AlekseyKorshuk commited on
Commit
d73d573
1 Parent(s): f9b4b82

End of training

Browse files
Files changed (3) hide show
  1. all_results.json +8 -0
  2. train_results.json +8 -0
  3. trainer_state.json +1330 -0
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.8617539679867097,
4
+ "train_runtime": 1017.4729,
5
+ "train_samples": 459,
6
+ "train_samples_per_second": 1.353,
7
+ "train_steps_per_second": 0.086
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.8617539679867097,
4
+ "train_runtime": 1017.4729,
5
+ "train_samples": 459,
6
+ "train_samples_per_second": 1.353,
7
+ "train_steps_per_second": 0.086
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 87,
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": 3e-05,
13
+ "loss": 2.6543,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "eval_accuracy": 0.05133992343294669,
19
+ "eval_loss": 2.611328125,
20
+ "eval_runtime": 3.8997,
21
+ "eval_samples_per_second": 22.053,
22
+ "eval_steps_per_second": 1.539,
23
+ "step": 1
24
+ },
25
+ {
26
+ "epoch": 0.07,
27
+ "learning_rate": 3e-05,
28
+ "loss": 2.6077,
29
+ "step": 2
30
+ },
31
+ {
32
+ "epoch": 0.07,
33
+ "eval_accuracy": 0.05133992343294669,
34
+ "eval_loss": 2.611328125,
35
+ "eval_runtime": 3.9386,
36
+ "eval_samples_per_second": 21.835,
37
+ "eval_steps_per_second": 1.523,
38
+ "step": 2
39
+ },
40
+ {
41
+ "epoch": 0.1,
42
+ "learning_rate": 2.9990221430845156e-05,
43
+ "loss": 2.5964,
44
+ "step": 3
45
+ },
46
+ {
47
+ "epoch": 0.1,
48
+ "eval_accuracy": 0.05192560425118565,
49
+ "eval_loss": 2.560546875,
50
+ "eval_runtime": 3.8896,
51
+ "eval_samples_per_second": 22.11,
52
+ "eval_steps_per_second": 1.543,
53
+ "step": 3
54
+ },
55
+ {
56
+ "epoch": 0.14,
57
+ "learning_rate": 2.996089847276925e-05,
58
+ "loss": 2.7302,
59
+ "step": 4
60
+ },
61
+ {
62
+ "epoch": 0.14,
63
+ "eval_accuracy": 0.052582709559453746,
64
+ "eval_loss": 2.5234375,
65
+ "eval_runtime": 4.1976,
66
+ "eval_samples_per_second": 20.488,
67
+ "eval_steps_per_second": 1.429,
68
+ "step": 4
69
+ },
70
+ {
71
+ "epoch": 0.17,
72
+ "learning_rate": 2.9912069357315394e-05,
73
+ "loss": 2.7004,
74
+ "step": 5
75
+ },
76
+ {
77
+ "epoch": 0.17,
78
+ "eval_accuracy": 0.05293983200959945,
79
+ "eval_loss": 2.5078125,
80
+ "eval_runtime": 4.2482,
81
+ "eval_samples_per_second": 20.244,
82
+ "eval_steps_per_second": 1.412,
83
+ "step": 5
84
+ },
85
+ {
86
+ "epoch": 0.21,
87
+ "learning_rate": 2.9843797748334563e-05,
88
+ "loss": 2.5681,
89
+ "step": 6
90
+ },
91
+ {
92
+ "epoch": 0.21,
93
+ "eval_accuracy": 0.05318267527569853,
94
+ "eval_loss": 2.494140625,
95
+ "eval_runtime": 4.2937,
96
+ "eval_samples_per_second": 20.029,
97
+ "eval_steps_per_second": 1.397,
98
+ "step": 6
99
+ },
100
+ {
101
+ "epoch": 0.24,
102
+ "learning_rate": 2.975617265898004e-05,
103
+ "loss": 2.6404,
104
+ "step": 7
105
+ },
106
+ {
107
+ "epoch": 0.24,
108
+ "eval_accuracy": 0.05342551854179761,
109
+ "eval_loss": 2.48828125,
110
+ "eval_runtime": 4.3127,
111
+ "eval_samples_per_second": 19.941,
112
+ "eval_steps_per_second": 1.391,
113
+ "step": 7
114
+ },
115
+ {
116
+ "epoch": 0.28,
117
+ "learning_rate": 2.96493083356513e-05,
118
+ "loss": 2.5325,
119
+ "step": 8
120
+ },
121
+ {
122
+ "epoch": 0.28,
123
+ "eval_accuracy": 0.053554082623850065,
124
+ "eval_loss": 2.48046875,
125
+ "eval_runtime": 4.2885,
126
+ "eval_samples_per_second": 20.053,
127
+ "eval_steps_per_second": 1.399,
128
+ "step": 8
129
+ },
130
+ {
131
+ "epoch": 0.31,
132
+ "learning_rate": 2.952334410903845e-05,
133
+ "loss": 2.7205,
134
+ "step": 9
135
+ },
136
+ {
137
+ "epoch": 0.31,
138
+ "eval_accuracy": 0.053554082623850065,
139
+ "eval_loss": 2.474609375,
140
+ "eval_runtime": 3.8974,
141
+ "eval_samples_per_second": 22.066,
142
+ "eval_steps_per_second": 1.539,
143
+ "step": 9
144
+ },
145
+ {
146
+ "epoch": 0.34,
147
+ "learning_rate": 2.937844421246162e-05,
148
+ "loss": 2.5149,
149
+ "step": 10
150
+ },
151
+ {
152
+ "epoch": 0.34,
153
+ "eval_accuracy": 0.05329695445974516,
154
+ "eval_loss": 2.46484375,
155
+ "eval_runtime": 4.2968,
156
+ "eval_samples_per_second": 20.015,
157
+ "eval_steps_per_second": 1.396,
158
+ "step": 10
159
+ },
160
+ {
161
+ "epoch": 0.38,
162
+ "learning_rate": 2.9214797567742036e-05,
163
+ "loss": 2.5017,
164
+ "step": 11
165
+ },
166
+ {
167
+ "epoch": 0.38,
168
+ "eval_accuracy": 0.05345408833780927,
169
+ "eval_loss": 2.451171875,
170
+ "eval_runtime": 4.2885,
171
+ "eval_samples_per_second": 20.054,
172
+ "eval_steps_per_second": 1.399,
173
+ "step": 11
174
+ },
175
+ {
176
+ "epoch": 0.41,
177
+ "learning_rate": 2.9032617538884018e-05,
178
+ "loss": 2.7026,
179
+ "step": 12
180
+ },
181
+ {
182
+ "epoch": 0.41,
183
+ "eval_accuracy": 0.053896920175989946,
184
+ "eval_loss": 2.439453125,
185
+ "eval_runtime": 4.2979,
186
+ "eval_samples_per_second": 20.01,
187
+ "eval_steps_per_second": 1.396,
188
+ "step": 12
189
+ },
190
+ {
191
+ "epoch": 0.45,
192
+ "learning_rate": 2.8832141653888998e-05,
193
+ "loss": 2.5259,
194
+ "step": 13
195
+ },
196
+ {
197
+ "epoch": 0.45,
198
+ "eval_accuracy": 0.05425404262613565,
199
+ "eval_loss": 2.431640625,
200
+ "eval_runtime": 4.2986,
201
+ "eval_samples_per_second": 20.007,
202
+ "eval_steps_per_second": 1.396,
203
+ "step": 13
204
+ },
205
+ {
206
+ "epoch": 0.48,
207
+ "learning_rate": 2.861363129506436e-05,
208
+ "loss": 2.563,
209
+ "step": 14
210
+ },
211
+ {
212
+ "epoch": 0.48,
213
+ "eval_accuracy": 0.05455402548425804,
214
+ "eval_loss": 2.421875,
215
+ "eval_runtime": 4.2879,
216
+ "eval_samples_per_second": 20.056,
217
+ "eval_steps_per_second": 1.399,
218
+ "step": 14
219
+ },
220
+ {
221
+ "epoch": 0.52,
222
+ "learning_rate": 2.8377371358230733e-05,
223
+ "loss": 2.5679,
224
+ "step": 15
225
+ },
226
+ {
227
+ "epoch": 0.52,
228
+ "eval_accuracy": 0.055011142220444544,
229
+ "eval_loss": 2.4140625,
230
+ "eval_runtime": 3.9042,
231
+ "eval_samples_per_second": 22.027,
232
+ "eval_steps_per_second": 1.537,
233
+ "step": 15
234
+ },
235
+ {
236
+ "epoch": 0.55,
237
+ "learning_rate": 2.8123669881272247e-05,
238
+ "loss": 2.3701,
239
+ "step": 16
240
+ },
241
+ {
242
+ "epoch": 0.55,
243
+ "eval_accuracy": 0.05512542140449117,
244
+ "eval_loss": 2.408203125,
245
+ "eval_runtime": 3.9081,
246
+ "eval_samples_per_second": 22.006,
247
+ "eval_steps_per_second": 1.535,
248
+ "step": 16
249
+ },
250
+ {
251
+ "epoch": 0.59,
252
+ "learning_rate": 2.8123669881272247e-05,
253
+ "loss": 2.4739,
254
+ "step": 17
255
+ },
256
+ {
257
+ "epoch": 0.59,
258
+ "eval_accuracy": 0.05512542140449117,
259
+ "eval_loss": 2.408203125,
260
+ "eval_runtime": 4.3256,
261
+ "eval_samples_per_second": 19.882,
262
+ "eval_steps_per_second": 1.387,
263
+ "step": 17
264
+ },
265
+ {
266
+ "epoch": 0.62,
267
+ "learning_rate": 2.7852857642513838e-05,
268
+ "loss": 2.481,
269
+ "step": 18
270
+ },
271
+ {
272
+ "epoch": 0.62,
273
+ "eval_accuracy": 0.054839723444374606,
274
+ "eval_loss": 2.40234375,
275
+ "eval_runtime": 3.8988,
276
+ "eval_samples_per_second": 22.058,
277
+ "eval_steps_per_second": 1.539,
278
+ "step": 18
279
+ },
280
+ {
281
+ "epoch": 0.66,
282
+ "learning_rate": 2.7565287729449473e-05,
283
+ "loss": 2.5795,
284
+ "step": 19
285
+ },
286
+ {
287
+ "epoch": 0.66,
288
+ "eval_accuracy": 0.054896863036397923,
289
+ "eval_loss": 2.39453125,
290
+ "eval_runtime": 4.2984,
291
+ "eval_samples_per_second": 20.007,
292
+ "eval_steps_per_second": 1.396,
293
+ "step": 19
294
+ },
295
+ {
296
+ "epoch": 0.69,
297
+ "learning_rate": 2.7261335078383377e-05,
298
+ "loss": 2.4902,
299
+ "step": 20
300
+ },
301
+ {
302
+ "epoch": 0.69,
303
+ "eval_accuracy": 0.05491114793440375,
304
+ "eval_loss": 2.38671875,
305
+ "eval_runtime": 4.3044,
306
+ "eval_samples_per_second": 19.98,
307
+ "eval_steps_per_second": 1.394,
308
+ "step": 20
309
+ },
310
+ {
311
+ "epoch": 0.72,
312
+ "learning_rate": 2.6941395985584656e-05,
313
+ "loss": 2.4509,
314
+ "step": 21
315
+ },
316
+ {
317
+ "epoch": 0.72,
318
+ "eval_accuracy": 0.055139706302497,
319
+ "eval_loss": 2.380859375,
320
+ "eval_runtime": 4.2949,
321
+ "eval_samples_per_second": 20.024,
322
+ "eval_steps_per_second": 1.397,
323
+ "step": 21
324
+ },
325
+ {
326
+ "epoch": 0.76,
327
+ "learning_rate": 2.6605887590592547e-05,
328
+ "loss": 2.6052,
329
+ "step": 22
330
+ },
331
+ {
332
+ "epoch": 0.76,
333
+ "eval_accuracy": 0.055325409976572766,
334
+ "eval_loss": 2.373046875,
335
+ "eval_runtime": 4.2966,
336
+ "eval_samples_per_second": 20.016,
337
+ "eval_steps_per_second": 1.396,
338
+ "step": 22
339
+ },
340
+ {
341
+ "epoch": 0.79,
342
+ "learning_rate": 2.6255247332346036e-05,
343
+ "loss": 2.3323,
344
+ "step": 23
345
+ },
346
+ {
347
+ "epoch": 0.79,
348
+ "eval_accuracy": 0.05552539854865436,
349
+ "eval_loss": 2.36328125,
350
+ "eval_runtime": 3.5753,
351
+ "eval_samples_per_second": 24.054,
352
+ "eval_steps_per_second": 1.678,
353
+ "step": 23
354
+ },
355
+ {
356
+ "epoch": 0.83,
357
+ "learning_rate": 2.5889932378846963e-05,
358
+ "loss": 2.5994,
359
+ "step": 24
360
+ },
361
+ {
362
+ "epoch": 0.83,
363
+ "eval_accuracy": 0.05563967773270099,
364
+ "eval_loss": 2.35546875,
365
+ "eval_runtime": 4.2029,
366
+ "eval_samples_per_second": 20.462,
367
+ "eval_steps_per_second": 1.428,
368
+ "step": 24
369
+ },
370
+ {
371
+ "epoch": 0.86,
372
+ "learning_rate": 2.5510419031100137e-05,
373
+ "loss": 2.3347,
374
+ "step": 25
375
+ },
376
+ {
377
+ "epoch": 0.86,
378
+ "eval_accuracy": 0.05558253814067768,
379
+ "eval_loss": 2.34765625,
380
+ "eval_runtime": 4.4126,
381
+ "eval_samples_per_second": 19.49,
382
+ "eval_steps_per_second": 1.36,
383
+ "step": 25
384
+ },
385
+ {
386
+ "epoch": 0.9,
387
+ "learning_rate": 2.5117202102107707e-05,
388
+ "loss": 2.421,
389
+ "step": 26
390
+ },
391
+ {
392
+ "epoch": 0.9,
393
+ "eval_accuracy": 0.05586823610079424,
394
+ "eval_loss": 2.33984375,
395
+ "eval_runtime": 4.3,
396
+ "eval_samples_per_second": 20.0,
397
+ "eval_steps_per_second": 1.395,
398
+ "step": 26
399
+ },
400
+ {
401
+ "epoch": 0.93,
402
+ "learning_rate": 2.4710794271727415e-05,
403
+ "loss": 2.5337,
404
+ "step": 27
405
+ },
406
+ {
407
+ "epoch": 0.93,
408
+ "eval_accuracy": 0.05603965487686418,
409
+ "eval_loss": 2.3359375,
410
+ "eval_runtime": 3.2828,
411
+ "eval_samples_per_second": 26.197,
412
+ "eval_steps_per_second": 1.828,
413
+ "step": 27
414
+ },
415
+ {
416
+ "epoch": 0.97,
417
+ "learning_rate": 2.4291725418235848e-05,
418
+ "loss": 2.4102,
419
+ "step": 28
420
+ },
421
+ {
422
+ "epoch": 0.97,
423
+ "eval_accuracy": 0.0562539283469516,
424
+ "eval_loss": 2.33203125,
425
+ "eval_runtime": 4.3038,
426
+ "eval_samples_per_second": 19.982,
427
+ "eval_steps_per_second": 1.394,
428
+ "step": 28
429
+ },
430
+ {
431
+ "epoch": 1.0,
432
+ "learning_rate": 2.3860541927468265e-05,
433
+ "loss": 2.4309,
434
+ "step": 29
435
+ },
436
+ {
437
+ "epoch": 1.0,
438
+ "eval_accuracy": 0.05643963202102737,
439
+ "eval_loss": 2.326171875,
440
+ "eval_runtime": 4.2875,
441
+ "eval_samples_per_second": 20.058,
442
+ "eval_steps_per_second": 1.399,
443
+ "step": 29
444
+ },
445
+ {
446
+ "epoch": 1.03,
447
+ "learning_rate": 2.341780598043574e-05,
448
+ "loss": 1.9305,
449
+ "step": 30
450
+ },
451
+ {
452
+ "epoch": 1.03,
453
+ "eval_accuracy": 0.0563539226329924,
454
+ "eval_loss": 2.322265625,
455
+ "eval_runtime": 4.2962,
456
+ "eval_samples_per_second": 20.017,
457
+ "eval_steps_per_second": 1.397,
458
+ "step": 30
459
+ },
460
+ {
461
+ "epoch": 1.07,
462
+ "learning_rate": 2.2964094820348302e-05,
463
+ "loss": 1.8601,
464
+ "step": 31
465
+ },
466
+ {
467
+ "epoch": 1.07,
468
+ "eval_accuracy": 0.056696760185132276,
469
+ "eval_loss": 2.3203125,
470
+ "eval_runtime": 4.32,
471
+ "eval_samples_per_second": 19.907,
472
+ "eval_steps_per_second": 1.389,
473
+ "step": 31
474
+ },
475
+ {
476
+ "epoch": 1.1,
477
+ "learning_rate": 2.25e-05,
478
+ "loss": 1.8682,
479
+ "step": 32
480
+ },
481
+ {
482
+ "epoch": 1.1,
483
+ "eval_accuracy": 0.056425347123021545,
484
+ "eval_loss": 2.328125,
485
+ "eval_runtime": 4.3035,
486
+ "eval_samples_per_second": 19.984,
487
+ "eval_steps_per_second": 1.394,
488
+ "step": 32
489
+ },
490
+ {
491
+ "epoch": 1.14,
492
+ "learning_rate": 2.2026126610496852e-05,
493
+ "loss": 1.8657,
494
+ "step": 33
495
+ },
496
+ {
497
+ "epoch": 1.14,
498
+ "eval_accuracy": 0.056382492429004054,
499
+ "eval_loss": 2.353515625,
500
+ "eval_runtime": 4.2925,
501
+ "eval_samples_per_second": 20.035,
502
+ "eval_steps_per_second": 1.398,
503
+ "step": 33
504
+ },
505
+ {
506
+ "epoch": 1.17,
507
+ "learning_rate": 2.154309249233351e-05,
508
+ "loss": 2.063,
509
+ "step": 34
510
+ },
511
+ {
512
+ "epoch": 1.17,
513
+ "eval_accuracy": 0.056696760185132276,
514
+ "eval_loss": 2.33984375,
515
+ "eval_runtime": 3.2879,
516
+ "eval_samples_per_second": 26.157,
517
+ "eval_steps_per_second": 1.825,
518
+ "step": 34
519
+ },
520
+ {
521
+ "epoch": 1.21,
522
+ "learning_rate": 2.105152742984713e-05,
523
+ "loss": 1.6443,
524
+ "step": 35
525
+ },
526
+ {
527
+ "epoch": 1.21,
528
+ "eval_accuracy": 0.05683960916519056,
529
+ "eval_loss": 2.32421875,
530
+ "eval_runtime": 4.3025,
531
+ "eval_samples_per_second": 19.988,
532
+ "eval_steps_per_second": 1.395,
533
+ "step": 35
534
+ },
535
+ {
536
+ "epoch": 1.24,
537
+ "learning_rate": 2.055207233009872e-05,
538
+ "loss": 1.7592,
539
+ "step": 36
540
+ },
541
+ {
542
+ "epoch": 1.24,
543
+ "eval_accuracy": 0.056882463859208046,
544
+ "eval_loss": 2.31640625,
545
+ "eval_runtime": 4.2776,
546
+ "eval_samples_per_second": 20.105,
547
+ "eval_steps_per_second": 1.403,
548
+ "step": 36
549
+ },
550
+ {
551
+ "epoch": 1.28,
552
+ "learning_rate": 2.0045378387252624e-05,
553
+ "loss": 1.8981,
554
+ "step": 37
555
+ },
556
+ {
557
+ "epoch": 1.28,
558
+ "eval_accuracy": 0.05686817896120222,
559
+ "eval_loss": 2.310546875,
560
+ "eval_runtime": 3.5855,
561
+ "eval_samples_per_second": 23.986,
562
+ "eval_steps_per_second": 1.673,
563
+ "step": 37
564
+ },
565
+ {
566
+ "epoch": 1.31,
567
+ "learning_rate": 1.953210623354359e-05,
568
+ "loss": 1.9379,
569
+ "step": 38
570
+ },
571
+ {
572
+ "epoch": 1.31,
573
+ "eval_accuracy": 0.05728244100337124,
574
+ "eval_loss": 2.3046875,
575
+ "eval_runtime": 4.2888,
576
+ "eval_samples_per_second": 20.052,
577
+ "eval_steps_per_second": 1.399,
578
+ "step": 38
579
+ },
580
+ {
581
+ "epoch": 1.34,
582
+ "learning_rate": 1.9012925077938318e-05,
583
+ "loss": 1.6008,
584
+ "step": 39
585
+ },
586
+ {
587
+ "epoch": 1.34,
588
+ "eval_accuracy": 0.057368150391406206,
589
+ "eval_loss": 2.302734375,
590
+ "eval_runtime": 4.1858,
591
+ "eval_samples_per_second": 20.546,
592
+ "eval_steps_per_second": 1.433,
593
+ "step": 39
594
+ },
595
+ {
596
+ "epoch": 1.38,
597
+ "learning_rate": 1.848851183361466e-05,
598
+ "loss": 1.595,
599
+ "step": 40
600
+ },
601
+ {
602
+ "epoch": 1.38,
603
+ "eval_accuracy": 0.057453859779441174,
604
+ "eval_loss": 2.302734375,
605
+ "eval_runtime": 4.3045,
606
+ "eval_samples_per_second": 19.979,
607
+ "eval_steps_per_second": 1.394,
608
+ "step": 40
609
+ },
610
+ {
611
+ "epoch": 1.41,
612
+ "learning_rate": 1.7959550235396002e-05,
613
+ "loss": 1.7096,
614
+ "step": 41
615
+ },
616
+ {
617
+ "epoch": 1.41,
618
+ "eval_accuracy": 0.05752528426947032,
619
+ "eval_loss": 2.302734375,
620
+ "eval_runtime": 4.3092,
621
+ "eval_samples_per_second": 19.957,
622
+ "eval_steps_per_second": 1.392,
623
+ "step": 41
624
+ },
625
+ {
626
+ "epoch": 1.45,
627
+ "learning_rate": 1.7426729948291474e-05,
628
+ "loss": 1.7245,
629
+ "step": 42
630
+ },
631
+ {
632
+ "epoch": 1.45,
633
+ "eval_accuracy": 0.0575681389634878,
634
+ "eval_loss": 2.302734375,
635
+ "eval_runtime": 3.5821,
636
+ "eval_samples_per_second": 24.008,
637
+ "eval_steps_per_second": 1.675,
638
+ "step": 42
639
+ },
640
+ {
641
+ "epoch": 1.48,
642
+ "learning_rate": 1.689074566830434e-05,
643
+ "loss": 1.795,
644
+ "step": 43
645
+ },
646
+ {
647
+ "epoch": 1.48,
648
+ "eval_accuracy": 0.05768241814753443,
649
+ "eval_loss": 2.30078125,
650
+ "eval_runtime": 4.2953,
651
+ "eval_samples_per_second": 20.022,
652
+ "eval_steps_per_second": 1.397,
653
+ "step": 43
654
+ },
655
+ {
656
+ "epoch": 1.52,
657
+ "learning_rate": 1.635229621668098e-05,
658
+ "loss": 1.7241,
659
+ "step": 44
660
+ },
661
+ {
662
+ "epoch": 1.52,
663
+ "eval_accuracy": 0.057610993657505286,
664
+ "eval_loss": 2.30078125,
665
+ "eval_runtime": 4.3019,
666
+ "eval_samples_per_second": 19.991,
667
+ "eval_steps_per_second": 1.395,
668
+ "step": 44
669
+ },
670
+ {
671
+ "epoch": 1.55,
672
+ "learning_rate": 1.5812083628781265e-05,
673
+ "loss": 1.6356,
674
+ "step": 45
675
+ },
676
+ {
677
+ "epoch": 1.55,
678
+ "eval_accuracy": 0.057639563453516944,
679
+ "eval_loss": 2.298828125,
680
+ "eval_runtime": 4.2961,
681
+ "eval_samples_per_second": 20.018,
682
+ "eval_steps_per_second": 1.397,
683
+ "step": 45
684
+ },
685
+ {
686
+ "epoch": 1.59,
687
+ "learning_rate": 1.5270812238758407e-05,
688
+ "loss": 1.77,
689
+ "step": 46
690
+ },
691
+ {
692
+ "epoch": 1.59,
693
+ "eval_accuracy": 0.057553854065481976,
694
+ "eval_loss": 2.296875,
695
+ "eval_runtime": 4.3142,
696
+ "eval_samples_per_second": 19.934,
697
+ "eval_steps_per_second": 1.391,
698
+ "step": 46
699
+ },
700
+ {
701
+ "epoch": 1.62,
702
+ "learning_rate": 1.4729187761241592e-05,
703
+ "loss": 1.6675,
704
+ "step": 47
705
+ },
706
+ {
707
+ "epoch": 1.62,
708
+ "eval_accuracy": 0.057668133249528596,
709
+ "eval_loss": 2.29296875,
710
+ "eval_runtime": 4.2943,
711
+ "eval_samples_per_second": 20.026,
712
+ "eval_steps_per_second": 1.397,
713
+ "step": 47
714
+ },
715
+ {
716
+ "epoch": 1.66,
717
+ "learning_rate": 1.4187916371218739e-05,
718
+ "loss": 1.6929,
719
+ "step": 48
720
+ },
721
+ {
722
+ "epoch": 1.66,
723
+ "eval_accuracy": 0.05771098794354608,
724
+ "eval_loss": 2.291015625,
725
+ "eval_runtime": 3.8871,
726
+ "eval_samples_per_second": 22.124,
727
+ "eval_steps_per_second": 1.544,
728
+ "step": 48
729
+ },
730
+ {
731
+ "epoch": 1.69,
732
+ "learning_rate": 1.3647703783319022e-05,
733
+ "loss": 1.6635,
734
+ "step": 49
735
+ },
736
+ {
737
+ "epoch": 1.69,
738
+ "eval_accuracy": 0.05762527855551111,
739
+ "eval_loss": 2.291015625,
740
+ "eval_runtime": 4.3038,
741
+ "eval_samples_per_second": 19.982,
742
+ "eval_steps_per_second": 1.394,
743
+ "step": 49
744
+ },
745
+ {
746
+ "epoch": 1.72,
747
+ "learning_rate": 1.310925433169566e-05,
748
+ "loss": 1.6093,
749
+ "step": 50
750
+ },
751
+ {
752
+ "epoch": 1.72,
753
+ "eval_accuracy": 0.05781098222958688,
754
+ "eval_loss": 2.291015625,
755
+ "eval_runtime": 3.8811,
756
+ "eval_samples_per_second": 22.159,
757
+ "eval_steps_per_second": 1.546,
758
+ "step": 50
759
+ },
760
+ {
761
+ "epoch": 1.76,
762
+ "learning_rate": 1.2573270051708529e-05,
763
+ "loss": 1.7362,
764
+ "step": 51
765
+ },
766
+ {
767
+ "epoch": 1.76,
768
+ "eval_accuracy": 0.05796811610765099,
769
+ "eval_loss": 2.2890625,
770
+ "eval_runtime": 4.3003,
771
+ "eval_samples_per_second": 19.998,
772
+ "eval_steps_per_second": 1.395,
773
+ "step": 51
774
+ },
775
+ {
776
+ "epoch": 1.79,
777
+ "learning_rate": 1.2040449764604002e-05,
778
+ "loss": 1.7015,
779
+ "step": 52
780
+ },
781
+ {
782
+ "epoch": 1.79,
783
+ "eval_accuracy": 0.05805382549568596,
784
+ "eval_loss": 2.28515625,
785
+ "eval_runtime": 4.2877,
786
+ "eval_samples_per_second": 20.057,
787
+ "eval_steps_per_second": 1.399,
788
+ "step": 52
789
+ },
790
+ {
791
+ "epoch": 1.83,
792
+ "learning_rate": 1.1511488166385349e-05,
793
+ "loss": 1.9515,
794
+ "step": 53
795
+ },
796
+ {
797
+ "epoch": 1.83,
798
+ "eval_accuracy": 0.05816810467973259,
799
+ "eval_loss": 2.28125,
800
+ "eval_runtime": 4.3062,
801
+ "eval_samples_per_second": 19.971,
802
+ "eval_steps_per_second": 1.393,
803
+ "step": 53
804
+ },
805
+ {
806
+ "epoch": 1.86,
807
+ "learning_rate": 1.098707492206169e-05,
808
+ "loss": 1.6494,
809
+ "step": 54
810
+ },
811
+ {
812
+ "epoch": 1.86,
813
+ "eval_accuracy": 0.05801097080166848,
814
+ "eval_loss": 2.27734375,
815
+ "eval_runtime": 3.8918,
816
+ "eval_samples_per_second": 22.098,
817
+ "eval_steps_per_second": 1.542,
818
+ "step": 54
819
+ },
820
+ {
821
+ "epoch": 1.9,
822
+ "learning_rate": 1.0467893766456408e-05,
823
+ "loss": 1.7522,
824
+ "step": 55
825
+ },
826
+ {
827
+ "epoch": 1.9,
828
+ "eval_accuracy": 0.05798240100565682,
829
+ "eval_loss": 2.2734375,
830
+ "eval_runtime": 4.1919,
831
+ "eval_samples_per_second": 20.516,
832
+ "eval_steps_per_second": 1.431,
833
+ "step": 55
834
+ },
835
+ {
836
+ "epoch": 1.93,
837
+ "learning_rate": 9.954621612747371e-06,
838
+ "loss": 1.7369,
839
+ "step": 56
840
+ },
841
+ {
842
+ "epoch": 1.93,
843
+ "eval_accuracy": 0.05806811039369179,
844
+ "eval_loss": 2.267578125,
845
+ "eval_runtime": 4.3092,
846
+ "eval_samples_per_second": 19.957,
847
+ "eval_steps_per_second": 1.392,
848
+ "step": 56
849
+ },
850
+ {
851
+ "epoch": 1.97,
852
+ "learning_rate": 9.447927669901284e-06,
853
+ "loss": 1.6528,
854
+ "step": 57
855
+ },
856
+ {
857
+ "epoch": 1.97,
858
+ "eval_accuracy": 0.058125249985715104,
859
+ "eval_loss": 2.263671875,
860
+ "eval_runtime": 3.8882,
861
+ "eval_samples_per_second": 22.118,
862
+ "eval_steps_per_second": 1.543,
863
+ "step": 57
864
+ },
865
+ {
866
+ "epoch": 2.0,
867
+ "learning_rate": 8.948472570152874e-06,
868
+ "loss": 1.51,
869
+ "step": 58
870
+ },
871
+ {
872
+ "epoch": 2.0,
873
+ "eval_accuracy": 0.05826809896577338,
874
+ "eval_loss": 2.26171875,
875
+ "eval_runtime": 4.2958,
876
+ "eval_samples_per_second": 20.019,
877
+ "eval_steps_per_second": 1.397,
878
+ "step": 58
879
+ },
880
+ {
881
+ "epoch": 2.03,
882
+ "learning_rate": 8.456907507666488e-06,
883
+ "loss": 1.4579,
884
+ "step": 59
885
+ },
886
+ {
887
+ "epoch": 2.03,
888
+ "eval_accuracy": 0.05845380263984915,
889
+ "eval_loss": 2.263671875,
890
+ "eval_runtime": 4.2959,
891
+ "eval_samples_per_second": 20.019,
892
+ "eval_steps_per_second": 1.397,
893
+ "step": 59
894
+ },
895
+ {
896
+ "epoch": 2.07,
897
+ "learning_rate": 7.97387338950315e-06,
898
+ "loss": 1.2645,
899
+ "step": 60
900
+ },
901
+ {
902
+ "epoch": 2.07,
903
+ "eval_accuracy": 0.05851094223187246,
904
+ "eval_loss": 2.26953125,
905
+ "eval_runtime": 4.2999,
906
+ "eval_samples_per_second": 20.0,
907
+ "eval_steps_per_second": 1.395,
908
+ "step": 60
909
+ },
910
+ {
911
+ "epoch": 2.1,
912
+ "learning_rate": 7.500000000000004e-06,
913
+ "loss": 1.2424,
914
+ "step": 61
915
+ },
916
+ {
917
+ "epoch": 2.1,
918
+ "eval_accuracy": 0.05839666304782584,
919
+ "eval_loss": 2.27734375,
920
+ "eval_runtime": 4.1932,
921
+ "eval_samples_per_second": 20.51,
922
+ "eval_steps_per_second": 1.431,
923
+ "step": 61
924
+ },
925
+ {
926
+ "epoch": 2.14,
927
+ "learning_rate": 7.035905179651701e-06,
928
+ "loss": 1.2117,
929
+ "step": 62
930
+ },
931
+ {
932
+ "epoch": 2.14,
933
+ "eval_accuracy": 0.058425232843837493,
934
+ "eval_loss": 2.2890625,
935
+ "eval_runtime": 4.3132,
936
+ "eval_samples_per_second": 19.939,
937
+ "eval_steps_per_second": 1.391,
938
+ "step": 62
939
+ },
940
+ {
941
+ "epoch": 2.17,
942
+ "learning_rate": 6.582194019564266e-06,
943
+ "loss": 1.4059,
944
+ "step": 63
945
+ },
946
+ {
947
+ "epoch": 2.17,
948
+ "eval_accuracy": 0.058039540597680135,
949
+ "eval_loss": 2.30078125,
950
+ "eval_runtime": 3.5756,
951
+ "eval_samples_per_second": 24.052,
952
+ "eval_steps_per_second": 1.678,
953
+ "step": 63
954
+ },
955
+ {
956
+ "epoch": 2.21,
957
+ "learning_rate": 6.1394580725317366e-06,
958
+ "loss": 1.328,
959
+ "step": 64
960
+ },
961
+ {
962
+ "epoch": 2.21,
963
+ "eval_accuracy": 0.05811096508770927,
964
+ "eval_loss": 2.314453125,
965
+ "eval_runtime": 4.2869,
966
+ "eval_samples_per_second": 20.061,
967
+ "eval_steps_per_second": 1.4,
968
+ "step": 64
969
+ },
970
+ {
971
+ "epoch": 2.24,
972
+ "learning_rate": 5.708274581764155e-06,
973
+ "loss": 1.3436,
974
+ "step": 65
975
+ },
976
+ {
977
+ "epoch": 2.24,
978
+ "eval_accuracy": 0.05795383120964517,
979
+ "eval_loss": 2.328125,
980
+ "eval_runtime": 3.8909,
981
+ "eval_samples_per_second": 22.103,
982
+ "eval_steps_per_second": 1.542,
983
+ "step": 65
984
+ },
985
+ {
986
+ "epoch": 2.28,
987
+ "learning_rate": 5.289205728272587e-06,
988
+ "loss": 1.389,
989
+ "step": 66
990
+ },
991
+ {
992
+ "epoch": 2.28,
993
+ "eval_accuracy": 0.058039540597680135,
994
+ "eval_loss": 2.337890625,
995
+ "eval_runtime": 4.3156,
996
+ "eval_samples_per_second": 19.928,
997
+ "eval_steps_per_second": 1.39,
998
+ "step": 66
999
+ },
1000
+ {
1001
+ "epoch": 2.31,
1002
+ "learning_rate": 4.882797897892293e-06,
1003
+ "loss": 1.2127,
1004
+ "step": 67
1005
+ },
1006
+ {
1007
+ "epoch": 2.31,
1008
+ "eval_accuracy": 0.0580252556996743,
1009
+ "eval_loss": 2.33984375,
1010
+ "eval_runtime": 4.3,
1011
+ "eval_samples_per_second": 20.0,
1012
+ "eval_steps_per_second": 1.395,
1013
+ "step": 67
1014
+ },
1015
+ {
1016
+ "epoch": 2.34,
1017
+ "learning_rate": 4.4895809688998655e-06,
1018
+ "loss": 1.3645,
1019
+ "step": 68
1020
+ },
1021
+ {
1022
+ "epoch": 2.34,
1023
+ "eval_accuracy": 0.058096680189703445,
1024
+ "eval_loss": 2.341796875,
1025
+ "eval_runtime": 4.216,
1026
+ "eval_samples_per_second": 20.398,
1027
+ "eval_steps_per_second": 1.423,
1028
+ "step": 68
1029
+ },
1030
+ {
1031
+ "epoch": 2.38,
1032
+ "learning_rate": 4.110067621153041e-06,
1033
+ "loss": 1.3389,
1034
+ "step": 69
1035
+ },
1036
+ {
1037
+ "epoch": 2.38,
1038
+ "eval_accuracy": 0.05805382549568596,
1039
+ "eval_loss": 2.337890625,
1040
+ "eval_runtime": 3.9033,
1041
+ "eval_samples_per_second": 22.032,
1042
+ "eval_steps_per_second": 1.537,
1043
+ "step": 69
1044
+ },
1045
+ {
1046
+ "epoch": 2.41,
1047
+ "learning_rate": 3.744752667653965e-06,
1048
+ "loss": 1.2549,
1049
+ "step": 70
1050
+ },
1051
+ {
1052
+ "epoch": 2.41,
1053
+ "eval_accuracy": 0.05808239529169762,
1054
+ "eval_loss": 2.33203125,
1055
+ "eval_runtime": 3.5918,
1056
+ "eval_samples_per_second": 23.943,
1057
+ "eval_steps_per_second": 1.67,
1058
+ "step": 70
1059
+ },
1060
+ {
1061
+ "epoch": 2.45,
1062
+ "learning_rate": 3.394112409407455e-06,
1063
+ "loss": 1.2193,
1064
+ "step": 71
1065
+ },
1066
+ {
1067
+ "epoch": 2.45,
1068
+ "eval_accuracy": 0.05816810467973259,
1069
+ "eval_loss": 2.328125,
1070
+ "eval_runtime": 4.2985,
1071
+ "eval_samples_per_second": 20.007,
1072
+ "eval_steps_per_second": 1.396,
1073
+ "step": 71
1074
+ },
1075
+ {
1076
+ "epoch": 2.48,
1077
+ "learning_rate": 3.0586040144153436e-06,
1078
+ "loss": 1.3617,
1079
+ "step": 72
1080
+ },
1081
+ {
1082
+ "epoch": 2.48,
1083
+ "eval_accuracy": 0.0583252385577967,
1084
+ "eval_loss": 2.322265625,
1085
+ "eval_runtime": 3.2953,
1086
+ "eval_samples_per_second": 26.097,
1087
+ "eval_steps_per_second": 1.821,
1088
+ "step": 72
1089
+ },
1090
+ {
1091
+ "epoch": 2.52,
1092
+ "learning_rate": 2.7386649216166233e-06,
1093
+ "loss": 1.2336,
1094
+ "step": 73
1095
+ },
1096
+ {
1097
+ "epoch": 2.52,
1098
+ "eval_accuracy": 0.058253814067767556,
1099
+ "eval_loss": 2.318359375,
1100
+ "eval_runtime": 4.1807,
1101
+ "eval_samples_per_second": 20.571,
1102
+ "eval_steps_per_second": 1.435,
1103
+ "step": 73
1104
+ },
1105
+ {
1106
+ "epoch": 2.55,
1107
+ "learning_rate": 2.4347122705505303e-06,
1108
+ "loss": 1.179,
1109
+ "step": 74
1110
+ },
1111
+ {
1112
+ "epoch": 2.55,
1113
+ "eval_accuracy": 0.05829666876178504,
1114
+ "eval_loss": 2.314453125,
1115
+ "eval_runtime": 3.5945,
1116
+ "eval_samples_per_second": 23.926,
1117
+ "eval_steps_per_second": 1.669,
1118
+ "step": 74
1119
+ },
1120
+ {
1121
+ "epoch": 2.59,
1122
+ "learning_rate": 2.1471423574861643e-06,
1123
+ "loss": 1.2468,
1124
+ "step": 75
1125
+ },
1126
+ {
1127
+ "epoch": 2.59,
1128
+ "eval_accuracy": 0.058282383863779215,
1129
+ "eval_loss": 2.3125,
1130
+ "eval_runtime": 3.5979,
1131
+ "eval_samples_per_second": 23.903,
1132
+ "eval_steps_per_second": 1.668,
1133
+ "step": 75
1134
+ },
1135
+ {
1136
+ "epoch": 2.62,
1137
+ "learning_rate": 1.8763301187277554e-06,
1138
+ "loss": 1.3325,
1139
+ "step": 76
1140
+ },
1141
+ {
1142
+ "epoch": 2.62,
1143
+ "eval_accuracy": 0.05829666876178504,
1144
+ "eval_loss": 2.30859375,
1145
+ "eval_runtime": 3.9093,
1146
+ "eval_samples_per_second": 21.999,
1147
+ "eval_steps_per_second": 1.535,
1148
+ "step": 76
1149
+ },
1150
+ {
1151
+ "epoch": 2.66,
1152
+ "learning_rate": 1.6226286417692666e-06,
1153
+ "loss": 1.1471,
1154
+ "step": 77
1155
+ },
1156
+ {
1157
+ "epoch": 2.66,
1158
+ "eval_accuracy": 0.058339523455802525,
1159
+ "eval_loss": 2.306640625,
1160
+ "eval_runtime": 4.2929,
1161
+ "eval_samples_per_second": 20.033,
1162
+ "eval_steps_per_second": 1.398,
1163
+ "step": 77
1164
+ },
1165
+ {
1166
+ "epoch": 2.69,
1167
+ "learning_rate": 1.3863687049356465e-06,
1168
+ "loss": 1.3123,
1169
+ "step": 78
1170
+ },
1171
+ {
1172
+ "epoch": 2.69,
1173
+ "eval_accuracy": 0.0583252385577967,
1174
+ "eval_loss": 2.306640625,
1175
+ "eval_runtime": 4.3098,
1176
+ "eval_samples_per_second": 19.955,
1177
+ "eval_steps_per_second": 1.392,
1178
+ "step": 78
1179
+ },
1180
+ {
1181
+ "epoch": 2.72,
1182
+ "learning_rate": 1.1678583461110026e-06,
1183
+ "loss": 1.3285,
1184
+ "step": 79
1185
+ },
1186
+ {
1187
+ "epoch": 2.72,
1188
+ "eval_accuracy": 0.05845380263984915,
1189
+ "eval_loss": 2.3046875,
1190
+ "eval_runtime": 4.3011,
1191
+ "eval_samples_per_second": 19.995,
1192
+ "eval_steps_per_second": 1.395,
1193
+ "step": 79
1194
+ },
1195
+ {
1196
+ "epoch": 2.76,
1197
+ "learning_rate": 9.67382461115986e-07,
1198
+ "loss": 1.3232,
1199
+ "step": 80
1200
+ },
1201
+ {
1202
+ "epoch": 2.76,
1203
+ "eval_accuracy": 0.05836809325181418,
1204
+ "eval_loss": 2.302734375,
1205
+ "eval_runtime": 4.2978,
1206
+ "eval_samples_per_second": 20.01,
1207
+ "eval_steps_per_second": 1.396,
1208
+ "step": 80
1209
+ },
1210
+ {
1211
+ "epoch": 2.79,
1212
+ "learning_rate": 7.852024322579649e-07,
1213
+ "loss": 1.1228,
1214
+ "step": 81
1215
+ },
1216
+ {
1217
+ "epoch": 2.79,
1218
+ "eval_accuracy": 0.05835380835380835,
1219
+ "eval_loss": 2.302734375,
1220
+ "eval_runtime": 3.6063,
1221
+ "eval_samples_per_second": 23.847,
1222
+ "eval_steps_per_second": 1.664,
1223
+ "step": 81
1224
+ },
1225
+ {
1226
+ "epoch": 2.83,
1227
+ "learning_rate": 6.215557875383804e-07,
1228
+ "loss": 1.3524,
1229
+ "step": 82
1230
+ },
1231
+ {
1232
+ "epoch": 2.83,
1233
+ "eval_accuracy": 0.05839666304782584,
1234
+ "eval_loss": 2.302734375,
1235
+ "eval_runtime": 4.1961,
1236
+ "eval_samples_per_second": 20.495,
1237
+ "eval_steps_per_second": 1.43,
1238
+ "step": 82
1239
+ },
1240
+ {
1241
+ "epoch": 2.86,
1242
+ "learning_rate": 4.766558909615504e-07,
1243
+ "loss": 1.2042,
1244
+ "step": 83
1245
+ },
1246
+ {
1247
+ "epoch": 2.86,
1248
+ "eval_accuracy": 0.058339523455802525,
1249
+ "eval_loss": 2.302734375,
1250
+ "eval_runtime": 4.3056,
1251
+ "eval_samples_per_second": 19.974,
1252
+ "eval_steps_per_second": 1.394,
1253
+ "step": 83
1254
+ },
1255
+ {
1256
+ "epoch": 2.9,
1257
+ "learning_rate": 3.5069166434870014e-07,
1258
+ "loss": 1.3588,
1259
+ "step": 84
1260
+ },
1261
+ {
1262
+ "epoch": 2.9,
1263
+ "eval_accuracy": 0.058339523455802525,
1264
+ "eval_loss": 2.30078125,
1265
+ "eval_runtime": 4.1957,
1266
+ "eval_samples_per_second": 20.497,
1267
+ "eval_steps_per_second": 1.43,
1268
+ "step": 84
1269
+ },
1270
+ {
1271
+ "epoch": 2.93,
1272
+ "learning_rate": 2.438273410199598e-07,
1273
+ "loss": 1.2982,
1274
+ "step": 85
1275
+ },
1276
+ {
1277
+ "epoch": 2.93,
1278
+ "eval_accuracy": 0.058425232843837493,
1279
+ "eval_loss": 2.30078125,
1280
+ "eval_runtime": 3.898,
1281
+ "eval_samples_per_second": 22.062,
1282
+ "eval_steps_per_second": 1.539,
1283
+ "step": 85
1284
+ },
1285
+ {
1286
+ "epoch": 2.97,
1287
+ "learning_rate": 1.5620225166544155e-07,
1288
+ "loss": 1.4373,
1289
+ "step": 86
1290
+ },
1291
+ {
1292
+ "epoch": 2.97,
1293
+ "eval_accuracy": 0.05845380263984915,
1294
+ "eval_loss": 2.30078125,
1295
+ "eval_runtime": 4.3019,
1296
+ "eval_samples_per_second": 19.991,
1297
+ "eval_steps_per_second": 1.395,
1298
+ "step": 86
1299
+ },
1300
+ {
1301
+ "epoch": 3.0,
1302
+ "learning_rate": 8.793064268460605e-08,
1303
+ "loss": 1.3562,
1304
+ "step": 87
1305
+ },
1306
+ {
1307
+ "epoch": 3.0,
1308
+ "eval_accuracy": 0.05841094794583167,
1309
+ "eval_loss": 2.30078125,
1310
+ "eval_runtime": 4.2946,
1311
+ "eval_samples_per_second": 20.025,
1312
+ "eval_steps_per_second": 1.397,
1313
+ "step": 87
1314
+ },
1315
+ {
1316
+ "epoch": 3.0,
1317
+ "step": 87,
1318
+ "total_flos": 4783591391232.0,
1319
+ "train_loss": 1.8617539679867097,
1320
+ "train_runtime": 1017.4729,
1321
+ "train_samples_per_second": 1.353,
1322
+ "train_steps_per_second": 0.086
1323
+ }
1324
+ ],
1325
+ "max_steps": 87,
1326
+ "num_train_epochs": 3,
1327
+ "total_flos": 4783591391232.0,
1328
+ "trial_name": null,
1329
+ "trial_params": null
1330
+ }