enoriega commited on
Commit
ea4d7ae
·
1 Parent(s): 06c1839

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "eval_loss": 0.12549959123134613,
4
+ "eval_runtime": 769.8015,
5
+ "eval_samples": 355106,
6
+ "eval_samples_per_second": 461.296,
7
+ "eval_steps_per_second": 57.663,
8
+ "train_loss": 0.1504179045798317,
9
+ "train_runtime": 22764.4166,
10
+ "train_samples": 500000,
11
+ "train_samples_per_second": 65.892,
12
+ "train_steps_per_second": 0.008
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "eval_loss": 0.12549959123134613,
4
+ "eval_runtime": 769.8015,
5
+ "eval_samples": 355106,
6
+ "eval_samples_per_second": 461.296,
7
+ "eval_steps_per_second": 57.663
8
+ }
runs/Jun06_15-43-20_rogue/events.out.tfevents.1654579087.rogue.404389.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e03e52d41733ca97bd0756268992aecfdfb04a4fc4528664da086c46b850133
3
+ size 311
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "train_loss": 0.1504179045798317,
4
+ "train_runtime": 22764.4166,
5
+ "train_samples": 500000,
6
+ "train_samples_per_second": 65.892,
7
+ "train_steps_per_second": 0.008
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1213 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.992,
5
+ "global_step": 186,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 4.973118279569893e-05,
13
+ "loss": 1.0487,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 4.9462365591397855e-05,
19
+ "loss": 0.4173,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 4.9193548387096775e-05,
25
+ "loss": 0.3554,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.06,
30
+ "learning_rate": 4.89247311827957e-05,
31
+ "loss": 0.2972,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.08,
36
+ "learning_rate": 4.865591397849463e-05,
37
+ "loss": 0.2958,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.1,
42
+ "learning_rate": 4.8387096774193554e-05,
43
+ "loss": 0.2668,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.11,
48
+ "learning_rate": 4.811827956989248e-05,
49
+ "loss": 0.2609,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.13,
54
+ "learning_rate": 4.78494623655914e-05,
55
+ "loss": 0.2561,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.14,
60
+ "learning_rate": 4.7580645161290326e-05,
61
+ "loss": 0.2567,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.16,
66
+ "learning_rate": 4.731182795698925e-05,
67
+ "loss": 0.2422,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.18,
72
+ "learning_rate": 4.704301075268818e-05,
73
+ "loss": 0.236,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.19,
78
+ "learning_rate": 4.67741935483871e-05,
79
+ "loss": 0.2299,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.21,
84
+ "learning_rate": 4.650537634408602e-05,
85
+ "loss": 0.2255,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.22,
90
+ "learning_rate": 4.6236559139784944e-05,
91
+ "loss": 0.2212,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.24,
96
+ "learning_rate": 4.596774193548387e-05,
97
+ "loss": 0.2031,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.26,
102
+ "learning_rate": 4.56989247311828e-05,
103
+ "loss": 0.2157,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.27,
108
+ "learning_rate": 4.543010752688172e-05,
109
+ "loss": 0.192,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 0.29,
114
+ "learning_rate": 4.516129032258064e-05,
115
+ "loss": 0.2297,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 0.3,
120
+ "learning_rate": 4.489247311827957e-05,
121
+ "loss": 0.2266,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 0.32,
126
+ "learning_rate": 4.4623655913978496e-05,
127
+ "loss": 0.1764,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 0.32,
132
+ "eval_loss": 0.2303156703710556,
133
+ "eval_runtime": 771.203,
134
+ "eval_samples_per_second": 460.457,
135
+ "eval_steps_per_second": 57.558,
136
+ "step": 20
137
+ },
138
+ {
139
+ "epoch": 0.34,
140
+ "learning_rate": 4.435483870967742e-05,
141
+ "loss": 0.2118,
142
+ "step": 21
143
+ },
144
+ {
145
+ "epoch": 0.35,
146
+ "learning_rate": 4.408602150537635e-05,
147
+ "loss": 0.2286,
148
+ "step": 22
149
+ },
150
+ {
151
+ "epoch": 0.37,
152
+ "learning_rate": 4.381720430107527e-05,
153
+ "loss": 0.1914,
154
+ "step": 23
155
+ },
156
+ {
157
+ "epoch": 0.38,
158
+ "learning_rate": 4.3548387096774194e-05,
159
+ "loss": 0.1766,
160
+ "step": 24
161
+ },
162
+ {
163
+ "epoch": 0.4,
164
+ "learning_rate": 4.327956989247312e-05,
165
+ "loss": 0.1839,
166
+ "step": 25
167
+ },
168
+ {
169
+ "epoch": 0.42,
170
+ "learning_rate": 4.301075268817205e-05,
171
+ "loss": 0.1815,
172
+ "step": 26
173
+ },
174
+ {
175
+ "epoch": 0.43,
176
+ "learning_rate": 4.2741935483870973e-05,
177
+ "loss": 0.1682,
178
+ "step": 27
179
+ },
180
+ {
181
+ "epoch": 0.45,
182
+ "learning_rate": 4.247311827956989e-05,
183
+ "loss": 0.1707,
184
+ "step": 28
185
+ },
186
+ {
187
+ "epoch": 0.46,
188
+ "learning_rate": 4.220430107526882e-05,
189
+ "loss": 0.1631,
190
+ "step": 29
191
+ },
192
+ {
193
+ "epoch": 0.48,
194
+ "learning_rate": 4.1935483870967746e-05,
195
+ "loss": 0.1576,
196
+ "step": 30
197
+ },
198
+ {
199
+ "epoch": 0.5,
200
+ "learning_rate": 4.166666666666667e-05,
201
+ "loss": 0.1616,
202
+ "step": 31
203
+ },
204
+ {
205
+ "epoch": 0.51,
206
+ "learning_rate": 4.13978494623656e-05,
207
+ "loss": 0.157,
208
+ "step": 32
209
+ },
210
+ {
211
+ "epoch": 0.53,
212
+ "learning_rate": 4.112903225806452e-05,
213
+ "loss": 0.1506,
214
+ "step": 33
215
+ },
216
+ {
217
+ "epoch": 0.54,
218
+ "learning_rate": 4.0860215053763444e-05,
219
+ "loss": 0.1507,
220
+ "step": 34
221
+ },
222
+ {
223
+ "epoch": 0.56,
224
+ "learning_rate": 4.0591397849462364e-05,
225
+ "loss": 0.154,
226
+ "step": 35
227
+ },
228
+ {
229
+ "epoch": 0.58,
230
+ "learning_rate": 4.032258064516129e-05,
231
+ "loss": 0.1496,
232
+ "step": 36
233
+ },
234
+ {
235
+ "epoch": 0.59,
236
+ "learning_rate": 4.005376344086022e-05,
237
+ "loss": 0.1494,
238
+ "step": 37
239
+ },
240
+ {
241
+ "epoch": 0.61,
242
+ "learning_rate": 3.978494623655914e-05,
243
+ "loss": 0.1526,
244
+ "step": 38
245
+ },
246
+ {
247
+ "epoch": 0.62,
248
+ "learning_rate": 3.951612903225806e-05,
249
+ "loss": 0.1472,
250
+ "step": 39
251
+ },
252
+ {
253
+ "epoch": 0.64,
254
+ "learning_rate": 3.924731182795699e-05,
255
+ "loss": 0.145,
256
+ "step": 40
257
+ },
258
+ {
259
+ "epoch": 0.64,
260
+ "eval_loss": 0.14696530997753143,
261
+ "eval_runtime": 771.7715,
262
+ "eval_samples_per_second": 460.118,
263
+ "eval_steps_per_second": 57.516,
264
+ "step": 40
265
+ },
266
+ {
267
+ "epoch": 0.66,
268
+ "learning_rate": 3.8978494623655915e-05,
269
+ "loss": 0.145,
270
+ "step": 41
271
+ },
272
+ {
273
+ "epoch": 0.67,
274
+ "learning_rate": 3.870967741935484e-05,
275
+ "loss": 0.1461,
276
+ "step": 42
277
+ },
278
+ {
279
+ "epoch": 0.69,
280
+ "learning_rate": 3.844086021505376e-05,
281
+ "loss": 0.1447,
282
+ "step": 43
283
+ },
284
+ {
285
+ "epoch": 0.7,
286
+ "learning_rate": 3.817204301075269e-05,
287
+ "loss": 0.1462,
288
+ "step": 44
289
+ },
290
+ {
291
+ "epoch": 0.72,
292
+ "learning_rate": 3.7903225806451614e-05,
293
+ "loss": 0.1391,
294
+ "step": 45
295
+ },
296
+ {
297
+ "epoch": 0.74,
298
+ "learning_rate": 3.763440860215054e-05,
299
+ "loss": 0.143,
300
+ "step": 46
301
+ },
302
+ {
303
+ "epoch": 0.75,
304
+ "learning_rate": 3.736559139784947e-05,
305
+ "loss": 0.1388,
306
+ "step": 47
307
+ },
308
+ {
309
+ "epoch": 0.77,
310
+ "learning_rate": 3.7096774193548386e-05,
311
+ "loss": 0.1448,
312
+ "step": 48
313
+ },
314
+ {
315
+ "epoch": 0.78,
316
+ "learning_rate": 3.682795698924731e-05,
317
+ "loss": 0.141,
318
+ "step": 49
319
+ },
320
+ {
321
+ "epoch": 0.8,
322
+ "learning_rate": 3.655913978494624e-05,
323
+ "loss": 0.1412,
324
+ "step": 50
325
+ },
326
+ {
327
+ "epoch": 0.82,
328
+ "learning_rate": 3.6290322580645165e-05,
329
+ "loss": 0.14,
330
+ "step": 51
331
+ },
332
+ {
333
+ "epoch": 0.83,
334
+ "learning_rate": 3.602150537634409e-05,
335
+ "loss": 0.136,
336
+ "step": 52
337
+ },
338
+ {
339
+ "epoch": 0.85,
340
+ "learning_rate": 3.575268817204301e-05,
341
+ "loss": 0.1401,
342
+ "step": 53
343
+ },
344
+ {
345
+ "epoch": 0.86,
346
+ "learning_rate": 3.548387096774194e-05,
347
+ "loss": 0.137,
348
+ "step": 54
349
+ },
350
+ {
351
+ "epoch": 0.88,
352
+ "learning_rate": 3.5215053763440864e-05,
353
+ "loss": 0.1386,
354
+ "step": 55
355
+ },
356
+ {
357
+ "epoch": 0.9,
358
+ "learning_rate": 3.494623655913979e-05,
359
+ "loss": 0.1348,
360
+ "step": 56
361
+ },
362
+ {
363
+ "epoch": 0.91,
364
+ "learning_rate": 3.467741935483872e-05,
365
+ "loss": 0.1352,
366
+ "step": 57
367
+ },
368
+ {
369
+ "epoch": 0.93,
370
+ "learning_rate": 3.4408602150537636e-05,
371
+ "loss": 0.1319,
372
+ "step": 58
373
+ },
374
+ {
375
+ "epoch": 0.94,
376
+ "learning_rate": 3.4139784946236556e-05,
377
+ "loss": 0.1329,
378
+ "step": 59
379
+ },
380
+ {
381
+ "epoch": 0.96,
382
+ "learning_rate": 3.387096774193548e-05,
383
+ "loss": 0.129,
384
+ "step": 60
385
+ },
386
+ {
387
+ "epoch": 0.96,
388
+ "eval_loss": 0.1321416050195694,
389
+ "eval_runtime": 771.9931,
390
+ "eval_samples_per_second": 459.986,
391
+ "eval_steps_per_second": 57.499,
392
+ "step": 60
393
+ },
394
+ {
395
+ "epoch": 0.98,
396
+ "learning_rate": 3.360215053763441e-05,
397
+ "loss": 0.1312,
398
+ "step": 61
399
+ },
400
+ {
401
+ "epoch": 0.99,
402
+ "learning_rate": 3.3333333333333335e-05,
403
+ "loss": 0.1328,
404
+ "step": 62
405
+ },
406
+ {
407
+ "epoch": 1.02,
408
+ "learning_rate": 3.306451612903226e-05,
409
+ "loss": 0.1941,
410
+ "step": 63
411
+ },
412
+ {
413
+ "epoch": 1.03,
414
+ "learning_rate": 3.279569892473118e-05,
415
+ "loss": 0.1338,
416
+ "step": 64
417
+ },
418
+ {
419
+ "epoch": 1.05,
420
+ "learning_rate": 3.252688172043011e-05,
421
+ "loss": 0.1349,
422
+ "step": 65
423
+ },
424
+ {
425
+ "epoch": 1.06,
426
+ "learning_rate": 3.2258064516129034e-05,
427
+ "loss": 0.1307,
428
+ "step": 66
429
+ },
430
+ {
431
+ "epoch": 1.08,
432
+ "learning_rate": 3.198924731182796e-05,
433
+ "loss": 0.129,
434
+ "step": 67
435
+ },
436
+ {
437
+ "epoch": 1.1,
438
+ "learning_rate": 3.172043010752688e-05,
439
+ "loss": 0.1308,
440
+ "step": 68
441
+ },
442
+ {
443
+ "epoch": 1.11,
444
+ "learning_rate": 3.1451612903225806e-05,
445
+ "loss": 0.1317,
446
+ "step": 69
447
+ },
448
+ {
449
+ "epoch": 1.13,
450
+ "learning_rate": 3.118279569892473e-05,
451
+ "loss": 0.1258,
452
+ "step": 70
453
+ },
454
+ {
455
+ "epoch": 1.14,
456
+ "learning_rate": 3.091397849462366e-05,
457
+ "loss": 0.1271,
458
+ "step": 71
459
+ },
460
+ {
461
+ "epoch": 1.16,
462
+ "learning_rate": 3.0645161290322585e-05,
463
+ "loss": 0.131,
464
+ "step": 72
465
+ },
466
+ {
467
+ "epoch": 1.18,
468
+ "learning_rate": 3.0376344086021508e-05,
469
+ "loss": 0.1274,
470
+ "step": 73
471
+ },
472
+ {
473
+ "epoch": 1.19,
474
+ "learning_rate": 3.010752688172043e-05,
475
+ "loss": 0.1269,
476
+ "step": 74
477
+ },
478
+ {
479
+ "epoch": 1.21,
480
+ "learning_rate": 2.9838709677419357e-05,
481
+ "loss": 0.1248,
482
+ "step": 75
483
+ },
484
+ {
485
+ "epoch": 1.22,
486
+ "learning_rate": 2.9569892473118284e-05,
487
+ "loss": 0.1286,
488
+ "step": 76
489
+ },
490
+ {
491
+ "epoch": 1.24,
492
+ "learning_rate": 2.9301075268817207e-05,
493
+ "loss": 0.1273,
494
+ "step": 77
495
+ },
496
+ {
497
+ "epoch": 1.26,
498
+ "learning_rate": 2.9032258064516133e-05,
499
+ "loss": 0.129,
500
+ "step": 78
501
+ },
502
+ {
503
+ "epoch": 1.27,
504
+ "learning_rate": 2.8763440860215056e-05,
505
+ "loss": 0.1291,
506
+ "step": 79
507
+ },
508
+ {
509
+ "epoch": 1.29,
510
+ "learning_rate": 2.8494623655913982e-05,
511
+ "loss": 0.1256,
512
+ "step": 80
513
+ },
514
+ {
515
+ "epoch": 1.29,
516
+ "eval_loss": 0.126514732837677,
517
+ "eval_runtime": 772.8684,
518
+ "eval_samples_per_second": 459.465,
519
+ "eval_steps_per_second": 57.434,
520
+ "step": 80
521
+ },
522
+ {
523
+ "epoch": 1.3,
524
+ "learning_rate": 2.822580645161291e-05,
525
+ "loss": 0.1245,
526
+ "step": 81
527
+ },
528
+ {
529
+ "epoch": 1.32,
530
+ "learning_rate": 2.7956989247311828e-05,
531
+ "loss": 0.1269,
532
+ "step": 82
533
+ },
534
+ {
535
+ "epoch": 1.34,
536
+ "learning_rate": 2.768817204301075e-05,
537
+ "loss": 0.1297,
538
+ "step": 83
539
+ },
540
+ {
541
+ "epoch": 1.35,
542
+ "learning_rate": 2.7419354838709678e-05,
543
+ "loss": 0.1273,
544
+ "step": 84
545
+ },
546
+ {
547
+ "epoch": 1.37,
548
+ "learning_rate": 2.71505376344086e-05,
549
+ "loss": 0.1231,
550
+ "step": 85
551
+ },
552
+ {
553
+ "epoch": 1.38,
554
+ "learning_rate": 2.6881720430107527e-05,
555
+ "loss": 0.1227,
556
+ "step": 86
557
+ },
558
+ {
559
+ "epoch": 1.4,
560
+ "learning_rate": 2.661290322580645e-05,
561
+ "loss": 0.1275,
562
+ "step": 87
563
+ },
564
+ {
565
+ "epoch": 1.42,
566
+ "learning_rate": 2.6344086021505376e-05,
567
+ "loss": 0.1244,
568
+ "step": 88
569
+ },
570
+ {
571
+ "epoch": 1.43,
572
+ "learning_rate": 2.6075268817204303e-05,
573
+ "loss": 0.1298,
574
+ "step": 89
575
+ },
576
+ {
577
+ "epoch": 1.45,
578
+ "learning_rate": 2.5806451612903226e-05,
579
+ "loss": 0.1246,
580
+ "step": 90
581
+ },
582
+ {
583
+ "epoch": 1.46,
584
+ "learning_rate": 2.5537634408602152e-05,
585
+ "loss": 0.1312,
586
+ "step": 91
587
+ },
588
+ {
589
+ "epoch": 1.48,
590
+ "learning_rate": 2.5268817204301075e-05,
591
+ "loss": 0.1228,
592
+ "step": 92
593
+ },
594
+ {
595
+ "epoch": 1.5,
596
+ "learning_rate": 2.5e-05,
597
+ "loss": 0.1283,
598
+ "step": 93
599
+ },
600
+ {
601
+ "epoch": 1.51,
602
+ "learning_rate": 2.4731182795698928e-05,
603
+ "loss": 0.1271,
604
+ "step": 94
605
+ },
606
+ {
607
+ "epoch": 1.53,
608
+ "learning_rate": 2.446236559139785e-05,
609
+ "loss": 0.126,
610
+ "step": 95
611
+ },
612
+ {
613
+ "epoch": 1.54,
614
+ "learning_rate": 2.4193548387096777e-05,
615
+ "loss": 0.1232,
616
+ "step": 96
617
+ },
618
+ {
619
+ "epoch": 1.56,
620
+ "learning_rate": 2.39247311827957e-05,
621
+ "loss": 0.1259,
622
+ "step": 97
623
+ },
624
+ {
625
+ "epoch": 1.58,
626
+ "learning_rate": 2.3655913978494626e-05,
627
+ "loss": 0.1262,
628
+ "step": 98
629
+ },
630
+ {
631
+ "epoch": 1.59,
632
+ "learning_rate": 2.338709677419355e-05,
633
+ "loss": 0.1253,
634
+ "step": 99
635
+ },
636
+ {
637
+ "epoch": 1.61,
638
+ "learning_rate": 2.3118279569892472e-05,
639
+ "loss": 0.1304,
640
+ "step": 100
641
+ },
642
+ {
643
+ "epoch": 1.61,
644
+ "eval_loss": 0.1252489686012268,
645
+ "eval_runtime": 772.5528,
646
+ "eval_samples_per_second": 459.653,
647
+ "eval_steps_per_second": 57.458,
648
+ "step": 100
649
+ },
650
+ {
651
+ "epoch": 1.62,
652
+ "learning_rate": 2.28494623655914e-05,
653
+ "loss": 0.1267,
654
+ "step": 101
655
+ },
656
+ {
657
+ "epoch": 1.64,
658
+ "learning_rate": 2.258064516129032e-05,
659
+ "loss": 0.1254,
660
+ "step": 102
661
+ },
662
+ {
663
+ "epoch": 1.66,
664
+ "learning_rate": 2.2311827956989248e-05,
665
+ "loss": 0.1265,
666
+ "step": 103
667
+ },
668
+ {
669
+ "epoch": 1.67,
670
+ "learning_rate": 2.2043010752688174e-05,
671
+ "loss": 0.1267,
672
+ "step": 104
673
+ },
674
+ {
675
+ "epoch": 1.69,
676
+ "learning_rate": 2.1774193548387097e-05,
677
+ "loss": 0.1301,
678
+ "step": 105
679
+ },
680
+ {
681
+ "epoch": 1.7,
682
+ "learning_rate": 2.1505376344086024e-05,
683
+ "loss": 0.1229,
684
+ "step": 106
685
+ },
686
+ {
687
+ "epoch": 1.72,
688
+ "learning_rate": 2.1236559139784946e-05,
689
+ "loss": 0.1254,
690
+ "step": 107
691
+ },
692
+ {
693
+ "epoch": 1.74,
694
+ "learning_rate": 2.0967741935483873e-05,
695
+ "loss": 0.125,
696
+ "step": 108
697
+ },
698
+ {
699
+ "epoch": 1.75,
700
+ "learning_rate": 2.06989247311828e-05,
701
+ "loss": 0.1217,
702
+ "step": 109
703
+ },
704
+ {
705
+ "epoch": 1.77,
706
+ "learning_rate": 2.0430107526881722e-05,
707
+ "loss": 0.1295,
708
+ "step": 110
709
+ },
710
+ {
711
+ "epoch": 1.78,
712
+ "learning_rate": 2.0161290322580645e-05,
713
+ "loss": 0.1268,
714
+ "step": 111
715
+ },
716
+ {
717
+ "epoch": 1.8,
718
+ "learning_rate": 1.989247311827957e-05,
719
+ "loss": 0.1277,
720
+ "step": 112
721
+ },
722
+ {
723
+ "epoch": 1.82,
724
+ "learning_rate": 1.9623655913978494e-05,
725
+ "loss": 0.1265,
726
+ "step": 113
727
+ },
728
+ {
729
+ "epoch": 1.83,
730
+ "learning_rate": 1.935483870967742e-05,
731
+ "loss": 0.1244,
732
+ "step": 114
733
+ },
734
+ {
735
+ "epoch": 1.85,
736
+ "learning_rate": 1.9086021505376344e-05,
737
+ "loss": 0.1275,
738
+ "step": 115
739
+ },
740
+ {
741
+ "epoch": 1.86,
742
+ "learning_rate": 1.881720430107527e-05,
743
+ "loss": 0.1267,
744
+ "step": 116
745
+ },
746
+ {
747
+ "epoch": 1.88,
748
+ "learning_rate": 1.8548387096774193e-05,
749
+ "loss": 0.1224,
750
+ "step": 117
751
+ },
752
+ {
753
+ "epoch": 1.9,
754
+ "learning_rate": 1.827956989247312e-05,
755
+ "loss": 0.1268,
756
+ "step": 118
757
+ },
758
+ {
759
+ "epoch": 1.91,
760
+ "learning_rate": 1.8010752688172046e-05,
761
+ "loss": 0.1238,
762
+ "step": 119
763
+ },
764
+ {
765
+ "epoch": 1.93,
766
+ "learning_rate": 1.774193548387097e-05,
767
+ "loss": 0.1235,
768
+ "step": 120
769
+ },
770
+ {
771
+ "epoch": 1.93,
772
+ "eval_loss": 0.12599562108516693,
773
+ "eval_runtime": 771.9019,
774
+ "eval_samples_per_second": 460.04,
775
+ "eval_steps_per_second": 57.506,
776
+ "step": 120
777
+ },
778
+ {
779
+ "epoch": 1.94,
780
+ "learning_rate": 1.7473118279569895e-05,
781
+ "loss": 0.1225,
782
+ "step": 121
783
+ },
784
+ {
785
+ "epoch": 1.96,
786
+ "learning_rate": 1.7204301075268818e-05,
787
+ "loss": 0.1245,
788
+ "step": 122
789
+ },
790
+ {
791
+ "epoch": 1.98,
792
+ "learning_rate": 1.693548387096774e-05,
793
+ "loss": 0.1284,
794
+ "step": 123
795
+ },
796
+ {
797
+ "epoch": 1.99,
798
+ "learning_rate": 1.6666666666666667e-05,
799
+ "loss": 0.1251,
800
+ "step": 124
801
+ },
802
+ {
803
+ "epoch": 2.02,
804
+ "learning_rate": 1.639784946236559e-05,
805
+ "loss": 0.1843,
806
+ "step": 125
807
+ },
808
+ {
809
+ "epoch": 2.03,
810
+ "learning_rate": 1.6129032258064517e-05,
811
+ "loss": 0.1263,
812
+ "step": 126
813
+ },
814
+ {
815
+ "epoch": 2.05,
816
+ "learning_rate": 1.586021505376344e-05,
817
+ "loss": 0.1284,
818
+ "step": 127
819
+ },
820
+ {
821
+ "epoch": 2.06,
822
+ "learning_rate": 1.5591397849462366e-05,
823
+ "loss": 0.1277,
824
+ "step": 128
825
+ },
826
+ {
827
+ "epoch": 2.08,
828
+ "learning_rate": 1.5322580645161292e-05,
829
+ "loss": 0.1223,
830
+ "step": 129
831
+ },
832
+ {
833
+ "epoch": 2.1,
834
+ "learning_rate": 1.5053763440860215e-05,
835
+ "loss": 0.1241,
836
+ "step": 130
837
+ },
838
+ {
839
+ "epoch": 2.11,
840
+ "learning_rate": 1.4784946236559142e-05,
841
+ "loss": 0.1233,
842
+ "step": 131
843
+ },
844
+ {
845
+ "epoch": 2.13,
846
+ "learning_rate": 1.4516129032258066e-05,
847
+ "loss": 0.1222,
848
+ "step": 132
849
+ },
850
+ {
851
+ "epoch": 2.14,
852
+ "learning_rate": 1.4247311827956991e-05,
853
+ "loss": 0.1195,
854
+ "step": 133
855
+ },
856
+ {
857
+ "epoch": 2.16,
858
+ "learning_rate": 1.3978494623655914e-05,
859
+ "loss": 0.1267,
860
+ "step": 134
861
+ },
862
+ {
863
+ "epoch": 2.18,
864
+ "learning_rate": 1.3709677419354839e-05,
865
+ "loss": 0.124,
866
+ "step": 135
867
+ },
868
+ {
869
+ "epoch": 2.19,
870
+ "learning_rate": 1.3440860215053763e-05,
871
+ "loss": 0.125,
872
+ "step": 136
873
+ },
874
+ {
875
+ "epoch": 2.21,
876
+ "learning_rate": 1.3172043010752688e-05,
877
+ "loss": 0.1242,
878
+ "step": 137
879
+ },
880
+ {
881
+ "epoch": 2.22,
882
+ "learning_rate": 1.2903225806451613e-05,
883
+ "loss": 0.1231,
884
+ "step": 138
885
+ },
886
+ {
887
+ "epoch": 2.24,
888
+ "learning_rate": 1.2634408602150537e-05,
889
+ "loss": 0.1208,
890
+ "step": 139
891
+ },
892
+ {
893
+ "epoch": 2.26,
894
+ "learning_rate": 1.2365591397849464e-05,
895
+ "loss": 0.125,
896
+ "step": 140
897
+ },
898
+ {
899
+ "epoch": 2.26,
900
+ "eval_loss": 0.12610550224781036,
901
+ "eval_runtime": 771.1365,
902
+ "eval_samples_per_second": 460.497,
903
+ "eval_steps_per_second": 57.563,
904
+ "step": 140
905
+ },
906
+ {
907
+ "epoch": 2.27,
908
+ "learning_rate": 1.2096774193548388e-05,
909
+ "loss": 0.1235,
910
+ "step": 141
911
+ },
912
+ {
913
+ "epoch": 2.29,
914
+ "learning_rate": 1.1827956989247313e-05,
915
+ "loss": 0.1267,
916
+ "step": 142
917
+ },
918
+ {
919
+ "epoch": 2.3,
920
+ "learning_rate": 1.1559139784946236e-05,
921
+ "loss": 0.1244,
922
+ "step": 143
923
+ },
924
+ {
925
+ "epoch": 2.32,
926
+ "learning_rate": 1.129032258064516e-05,
927
+ "loss": 0.1215,
928
+ "step": 144
929
+ },
930
+ {
931
+ "epoch": 2.34,
932
+ "learning_rate": 1.1021505376344087e-05,
933
+ "loss": 0.1261,
934
+ "step": 145
935
+ },
936
+ {
937
+ "epoch": 2.35,
938
+ "learning_rate": 1.0752688172043012e-05,
939
+ "loss": 0.1226,
940
+ "step": 146
941
+ },
942
+ {
943
+ "epoch": 2.37,
944
+ "learning_rate": 1.0483870967741936e-05,
945
+ "loss": 0.1277,
946
+ "step": 147
947
+ },
948
+ {
949
+ "epoch": 2.38,
950
+ "learning_rate": 1.0215053763440861e-05,
951
+ "loss": 0.1262,
952
+ "step": 148
953
+ },
954
+ {
955
+ "epoch": 2.4,
956
+ "learning_rate": 9.946236559139786e-06,
957
+ "loss": 0.1248,
958
+ "step": 149
959
+ },
960
+ {
961
+ "epoch": 2.42,
962
+ "learning_rate": 9.67741935483871e-06,
963
+ "loss": 0.1293,
964
+ "step": 150
965
+ },
966
+ {
967
+ "epoch": 2.43,
968
+ "learning_rate": 9.408602150537635e-06,
969
+ "loss": 0.1266,
970
+ "step": 151
971
+ },
972
+ {
973
+ "epoch": 2.45,
974
+ "learning_rate": 9.13978494623656e-06,
975
+ "loss": 0.1276,
976
+ "step": 152
977
+ },
978
+ {
979
+ "epoch": 2.46,
980
+ "learning_rate": 8.870967741935484e-06,
981
+ "loss": 0.1246,
982
+ "step": 153
983
+ },
984
+ {
985
+ "epoch": 2.48,
986
+ "learning_rate": 8.602150537634409e-06,
987
+ "loss": 0.1263,
988
+ "step": 154
989
+ },
990
+ {
991
+ "epoch": 2.5,
992
+ "learning_rate": 8.333333333333334e-06,
993
+ "loss": 0.1251,
994
+ "step": 155
995
+ },
996
+ {
997
+ "epoch": 2.51,
998
+ "learning_rate": 8.064516129032258e-06,
999
+ "loss": 0.1263,
1000
+ "step": 156
1001
+ },
1002
+ {
1003
+ "epoch": 2.53,
1004
+ "learning_rate": 7.795698924731183e-06,
1005
+ "loss": 0.1285,
1006
+ "step": 157
1007
+ },
1008
+ {
1009
+ "epoch": 2.54,
1010
+ "learning_rate": 7.526881720430108e-06,
1011
+ "loss": 0.1238,
1012
+ "step": 158
1013
+ },
1014
+ {
1015
+ "epoch": 2.56,
1016
+ "learning_rate": 7.258064516129033e-06,
1017
+ "loss": 0.1261,
1018
+ "step": 159
1019
+ },
1020
+ {
1021
+ "epoch": 2.58,
1022
+ "learning_rate": 6.989247311827957e-06,
1023
+ "loss": 0.1263,
1024
+ "step": 160
1025
+ },
1026
+ {
1027
+ "epoch": 2.58,
1028
+ "eval_loss": 0.12619347870349884,
1029
+ "eval_runtime": 771.1616,
1030
+ "eval_samples_per_second": 460.482,
1031
+ "eval_steps_per_second": 57.561,
1032
+ "step": 160
1033
+ },
1034
+ {
1035
+ "epoch": 2.59,
1036
+ "learning_rate": 6.720430107526882e-06,
1037
+ "loss": 0.1228,
1038
+ "step": 161
1039
+ },
1040
+ {
1041
+ "epoch": 2.61,
1042
+ "learning_rate": 6.451612903225806e-06,
1043
+ "loss": 0.1245,
1044
+ "step": 162
1045
+ },
1046
+ {
1047
+ "epoch": 2.62,
1048
+ "learning_rate": 6.182795698924732e-06,
1049
+ "loss": 0.1229,
1050
+ "step": 163
1051
+ },
1052
+ {
1053
+ "epoch": 2.64,
1054
+ "learning_rate": 5.9139784946236566e-06,
1055
+ "loss": 0.1216,
1056
+ "step": 164
1057
+ },
1058
+ {
1059
+ "epoch": 2.66,
1060
+ "learning_rate": 5.64516129032258e-06,
1061
+ "loss": 0.1268,
1062
+ "step": 165
1063
+ },
1064
+ {
1065
+ "epoch": 2.67,
1066
+ "learning_rate": 5.376344086021506e-06,
1067
+ "loss": 0.1237,
1068
+ "step": 166
1069
+ },
1070
+ {
1071
+ "epoch": 2.69,
1072
+ "learning_rate": 5.1075268817204305e-06,
1073
+ "loss": 0.1251,
1074
+ "step": 167
1075
+ },
1076
+ {
1077
+ "epoch": 2.7,
1078
+ "learning_rate": 4.838709677419355e-06,
1079
+ "loss": 0.1241,
1080
+ "step": 168
1081
+ },
1082
+ {
1083
+ "epoch": 2.72,
1084
+ "learning_rate": 4.56989247311828e-06,
1085
+ "loss": 0.1247,
1086
+ "step": 169
1087
+ },
1088
+ {
1089
+ "epoch": 2.74,
1090
+ "learning_rate": 4.3010752688172045e-06,
1091
+ "loss": 0.1241,
1092
+ "step": 170
1093
+ },
1094
+ {
1095
+ "epoch": 2.75,
1096
+ "learning_rate": 4.032258064516129e-06,
1097
+ "loss": 0.1242,
1098
+ "step": 171
1099
+ },
1100
+ {
1101
+ "epoch": 2.77,
1102
+ "learning_rate": 3.763440860215054e-06,
1103
+ "loss": 0.1265,
1104
+ "step": 172
1105
+ },
1106
+ {
1107
+ "epoch": 2.78,
1108
+ "learning_rate": 3.4946236559139785e-06,
1109
+ "loss": 0.1209,
1110
+ "step": 173
1111
+ },
1112
+ {
1113
+ "epoch": 2.8,
1114
+ "learning_rate": 3.225806451612903e-06,
1115
+ "loss": 0.1234,
1116
+ "step": 174
1117
+ },
1118
+ {
1119
+ "epoch": 2.82,
1120
+ "learning_rate": 2.9569892473118283e-06,
1121
+ "loss": 0.1231,
1122
+ "step": 175
1123
+ },
1124
+ {
1125
+ "epoch": 2.83,
1126
+ "learning_rate": 2.688172043010753e-06,
1127
+ "loss": 0.1274,
1128
+ "step": 176
1129
+ },
1130
+ {
1131
+ "epoch": 2.85,
1132
+ "learning_rate": 2.4193548387096776e-06,
1133
+ "loss": 0.1247,
1134
+ "step": 177
1135
+ },
1136
+ {
1137
+ "epoch": 2.86,
1138
+ "learning_rate": 2.1505376344086023e-06,
1139
+ "loss": 0.1251,
1140
+ "step": 178
1141
+ },
1142
+ {
1143
+ "epoch": 2.88,
1144
+ "learning_rate": 1.881720430107527e-06,
1145
+ "loss": 0.1229,
1146
+ "step": 179
1147
+ },
1148
+ {
1149
+ "epoch": 2.9,
1150
+ "learning_rate": 1.6129032258064516e-06,
1151
+ "loss": 0.1244,
1152
+ "step": 180
1153
+ },
1154
+ {
1155
+ "epoch": 2.9,
1156
+ "eval_loss": 0.1256006360054016,
1157
+ "eval_runtime": 771.5064,
1158
+ "eval_samples_per_second": 460.276,
1159
+ "eval_steps_per_second": 57.535,
1160
+ "step": 180
1161
+ },
1162
+ {
1163
+ "epoch": 2.91,
1164
+ "learning_rate": 1.3440860215053765e-06,
1165
+ "loss": 0.1266,
1166
+ "step": 181
1167
+ },
1168
+ {
1169
+ "epoch": 2.93,
1170
+ "learning_rate": 1.0752688172043011e-06,
1171
+ "loss": 0.1213,
1172
+ "step": 182
1173
+ },
1174
+ {
1175
+ "epoch": 2.94,
1176
+ "learning_rate": 8.064516129032258e-07,
1177
+ "loss": 0.1245,
1178
+ "step": 183
1179
+ },
1180
+ {
1181
+ "epoch": 2.96,
1182
+ "learning_rate": 5.376344086021506e-07,
1183
+ "loss": 0.1244,
1184
+ "step": 184
1185
+ },
1186
+ {
1187
+ "epoch": 2.98,
1188
+ "learning_rate": 2.688172043010753e-07,
1189
+ "loss": 0.1278,
1190
+ "step": 185
1191
+ },
1192
+ {
1193
+ "epoch": 2.99,
1194
+ "learning_rate": 0.0,
1195
+ "loss": 0.1247,
1196
+ "step": 186
1197
+ },
1198
+ {
1199
+ "epoch": 2.99,
1200
+ "step": 186,
1201
+ "total_flos": 3.806816568043341e+17,
1202
+ "train_loss": 0.1504179045798317,
1203
+ "train_runtime": 22764.4166,
1204
+ "train_samples_per_second": 65.892,
1205
+ "train_steps_per_second": 0.008
1206
+ }
1207
+ ],
1208
+ "max_steps": 186,
1209
+ "num_train_epochs": 3,
1210
+ "total_flos": 3.806816568043341e+17,
1211
+ "trial_name": null,
1212
+ "trial_params": null
1213
+ }