,
161
+ "learning_rate": 7.077075065009433e-05,
162
+ "loss": 0.0912,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.4136690647482014,
167
+ "grad_norm": 0.7429100275039673,
168
+ "learning_rate": 6.80294424351225e-05,
169
+ "loss": 0.0843,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.4316546762589928,
174
+ "grad_norm": 1.2504783868789673,
175
+ "learning_rate": 6.522432461349536e-05,
176
+ "loss": 0.0816,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.44964028776978415,
181
+ "grad_norm": 0.5470308065414429,
182
+ "learning_rate": 6.236532502771078e-05,
183
+ "loss": 0.0806,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.4676258992805755,
188
+ "grad_norm": 1.6046953201293945,
189
+ "learning_rate": 5.946256221802051e-05,
190
+ "loss": 0.0792,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.4856115107913669,
195
+ "grad_norm": 0.6300632357597351,
196
+ "learning_rate": 5.6526309611002594e-05,
197
+ "loss": 0.0928,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.5035971223021583,
202
+ "grad_norm": 0.7121009826660156,
203
+ "learning_rate": 5.3566959159961615e-05,
204
+ "loss": 0.0789,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.5215827338129496,
209
+ "grad_norm": 1.3714096546173096,
210
+ "learning_rate": 5.059498456584072e-05,
211
+ "loss": 0.0845,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.539568345323741,
216
+ "grad_norm": 1.2124353647232056,
217
+ "learning_rate": 4.762090420881289e-05,
218
+ "loss": 0.0759,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.5575539568345323,
223
+ "grad_norm": 0.8162264823913574,
224
+ "learning_rate": 4.4655243921744374e-05,
225
+ "loss": 0.0826,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.5755395683453237,
230
+ "grad_norm": 0.7335633039474487,
231
+ "learning_rate": 4.17084997372813e-05,
232
+ "loss": 0.0819,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.5935251798561151,
237
+ "grad_norm": 1.067676067352295,
238
+ "learning_rate": 3.879110074040514e-05,
239
+ "loss": 0.0793,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.6115107913669064,
244
+ "grad_norm": 0.768426239490509,
245
+ "learning_rate": 3.591337215792852e-05,
246
+ "loss": 0.072,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.6294964028776978,
251
+ "grad_norm": 0.6882124543190002,
252
+ "learning_rate": 3.3085498815564645e-05,
253
+ "loss": 0.0732,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.6474820143884892,
258
+ "grad_norm": 0.7045790553092957,
259
+ "learning_rate": 3.0317489091902935e-05,
260
+ "loss": 0.0751,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.6654676258992805,
265
+ "grad_norm": 0.6171532869338989,
266
+ "learning_rate": 2.7619139496864378e-05,
267
+ "loss": 0.0723,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.6834532374100719,
272
+ "grad_norm": 1.0635524988174438,
273
+ "learning_rate": 2.500000000000001e-05,
274
+ "loss": 0.0727,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.7014388489208633,
279
+ "grad_norm": 1.7545868158340454,
280
+ "learning_rate": 2.246934023134257e-05,
281
+ "loss": 0.0715,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.7194244604316546,
286
+ "grad_norm": 0.6268433332443237,
287
+ "learning_rate": 2.0036116674432654e-05,
288
+ "loss": 0.0704,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.737410071942446,
293
+ "grad_norm": 0.6153240203857422,
294
+ "learning_rate": 1.7708940967629567e-05,
295
+ "loss": 0.0776,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.7553956834532374,
300
+ "grad_norm": 0.6164385080337524,
301
+ "learning_rate": 1.549604942589441e-05,
302
+ "loss": 0.0681,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.7733812949640287,
307
+ "grad_norm": 0.780419647693634,
308
+ "learning_rate": 1.340527389091374e-05,
309
+ "loss": 0.0749,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.7913669064748201,
314
+ "grad_norm": 0.8401429653167725,
315
+ "learning_rate": 1.144401401273062e-05,
316
+ "loss": 0.0697,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.8093525179856115,
321
+ "grad_norm": 0.716050386428833,
322
+ "learning_rate": 9.619211060983675e-06,
323
+ "loss": 0.0644,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.8273381294964028,
328
+ "grad_norm": 0.7171333432197571,
329
+ "learning_rate": 7.937323358440935e-06,
330
+ "loss": 0.0689,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.8453237410071942,
335
+ "grad_norm": 0.8431842923164368,
336
+ "learning_rate": 6.40430342377426e-06,
337
+ "loss": 0.0754,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.8633093525179856,
342
+ "grad_norm": 0.47953060269355774,
343
+ "learning_rate": 5.025576904469842e-06,
344
+ "loss": 0.0747,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.8812949640287769,
349
+ "grad_norm": 0.793252170085907,
350
+ "learning_rate": 3.8060233744356633e-06,
351
+ "loss": 0.0694,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.8992805755395683,
356
+ "grad_norm": 0.5008072853088379,
357
+ "learning_rate": 2.7499590642665774e-06,
358
+ "loss": 0.0637,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.9172661870503597,
363
+ "grad_norm": 0.7334234118461609,
364
+ "learning_rate": 1.8611215852879005e-06,
365
+ "loss": 0.0753,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.935251798561151,
370
+ "grad_norm": 0.64471435546875,
371
+ "learning_rate": 1.1426567014420297e-06,
372
+ "loss": 0.0632,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.9532374100719424,
377
+ "grad_norm": 1.301717758178711,
378
+ "learning_rate": 5.971071958349228e-07,
379
+ "loss": 0.0709,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.9712230215827338,
384
+ "grad_norm": 0.5564674139022827,
385
+ "learning_rate": 2.2640387134577058e-07,
386
+ "loss": 0.0647,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.9892086330935251,
391
+ "grad_norm": 0.42517802119255066,
392
+ "learning_rate": 3.185871715041255e-08,
393
+ "loss": 0.0666,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 1.0,
398
+ "step": 556,
399
+ "total_flos": 3.809504209392605e+16,
400
+ "train_loss": 0.14214680440348687,
401
+ "train_runtime": 180.0894,
402
+ "train_samples_per_second": 98.768,
403
+ "train_steps_per_second": 3.087
404
+ }
405
+ ],
406
+ "logging_steps": 10,
407
+ "max_steps": 556,
408
+ "num_input_tokens_seen": 0,
409
+ "num_train_epochs": 1,
410
+ "save_steps": 20000,
411
+ "stateful_callbacks": {
412
+ "TrainerControl": {
413
+ "args": {
414
+ "should_epoch_stop": false,
415
+ "should_evaluate": false,
416
+ "should_log": false,
417
+ "should_save": true,
418
+ "should_training_stop": true
419
+ },
420
+ "attributes": {}
421
+ }
422
+ },
423
+ "total_flos": 3.809504209392605e+16,
424
+ "train_batch_size": 32,
425
+ "trial_name": null,
426
+ "trial_params": null
427
+ }