Hritik commited on
Commit
1190ba8
1 Parent(s): ddb522b

first commit

Browse files
Files changed (7) hide show
  1. optimizer.pt +3 -0
  2. pytorch_model.bin +3 -0
  3. rng_state.pth +3 -0
  4. scaler.pt +3 -0
  5. scheduler.pt +3 -0
  6. trainer_state.json +1136 -0
  7. training_args.bin +3 -0
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:696f417871a7eff2eae730c1cbcd3a1479c1b6c391b495c5c70aba3255258001
3
+ size 134358981
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6e7ad47146dbcce59f74751037515cff750ff709dcc6d71ebef6d340b8f88a0
3
+ size 7074279085
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:745ed0043b7a7fa03a50be03fe437716dd2f930ef5ae25e87fba773dc6aff9cf
3
+ size 14575
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2e8bcb49fc82058bd85617725fb069d7206d3cd1f05a30c977d7d843b24b413
3
+ size 557
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd9d8496494b6cfd49dd20d1d756e87ccf1c8cfb6f95e8601a561ca6104b1cf1
3
+ size 627
trainer_state.json ADDED
@@ -0,0 +1,1136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.8831531217300315,
5
+ "global_step": 7000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 4.4000000000000006e-05,
13
+ "loss": 0.2783,
14
+ "step": 50
15
+ },
16
+ {
17
+ "epoch": 0.07,
18
+ "learning_rate": 9.4e-05,
19
+ "loss": 0.079,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.1,
24
+ "learning_rate": 9.937721160651098e-05,
25
+ "loss": 0.0502,
26
+ "step": 150
27
+ },
28
+ {
29
+ "epoch": 0.14,
30
+ "learning_rate": 9.86694975230007e-05,
31
+ "loss": 0.0464,
32
+ "step": 200
33
+ },
34
+ {
35
+ "epoch": 0.14,
36
+ "eval_loss": 0.029130851849913597,
37
+ "eval_runtime": 1098.3952,
38
+ "eval_samples_per_second": 6.711,
39
+ "eval_steps_per_second": 0.28,
40
+ "step": 200
41
+ },
42
+ {
43
+ "epoch": 0.17,
44
+ "learning_rate": 9.796178343949045e-05,
45
+ "loss": 0.0331,
46
+ "step": 250
47
+ },
48
+ {
49
+ "epoch": 0.21,
50
+ "learning_rate": 9.725406935598019e-05,
51
+ "loss": 0.0292,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.24,
56
+ "learning_rate": 9.654635527246992e-05,
57
+ "loss": 0.0292,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.28,
62
+ "learning_rate": 9.583864118895966e-05,
63
+ "loss": 0.032,
64
+ "step": 400
65
+ },
66
+ {
67
+ "epoch": 0.28,
68
+ "eval_loss": 0.02652278169989586,
69
+ "eval_runtime": 1094.8086,
70
+ "eval_samples_per_second": 6.733,
71
+ "eval_steps_per_second": 0.281,
72
+ "step": 400
73
+ },
74
+ {
75
+ "epoch": 0.31,
76
+ "learning_rate": 9.513092710544941e-05,
77
+ "loss": 0.0272,
78
+ "step": 450
79
+ },
80
+ {
81
+ "epoch": 0.35,
82
+ "learning_rate": 9.442321302193913e-05,
83
+ "loss": 0.026,
84
+ "step": 500
85
+ },
86
+ {
87
+ "epoch": 0.38,
88
+ "learning_rate": 9.371549893842888e-05,
89
+ "loss": 0.0227,
90
+ "step": 550
91
+ },
92
+ {
93
+ "epoch": 0.42,
94
+ "learning_rate": 9.300778485491862e-05,
95
+ "loss": 0.0246,
96
+ "step": 600
97
+ },
98
+ {
99
+ "epoch": 0.42,
100
+ "eval_loss": 0.02081386186182499,
101
+ "eval_runtime": 1091.9223,
102
+ "eval_samples_per_second": 6.75,
103
+ "eval_steps_per_second": 0.282,
104
+ "step": 600
105
+ },
106
+ {
107
+ "epoch": 0.45,
108
+ "learning_rate": 9.230007077140835e-05,
109
+ "loss": 0.0263,
110
+ "step": 650
111
+ },
112
+ {
113
+ "epoch": 0.49,
114
+ "learning_rate": 9.159235668789809e-05,
115
+ "loss": 0.0252,
116
+ "step": 700
117
+ },
118
+ {
119
+ "epoch": 0.52,
120
+ "learning_rate": 9.088464260438783e-05,
121
+ "loss": 0.0253,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 0.56,
126
+ "learning_rate": 9.017692852087756e-05,
127
+ "loss": 0.0274,
128
+ "step": 800
129
+ },
130
+ {
131
+ "epoch": 0.56,
132
+ "eval_loss": 0.019421329721808434,
133
+ "eval_runtime": 1091.9762,
134
+ "eval_samples_per_second": 6.75,
135
+ "eval_steps_per_second": 0.282,
136
+ "step": 800
137
+ },
138
+ {
139
+ "epoch": 0.59,
140
+ "learning_rate": 8.946921443736731e-05,
141
+ "loss": 0.0211,
142
+ "step": 850
143
+ },
144
+ {
145
+ "epoch": 0.63,
146
+ "learning_rate": 8.876150035385705e-05,
147
+ "loss": 0.0234,
148
+ "step": 900
149
+ },
150
+ {
151
+ "epoch": 0.66,
152
+ "learning_rate": 8.805378627034678e-05,
153
+ "loss": 0.0227,
154
+ "step": 950
155
+ },
156
+ {
157
+ "epoch": 0.7,
158
+ "learning_rate": 8.734607218683652e-05,
159
+ "loss": 0.02,
160
+ "step": 1000
161
+ },
162
+ {
163
+ "epoch": 0.7,
164
+ "eval_loss": 0.01722061075270176,
165
+ "eval_runtime": 1091.8825,
166
+ "eval_samples_per_second": 6.751,
167
+ "eval_steps_per_second": 0.282,
168
+ "step": 1000
169
+ },
170
+ {
171
+ "epoch": 0.73,
172
+ "learning_rate": 8.663835810332625e-05,
173
+ "loss": 0.0229,
174
+ "step": 1050
175
+ },
176
+ {
177
+ "epoch": 0.77,
178
+ "learning_rate": 8.5930644019816e-05,
179
+ "loss": 0.0189,
180
+ "step": 1100
181
+ },
182
+ {
183
+ "epoch": 0.8,
184
+ "learning_rate": 8.522292993630574e-05,
185
+ "loss": 0.0188,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 0.84,
190
+ "learning_rate": 8.451521585279548e-05,
191
+ "loss": 0.018,
192
+ "step": 1200
193
+ },
194
+ {
195
+ "epoch": 0.84,
196
+ "eval_loss": 0.017043238505721092,
197
+ "eval_runtime": 1091.8986,
198
+ "eval_samples_per_second": 6.751,
199
+ "eval_steps_per_second": 0.282,
200
+ "step": 1200
201
+ },
202
+ {
203
+ "epoch": 0.87,
204
+ "learning_rate": 8.380750176928521e-05,
205
+ "loss": 0.0183,
206
+ "step": 1250
207
+ },
208
+ {
209
+ "epoch": 0.91,
210
+ "learning_rate": 8.309978768577495e-05,
211
+ "loss": 0.0207,
212
+ "step": 1300
213
+ },
214
+ {
215
+ "epoch": 0.94,
216
+ "learning_rate": 8.239207360226468e-05,
217
+ "loss": 0.0147,
218
+ "step": 1350
219
+ },
220
+ {
221
+ "epoch": 0.98,
222
+ "learning_rate": 8.168435951875443e-05,
223
+ "loss": 0.0151,
224
+ "step": 1400
225
+ },
226
+ {
227
+ "epoch": 0.98,
228
+ "eval_loss": 0.015169227495789528,
229
+ "eval_runtime": 1091.8795,
230
+ "eval_samples_per_second": 6.751,
231
+ "eval_steps_per_second": 0.282,
232
+ "step": 1400
233
+ },
234
+ {
235
+ "epoch": 1.01,
236
+ "learning_rate": 8.097664543524417e-05,
237
+ "loss": 0.0148,
238
+ "step": 1450
239
+ },
240
+ {
241
+ "epoch": 1.05,
242
+ "learning_rate": 8.02689313517339e-05,
243
+ "loss": 0.0125,
244
+ "step": 1500
245
+ },
246
+ {
247
+ "epoch": 1.08,
248
+ "learning_rate": 7.956121726822364e-05,
249
+ "loss": 0.0129,
250
+ "step": 1550
251
+ },
252
+ {
253
+ "epoch": 1.12,
254
+ "learning_rate": 7.885350318471339e-05,
255
+ "loss": 0.0138,
256
+ "step": 1600
257
+ },
258
+ {
259
+ "epoch": 1.12,
260
+ "eval_loss": 0.014357708394527435,
261
+ "eval_runtime": 1091.7946,
262
+ "eval_samples_per_second": 6.751,
263
+ "eval_steps_per_second": 0.282,
264
+ "step": 1600
265
+ },
266
+ {
267
+ "epoch": 1.15,
268
+ "learning_rate": 7.814578910120311e-05,
269
+ "loss": 0.015,
270
+ "step": 1650
271
+ },
272
+ {
273
+ "epoch": 1.19,
274
+ "learning_rate": 7.743807501769286e-05,
275
+ "loss": 0.0152,
276
+ "step": 1700
277
+ },
278
+ {
279
+ "epoch": 1.22,
280
+ "learning_rate": 7.67303609341826e-05,
281
+ "loss": 0.014,
282
+ "step": 1750
283
+ },
284
+ {
285
+ "epoch": 1.26,
286
+ "learning_rate": 7.602264685067233e-05,
287
+ "loss": 0.015,
288
+ "step": 1800
289
+ },
290
+ {
291
+ "epoch": 1.26,
292
+ "eval_loss": 0.013772633858025074,
293
+ "eval_runtime": 1091.9078,
294
+ "eval_samples_per_second": 6.751,
295
+ "eval_steps_per_second": 0.282,
296
+ "step": 1800
297
+ },
298
+ {
299
+ "epoch": 1.29,
300
+ "learning_rate": 7.531493276716207e-05,
301
+ "loss": 0.0134,
302
+ "step": 1850
303
+ },
304
+ {
305
+ "epoch": 1.33,
306
+ "learning_rate": 7.46072186836518e-05,
307
+ "loss": 0.0116,
308
+ "step": 1900
309
+ },
310
+ {
311
+ "epoch": 1.36,
312
+ "learning_rate": 7.389950460014154e-05,
313
+ "loss": 0.0136,
314
+ "step": 1950
315
+ },
316
+ {
317
+ "epoch": 1.4,
318
+ "learning_rate": 7.319179051663129e-05,
319
+ "loss": 0.0134,
320
+ "step": 2000
321
+ },
322
+ {
323
+ "epoch": 1.4,
324
+ "eval_loss": 0.013411927036941051,
325
+ "eval_runtime": 1091.4452,
326
+ "eval_samples_per_second": 6.753,
327
+ "eval_steps_per_second": 0.282,
328
+ "step": 2000
329
+ },
330
+ {
331
+ "epoch": 1.43,
332
+ "learning_rate": 7.248407643312101e-05,
333
+ "loss": 0.0129,
334
+ "step": 2050
335
+ },
336
+ {
337
+ "epoch": 1.46,
338
+ "learning_rate": 7.177636234961076e-05,
339
+ "loss": 0.0144,
340
+ "step": 2100
341
+ },
342
+ {
343
+ "epoch": 1.5,
344
+ "learning_rate": 7.10686482661005e-05,
345
+ "loss": 0.0119,
346
+ "step": 2150
347
+ },
348
+ {
349
+ "epoch": 1.53,
350
+ "learning_rate": 7.036093418259024e-05,
351
+ "loss": 0.0152,
352
+ "step": 2200
353
+ },
354
+ {
355
+ "epoch": 1.53,
356
+ "eval_loss": 0.014650107361376286,
357
+ "eval_runtime": 1091.4623,
358
+ "eval_samples_per_second": 6.753,
359
+ "eval_steps_per_second": 0.282,
360
+ "step": 2200
361
+ },
362
+ {
363
+ "epoch": 1.57,
364
+ "learning_rate": 6.965322009907997e-05,
365
+ "loss": 0.0145,
366
+ "step": 2250
367
+ },
368
+ {
369
+ "epoch": 1.6,
370
+ "learning_rate": 6.894550601556972e-05,
371
+ "loss": 0.0154,
372
+ "step": 2300
373
+ },
374
+ {
375
+ "epoch": 1.64,
376
+ "learning_rate": 6.823779193205944e-05,
377
+ "loss": 0.013,
378
+ "step": 2350
379
+ },
380
+ {
381
+ "epoch": 1.67,
382
+ "learning_rate": 6.75300778485492e-05,
383
+ "loss": 0.0123,
384
+ "step": 2400
385
+ },
386
+ {
387
+ "epoch": 1.67,
388
+ "eval_loss": 0.013516324572265148,
389
+ "eval_runtime": 1091.733,
390
+ "eval_samples_per_second": 6.752,
391
+ "eval_steps_per_second": 0.282,
392
+ "step": 2400
393
+ },
394
+ {
395
+ "epoch": 1.71,
396
+ "learning_rate": 6.682236376503893e-05,
397
+ "loss": 0.0123,
398
+ "step": 2450
399
+ },
400
+ {
401
+ "epoch": 1.74,
402
+ "learning_rate": 6.611464968152867e-05,
403
+ "loss": 0.0132,
404
+ "step": 2500
405
+ },
406
+ {
407
+ "epoch": 1.78,
408
+ "learning_rate": 6.54069355980184e-05,
409
+ "loss": 0.0172,
410
+ "step": 2550
411
+ },
412
+ {
413
+ "epoch": 1.81,
414
+ "learning_rate": 6.469922151450815e-05,
415
+ "loss": 0.0121,
416
+ "step": 2600
417
+ },
418
+ {
419
+ "epoch": 1.81,
420
+ "eval_loss": 0.013174451887607574,
421
+ "eval_runtime": 1091.4369,
422
+ "eval_samples_per_second": 6.753,
423
+ "eval_steps_per_second": 0.282,
424
+ "step": 2600
425
+ },
426
+ {
427
+ "epoch": 1.85,
428
+ "learning_rate": 6.399150743099787e-05,
429
+ "loss": 0.0112,
430
+ "step": 2650
431
+ },
432
+ {
433
+ "epoch": 1.88,
434
+ "learning_rate": 6.328379334748762e-05,
435
+ "loss": 0.0137,
436
+ "step": 2700
437
+ },
438
+ {
439
+ "epoch": 1.92,
440
+ "learning_rate": 6.257607926397736e-05,
441
+ "loss": 0.0107,
442
+ "step": 2750
443
+ },
444
+ {
445
+ "epoch": 1.95,
446
+ "learning_rate": 6.18683651804671e-05,
447
+ "loss": 0.0115,
448
+ "step": 2800
449
+ },
450
+ {
451
+ "epoch": 1.95,
452
+ "eval_loss": 0.01469574123620987,
453
+ "eval_runtime": 1091.5407,
454
+ "eval_samples_per_second": 6.753,
455
+ "eval_steps_per_second": 0.282,
456
+ "step": 2800
457
+ },
458
+ {
459
+ "epoch": 1.99,
460
+ "learning_rate": 6.116065109695683e-05,
461
+ "loss": 0.015,
462
+ "step": 2850
463
+ },
464
+ {
465
+ "epoch": 2.02,
466
+ "learning_rate": 6.045293701344657e-05,
467
+ "loss": 0.0123,
468
+ "step": 2900
469
+ },
470
+ {
471
+ "epoch": 2.06,
472
+ "learning_rate": 5.974522292993631e-05,
473
+ "loss": 0.0101,
474
+ "step": 2950
475
+ },
476
+ {
477
+ "epoch": 2.09,
478
+ "learning_rate": 5.9037508846426045e-05,
479
+ "loss": 0.0115,
480
+ "step": 3000
481
+ },
482
+ {
483
+ "epoch": 2.09,
484
+ "eval_loss": 0.01295243389904499,
485
+ "eval_runtime": 1091.2144,
486
+ "eval_samples_per_second": 6.755,
487
+ "eval_steps_per_second": 0.282,
488
+ "step": 3000
489
+ },
490
+ {
491
+ "epoch": 2.13,
492
+ "learning_rate": 5.832979476291578e-05,
493
+ "loss": 0.0116,
494
+ "step": 3050
495
+ },
496
+ {
497
+ "epoch": 2.16,
498
+ "learning_rate": 5.7622080679405524e-05,
499
+ "loss": 0.0117,
500
+ "step": 3100
501
+ },
502
+ {
503
+ "epoch": 2.2,
504
+ "learning_rate": 5.6914366595895267e-05,
505
+ "loss": 0.0148,
506
+ "step": 3150
507
+ },
508
+ {
509
+ "epoch": 2.23,
510
+ "learning_rate": 5.6206652512384996e-05,
511
+ "loss": 0.0104,
512
+ "step": 3200
513
+ },
514
+ {
515
+ "epoch": 2.23,
516
+ "eval_loss": 0.012959270738065243,
517
+ "eval_runtime": 1091.2195,
518
+ "eval_samples_per_second": 6.755,
519
+ "eval_steps_per_second": 0.282,
520
+ "step": 3200
521
+ },
522
+ {
523
+ "epoch": 2.27,
524
+ "learning_rate": 5.549893842887474e-05,
525
+ "loss": 0.01,
526
+ "step": 3250
527
+ },
528
+ {
529
+ "epoch": 2.3,
530
+ "learning_rate": 5.479122434536448e-05,
531
+ "loss": 0.0095,
532
+ "step": 3300
533
+ },
534
+ {
535
+ "epoch": 2.34,
536
+ "learning_rate": 5.408351026185421e-05,
537
+ "loss": 0.0108,
538
+ "step": 3350
539
+ },
540
+ {
541
+ "epoch": 2.37,
542
+ "learning_rate": 5.337579617834395e-05,
543
+ "loss": 0.0112,
544
+ "step": 3400
545
+ },
546
+ {
547
+ "epoch": 2.37,
548
+ "eval_loss": 0.013205363415181637,
549
+ "eval_runtime": 1091.0349,
550
+ "eval_samples_per_second": 6.756,
551
+ "eval_steps_per_second": 0.282,
552
+ "step": 3400
553
+ },
554
+ {
555
+ "epoch": 2.41,
556
+ "learning_rate": 5.2668082094833696e-05,
557
+ "loss": 0.0119,
558
+ "step": 3450
559
+ },
560
+ {
561
+ "epoch": 2.44,
562
+ "learning_rate": 5.1960368011323425e-05,
563
+ "loss": 0.0107,
564
+ "step": 3500
565
+ },
566
+ {
567
+ "epoch": 2.48,
568
+ "learning_rate": 5.125265392781317e-05,
569
+ "loss": 0.0129,
570
+ "step": 3550
571
+ },
572
+ {
573
+ "epoch": 2.51,
574
+ "learning_rate": 5.054493984430291e-05,
575
+ "loss": 0.0103,
576
+ "step": 3600
577
+ },
578
+ {
579
+ "epoch": 2.51,
580
+ "eval_loss": 0.013000305742025375,
581
+ "eval_runtime": 1091.2105,
582
+ "eval_samples_per_second": 6.755,
583
+ "eval_steps_per_second": 0.282,
584
+ "step": 3600
585
+ },
586
+ {
587
+ "epoch": 2.55,
588
+ "learning_rate": 4.9837225760792646e-05,
589
+ "loss": 0.0114,
590
+ "step": 3650
591
+ },
592
+ {
593
+ "epoch": 2.58,
594
+ "learning_rate": 4.912951167728238e-05,
595
+ "loss": 0.0116,
596
+ "step": 3700
597
+ },
598
+ {
599
+ "epoch": 2.62,
600
+ "learning_rate": 4.842179759377212e-05,
601
+ "loss": 0.0106,
602
+ "step": 3750
603
+ },
604
+ {
605
+ "epoch": 2.65,
606
+ "learning_rate": 4.771408351026186e-05,
607
+ "loss": 0.0099,
608
+ "step": 3800
609
+ },
610
+ {
611
+ "epoch": 2.65,
612
+ "eval_loss": 0.01273022498935461,
613
+ "eval_runtime": 1091.0891,
614
+ "eval_samples_per_second": 6.756,
615
+ "eval_steps_per_second": 0.282,
616
+ "step": 3800
617
+ },
618
+ {
619
+ "epoch": 2.69,
620
+ "learning_rate": 4.70063694267516e-05,
621
+ "loss": 0.0109,
622
+ "step": 3850
623
+ },
624
+ {
625
+ "epoch": 2.72,
626
+ "learning_rate": 4.629865534324133e-05,
627
+ "loss": 0.0125,
628
+ "step": 3900
629
+ },
630
+ {
631
+ "epoch": 2.76,
632
+ "learning_rate": 4.559094125973107e-05,
633
+ "loss": 0.0135,
634
+ "step": 3950
635
+ },
636
+ {
637
+ "epoch": 2.79,
638
+ "learning_rate": 4.488322717622081e-05,
639
+ "loss": 0.0111,
640
+ "step": 4000
641
+ },
642
+ {
643
+ "epoch": 2.79,
644
+ "eval_loss": 0.012479917146265507,
645
+ "eval_runtime": 1091.0461,
646
+ "eval_samples_per_second": 6.756,
647
+ "eval_steps_per_second": 0.282,
648
+ "step": 4000
649
+ },
650
+ {
651
+ "epoch": 2.83,
652
+ "learning_rate": 4.417551309271055e-05,
653
+ "loss": 0.0106,
654
+ "step": 4050
655
+ },
656
+ {
657
+ "epoch": 2.86,
658
+ "learning_rate": 4.3467799009200284e-05,
659
+ "loss": 0.0115,
660
+ "step": 4100
661
+ },
662
+ {
663
+ "epoch": 2.9,
664
+ "learning_rate": 4.2760084925690026e-05,
665
+ "loss": 0.0107,
666
+ "step": 4150
667
+ },
668
+ {
669
+ "epoch": 2.93,
670
+ "learning_rate": 4.205237084217976e-05,
671
+ "loss": 0.0094,
672
+ "step": 4200
673
+ },
674
+ {
675
+ "epoch": 2.93,
676
+ "eval_loss": 0.0125270402058959,
677
+ "eval_runtime": 1091.2851,
678
+ "eval_samples_per_second": 6.754,
679
+ "eval_steps_per_second": 0.282,
680
+ "step": 4200
681
+ },
682
+ {
683
+ "epoch": 2.96,
684
+ "learning_rate": 4.13446567586695e-05,
685
+ "loss": 0.0127,
686
+ "step": 4250
687
+ },
688
+ {
689
+ "epoch": 3.0,
690
+ "learning_rate": 4.063694267515924e-05,
691
+ "loss": 0.0088,
692
+ "step": 4300
693
+ },
694
+ {
695
+ "epoch": 3.03,
696
+ "learning_rate": 3.992922859164898e-05,
697
+ "loss": 0.008,
698
+ "step": 4350
699
+ },
700
+ {
701
+ "epoch": 3.07,
702
+ "learning_rate": 3.922151450813871e-05,
703
+ "loss": 0.0083,
704
+ "step": 4400
705
+ },
706
+ {
707
+ "epoch": 3.07,
708
+ "eval_loss": 0.012798724696040154,
709
+ "eval_runtime": 1090.8595,
710
+ "eval_samples_per_second": 6.757,
711
+ "eval_steps_per_second": 0.282,
712
+ "step": 4400
713
+ },
714
+ {
715
+ "epoch": 3.1,
716
+ "learning_rate": 3.8513800424628456e-05,
717
+ "loss": 0.0101,
718
+ "step": 4450
719
+ },
720
+ {
721
+ "epoch": 3.14,
722
+ "learning_rate": 3.780608634111819e-05,
723
+ "loss": 0.0085,
724
+ "step": 4500
725
+ },
726
+ {
727
+ "epoch": 3.17,
728
+ "learning_rate": 3.709837225760793e-05,
729
+ "loss": 0.0083,
730
+ "step": 4550
731
+ },
732
+ {
733
+ "epoch": 3.21,
734
+ "learning_rate": 3.6390658174097663e-05,
735
+ "loss": 0.0093,
736
+ "step": 4600
737
+ },
738
+ {
739
+ "epoch": 3.21,
740
+ "eval_loss": 0.012657725252211094,
741
+ "eval_runtime": 1090.8271,
742
+ "eval_samples_per_second": 6.757,
743
+ "eval_steps_per_second": 0.282,
744
+ "step": 4600
745
+ },
746
+ {
747
+ "epoch": 3.24,
748
+ "learning_rate": 3.5682944090587406e-05,
749
+ "loss": 0.0099,
750
+ "step": 4650
751
+ },
752
+ {
753
+ "epoch": 3.28,
754
+ "learning_rate": 3.497523000707714e-05,
755
+ "loss": 0.0107,
756
+ "step": 4700
757
+ },
758
+ {
759
+ "epoch": 3.31,
760
+ "learning_rate": 3.426751592356688e-05,
761
+ "loss": 0.0097,
762
+ "step": 4750
763
+ },
764
+ {
765
+ "epoch": 3.35,
766
+ "learning_rate": 3.355980184005662e-05,
767
+ "loss": 0.0099,
768
+ "step": 4800
769
+ },
770
+ {
771
+ "epoch": 3.35,
772
+ "eval_loss": 0.012795734219253063,
773
+ "eval_runtime": 1090.6151,
774
+ "eval_samples_per_second": 6.759,
775
+ "eval_steps_per_second": 0.282,
776
+ "step": 4800
777
+ },
778
+ {
779
+ "epoch": 3.38,
780
+ "learning_rate": 3.285208775654636e-05,
781
+ "loss": 0.0094,
782
+ "step": 4850
783
+ },
784
+ {
785
+ "epoch": 3.42,
786
+ "learning_rate": 3.214437367303609e-05,
787
+ "loss": 0.0099,
788
+ "step": 4900
789
+ },
790
+ {
791
+ "epoch": 3.45,
792
+ "learning_rate": 3.1436659589525835e-05,
793
+ "loss": 0.0097,
794
+ "step": 4950
795
+ },
796
+ {
797
+ "epoch": 3.49,
798
+ "learning_rate": 3.072894550601557e-05,
799
+ "loss": 0.0109,
800
+ "step": 5000
801
+ },
802
+ {
803
+ "epoch": 3.49,
804
+ "eval_loss": 0.01254407037049532,
805
+ "eval_runtime": 1090.2952,
806
+ "eval_samples_per_second": 6.761,
807
+ "eval_steps_per_second": 0.282,
808
+ "step": 5000
809
+ },
810
+ {
811
+ "epoch": 3.52,
812
+ "learning_rate": 3.0021231422505307e-05,
813
+ "loss": 0.009,
814
+ "step": 5050
815
+ },
816
+ {
817
+ "epoch": 3.56,
818
+ "learning_rate": 2.931351733899505e-05,
819
+ "loss": 0.0106,
820
+ "step": 5100
821
+ },
822
+ {
823
+ "epoch": 3.59,
824
+ "learning_rate": 2.8605803255484786e-05,
825
+ "loss": 0.0112,
826
+ "step": 5150
827
+ },
828
+ {
829
+ "epoch": 3.63,
830
+ "learning_rate": 2.7898089171974522e-05,
831
+ "loss": 0.0089,
832
+ "step": 5200
833
+ },
834
+ {
835
+ "epoch": 3.63,
836
+ "eval_loss": 0.012657931074500084,
837
+ "eval_runtime": 1090.5111,
838
+ "eval_samples_per_second": 6.759,
839
+ "eval_steps_per_second": 0.282,
840
+ "step": 5200
841
+ },
842
+ {
843
+ "epoch": 3.66,
844
+ "learning_rate": 2.7190375088464258e-05,
845
+ "loss": 0.0088,
846
+ "step": 5250
847
+ },
848
+ {
849
+ "epoch": 3.7,
850
+ "learning_rate": 2.6482661004954e-05,
851
+ "loss": 0.0113,
852
+ "step": 5300
853
+ },
854
+ {
855
+ "epoch": 3.73,
856
+ "learning_rate": 2.5774946921443737e-05,
857
+ "loss": 0.0105,
858
+ "step": 5350
859
+ },
860
+ {
861
+ "epoch": 3.77,
862
+ "learning_rate": 2.5067232837933473e-05,
863
+ "loss": 0.0095,
864
+ "step": 5400
865
+ },
866
+ {
867
+ "epoch": 3.77,
868
+ "eval_loss": 0.012662014923989773,
869
+ "eval_runtime": 1090.6147,
870
+ "eval_samples_per_second": 6.759,
871
+ "eval_steps_per_second": 0.282,
872
+ "step": 5400
873
+ },
874
+ {
875
+ "epoch": 3.8,
876
+ "learning_rate": 2.4359518754423212e-05,
877
+ "loss": 0.0098,
878
+ "step": 5450
879
+ },
880
+ {
881
+ "epoch": 3.84,
882
+ "learning_rate": 2.365180467091295e-05,
883
+ "loss": 0.0083,
884
+ "step": 5500
885
+ },
886
+ {
887
+ "epoch": 3.87,
888
+ "learning_rate": 2.294409058740269e-05,
889
+ "loss": 0.0109,
890
+ "step": 5550
891
+ },
892
+ {
893
+ "epoch": 3.91,
894
+ "learning_rate": 2.2236376503892427e-05,
895
+ "loss": 0.0101,
896
+ "step": 5600
897
+ },
898
+ {
899
+ "epoch": 3.91,
900
+ "eval_loss": 0.012734564021229744,
901
+ "eval_runtime": 1090.6719,
902
+ "eval_samples_per_second": 6.758,
903
+ "eval_steps_per_second": 0.282,
904
+ "step": 5600
905
+ },
906
+ {
907
+ "epoch": 3.94,
908
+ "learning_rate": 2.1528662420382166e-05,
909
+ "loss": 0.0103,
910
+ "step": 5650
911
+ },
912
+ {
913
+ "epoch": 3.98,
914
+ "learning_rate": 2.0820948336871905e-05,
915
+ "loss": 0.0084,
916
+ "step": 5700
917
+ },
918
+ {
919
+ "epoch": 4.01,
920
+ "learning_rate": 2.0113234253361645e-05,
921
+ "loss": 0.0092,
922
+ "step": 5750
923
+ },
924
+ {
925
+ "epoch": 4.05,
926
+ "learning_rate": 1.9405520169851384e-05,
927
+ "loss": 0.0098,
928
+ "step": 5800
929
+ },
930
+ {
931
+ "epoch": 4.05,
932
+ "eval_loss": 0.01291683316230774,
933
+ "eval_runtime": 1090.3976,
934
+ "eval_samples_per_second": 6.76,
935
+ "eval_steps_per_second": 0.282,
936
+ "step": 5800
937
+ },
938
+ {
939
+ "epoch": 4.08,
940
+ "learning_rate": 1.869780608634112e-05,
941
+ "loss": 0.0081,
942
+ "step": 5850
943
+ },
944
+ {
945
+ "epoch": 4.12,
946
+ "learning_rate": 1.799009200283086e-05,
947
+ "loss": 0.0082,
948
+ "step": 5900
949
+ },
950
+ {
951
+ "epoch": 4.15,
952
+ "learning_rate": 1.7282377919320595e-05,
953
+ "loss": 0.0074,
954
+ "step": 5950
955
+ },
956
+ {
957
+ "epoch": 4.19,
958
+ "learning_rate": 1.6574663835810334e-05,
959
+ "loss": 0.0084,
960
+ "step": 6000
961
+ },
962
+ {
963
+ "epoch": 4.19,
964
+ "eval_loss": 0.012655936181545258,
965
+ "eval_runtime": 1090.3951,
966
+ "eval_samples_per_second": 6.76,
967
+ "eval_steps_per_second": 0.282,
968
+ "step": 6000
969
+ },
970
+ {
971
+ "epoch": 4.22,
972
+ "learning_rate": 1.5866949752300074e-05,
973
+ "loss": 0.0084,
974
+ "step": 6050
975
+ },
976
+ {
977
+ "epoch": 4.26,
978
+ "learning_rate": 1.515923566878981e-05,
979
+ "loss": 0.0095,
980
+ "step": 6100
981
+ },
982
+ {
983
+ "epoch": 4.29,
984
+ "learning_rate": 1.4451521585279547e-05,
985
+ "loss": 0.0108,
986
+ "step": 6150
987
+ },
988
+ {
989
+ "epoch": 4.33,
990
+ "learning_rate": 1.3743807501769285e-05,
991
+ "loss": 0.0082,
992
+ "step": 6200
993
+ },
994
+ {
995
+ "epoch": 4.33,
996
+ "eval_loss": 0.012752115726470947,
997
+ "eval_runtime": 1090.0724,
998
+ "eval_samples_per_second": 6.762,
999
+ "eval_steps_per_second": 0.283,
1000
+ "step": 6200
1001
+ },
1002
+ {
1003
+ "epoch": 4.36,
1004
+ "learning_rate": 1.3036093418259024e-05,
1005
+ "loss": 0.0091,
1006
+ "step": 6250
1007
+ },
1008
+ {
1009
+ "epoch": 4.39,
1010
+ "learning_rate": 1.2328379334748762e-05,
1011
+ "loss": 0.0083,
1012
+ "step": 6300
1013
+ },
1014
+ {
1015
+ "epoch": 4.43,
1016
+ "learning_rate": 1.1620665251238501e-05,
1017
+ "loss": 0.0082,
1018
+ "step": 6350
1019
+ },
1020
+ {
1021
+ "epoch": 4.46,
1022
+ "learning_rate": 1.0912951167728239e-05,
1023
+ "loss": 0.0088,
1024
+ "step": 6400
1025
+ },
1026
+ {
1027
+ "epoch": 4.46,
1028
+ "eval_loss": 0.013044040650129318,
1029
+ "eval_runtime": 1090.0452,
1030
+ "eval_samples_per_second": 6.762,
1031
+ "eval_steps_per_second": 0.283,
1032
+ "step": 6400
1033
+ },
1034
+ {
1035
+ "epoch": 4.5,
1036
+ "learning_rate": 1.0205237084217977e-05,
1037
+ "loss": 0.0084,
1038
+ "step": 6450
1039
+ },
1040
+ {
1041
+ "epoch": 4.53,
1042
+ "learning_rate": 9.497523000707714e-06,
1043
+ "loss": 0.0086,
1044
+ "step": 6500
1045
+ },
1046
+ {
1047
+ "epoch": 4.57,
1048
+ "learning_rate": 8.789808917197452e-06,
1049
+ "loss": 0.0089,
1050
+ "step": 6550
1051
+ },
1052
+ {
1053
+ "epoch": 4.6,
1054
+ "learning_rate": 8.082094833687191e-06,
1055
+ "loss": 0.0077,
1056
+ "step": 6600
1057
+ },
1058
+ {
1059
+ "epoch": 4.6,
1060
+ "eval_loss": 0.012679013423621655,
1061
+ "eval_runtime": 1089.906,
1062
+ "eval_samples_per_second": 6.763,
1063
+ "eval_steps_per_second": 0.283,
1064
+ "step": 6600
1065
+ },
1066
+ {
1067
+ "epoch": 4.64,
1068
+ "learning_rate": 7.374380750176929e-06,
1069
+ "loss": 0.0086,
1070
+ "step": 6650
1071
+ },
1072
+ {
1073
+ "epoch": 4.67,
1074
+ "learning_rate": 6.666666666666667e-06,
1075
+ "loss": 0.0075,
1076
+ "step": 6700
1077
+ },
1078
+ {
1079
+ "epoch": 4.71,
1080
+ "learning_rate": 5.958952583156405e-06,
1081
+ "loss": 0.008,
1082
+ "step": 6750
1083
+ },
1084
+ {
1085
+ "epoch": 4.74,
1086
+ "learning_rate": 5.251238499646143e-06,
1087
+ "loss": 0.0082,
1088
+ "step": 6800
1089
+ },
1090
+ {
1091
+ "epoch": 4.74,
1092
+ "eval_loss": 0.01266387477517128,
1093
+ "eval_runtime": 1090.0881,
1094
+ "eval_samples_per_second": 6.762,
1095
+ "eval_steps_per_second": 0.283,
1096
+ "step": 6800
1097
+ },
1098
+ {
1099
+ "epoch": 4.78,
1100
+ "learning_rate": 4.543524416135881e-06,
1101
+ "loss": 0.0075,
1102
+ "step": 6850
1103
+ },
1104
+ {
1105
+ "epoch": 4.81,
1106
+ "learning_rate": 3.83581033262562e-06,
1107
+ "loss": 0.0077,
1108
+ "step": 6900
1109
+ },
1110
+ {
1111
+ "epoch": 4.85,
1112
+ "learning_rate": 3.1280962491153574e-06,
1113
+ "loss": 0.0089,
1114
+ "step": 6950
1115
+ },
1116
+ {
1117
+ "epoch": 4.88,
1118
+ "learning_rate": 2.420382165605096e-06,
1119
+ "loss": 0.0093,
1120
+ "step": 7000
1121
+ },
1122
+ {
1123
+ "epoch": 4.88,
1124
+ "eval_loss": 0.012687072157859802,
1125
+ "eval_runtime": 1089.9152,
1126
+ "eval_samples_per_second": 6.763,
1127
+ "eval_steps_per_second": 0.283,
1128
+ "step": 7000
1129
+ }
1130
+ ],
1131
+ "max_steps": 7165,
1132
+ "num_train_epochs": 5,
1133
+ "total_flos": 1.675440592244441e+18,
1134
+ "trial_name": null,
1135
+ "trial_params": null
1136
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be85c4a7453a1f2bd13f0244b0d92e7983d9faea99128c2b72b90646cad396f2
3
+ size 4027