Piro17 commited on
Commit
1fa3dcd
1 Parent(s): 26d6745

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 13.0,
3
+ "eval_accuracy": 0.6224824355971897,
4
+ "eval_f1": 0.6166583378418276,
5
+ "eval_loss": 1.0525678396224976,
6
+ "eval_precision": 0.6161446612041342,
7
+ "eval_recall": 0.6224824355971897,
8
+ "eval_runtime": 66.6004,
9
+ "eval_samples_per_second": 32.057,
10
+ "eval_steps_per_second": 1.006,
11
+ "total_flos": 1.7197173028930535e+19,
12
+ "train_loss": 1.109088444861461,
13
+ "train_runtime": 12754.1331,
14
+ "train_samples_per_second": 17.403,
15
+ "train_steps_per_second": 0.136
16
+ }
eval_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 13.0,
3
+ "eval_accuracy": 0.6224824355971897,
4
+ "eval_f1": 0.6166583378418276,
5
+ "eval_loss": 1.0525678396224976,
6
+ "eval_precision": 0.6161446612041342,
7
+ "eval_recall": 0.6224824355971897,
8
+ "eval_runtime": 66.6004,
9
+ "eval_samples_per_second": 32.057,
10
+ "eval_steps_per_second": 1.006
11
+ }
runs/Feb15_10-39-54_2d5c245a2246/events.out.tfevents.1676470510.2d5c245a2246.108.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a53c3b756ddabe6ad951239aa7e8455040be709999c299b776f0056eeb4c9be9
3
+ size 512
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 13.0,
3
+ "total_flos": 1.7197173028930535e+19,
4
+ "train_loss": 1.109088444861461,
5
+ "train_runtime": 12754.1331,
6
+ "train_samples_per_second": 17.403,
7
+ "train_steps_per_second": 0.136
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1213 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6166583378418276,
3
+ "best_model_checkpoint": "13E-affecthq-fer-balanced/checkpoint-1729",
4
+ "epoch": 12.99625468164794,
5
+ "global_step": 1729,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.07,
12
+ "learning_rate": 5.780346820809249e-07,
13
+ "loss": 1.9634,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.15,
18
+ "learning_rate": 1.1560693641618499e-06,
19
+ "loss": 1.9552,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.22,
24
+ "learning_rate": 1.7341040462427746e-06,
25
+ "loss": 1.96,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.3,
30
+ "learning_rate": 2.3121387283236997e-06,
31
+ "loss": 1.9515,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.37,
36
+ "learning_rate": 2.890173410404625e-06,
37
+ "loss": 1.9503,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.45,
42
+ "learning_rate": 3.468208092485549e-06,
43
+ "loss": 1.9404,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.52,
48
+ "learning_rate": 4.046242774566474e-06,
49
+ "loss": 1.9321,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.6,
54
+ "learning_rate": 4.6242774566473994e-06,
55
+ "loss": 1.92,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.67,
60
+ "learning_rate": 5.202312138728324e-06,
61
+ "loss": 1.9068,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.75,
66
+ "learning_rate": 5.78034682080925e-06,
67
+ "loss": 1.8931,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.82,
72
+ "learning_rate": 6.358381502890174e-06,
73
+ "loss": 1.8641,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.9,
78
+ "learning_rate": 6.936416184971098e-06,
79
+ "loss": 1.8289,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.97,
84
+ "learning_rate": 7.5144508670520235e-06,
85
+ "loss": 1.7863,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 1.0,
90
+ "eval_accuracy": 0.40046838407494145,
91
+ "eval_f1": 0.3058443007386033,
92
+ "eval_loss": 1.7632062435150146,
93
+ "eval_precision": 0.36173402622129774,
94
+ "eval_recall": 0.40046838407494145,
95
+ "eval_runtime": 67.8192,
96
+ "eval_samples_per_second": 31.481,
97
+ "eval_steps_per_second": 0.988,
98
+ "step": 133
99
+ },
100
+ {
101
+ "epoch": 1.05,
102
+ "learning_rate": 8.092485549132949e-06,
103
+ "loss": 1.8586,
104
+ "step": 140
105
+ },
106
+ {
107
+ "epoch": 1.13,
108
+ "learning_rate": 8.670520231213873e-06,
109
+ "loss": 1.7136,
110
+ "step": 150
111
+ },
112
+ {
113
+ "epoch": 1.2,
114
+ "learning_rate": 9.248554913294799e-06,
115
+ "loss": 1.6728,
116
+ "step": 160
117
+ },
118
+ {
119
+ "epoch": 1.28,
120
+ "learning_rate": 9.826589595375723e-06,
121
+ "loss": 1.6475,
122
+ "step": 170
123
+ },
124
+ {
125
+ "epoch": 1.35,
126
+ "learning_rate": 9.955012853470438e-06,
127
+ "loss": 1.5912,
128
+ "step": 180
129
+ },
130
+ {
131
+ "epoch": 1.43,
132
+ "learning_rate": 9.890745501285348e-06,
133
+ "loss": 1.545,
134
+ "step": 190
135
+ },
136
+ {
137
+ "epoch": 1.5,
138
+ "learning_rate": 9.826478149100258e-06,
139
+ "loss": 1.5187,
140
+ "step": 200
141
+ },
142
+ {
143
+ "epoch": 1.58,
144
+ "learning_rate": 9.762210796915168e-06,
145
+ "loss": 1.5123,
146
+ "step": 210
147
+ },
148
+ {
149
+ "epoch": 1.65,
150
+ "learning_rate": 9.697943444730078e-06,
151
+ "loss": 1.4574,
152
+ "step": 220
153
+ },
154
+ {
155
+ "epoch": 1.73,
156
+ "learning_rate": 9.633676092544988e-06,
157
+ "loss": 1.4483,
158
+ "step": 230
159
+ },
160
+ {
161
+ "epoch": 1.8,
162
+ "learning_rate": 9.569408740359899e-06,
163
+ "loss": 1.4114,
164
+ "step": 240
165
+ },
166
+ {
167
+ "epoch": 1.88,
168
+ "learning_rate": 9.505141388174809e-06,
169
+ "loss": 1.3853,
170
+ "step": 250
171
+ },
172
+ {
173
+ "epoch": 1.95,
174
+ "learning_rate": 9.440874035989717e-06,
175
+ "loss": 1.3653,
176
+ "step": 260
177
+ },
178
+ {
179
+ "epoch": 2.0,
180
+ "eval_accuracy": 0.5049180327868853,
181
+ "eval_f1": 0.44452526643283874,
182
+ "eval_loss": 1.3629579544067383,
183
+ "eval_precision": 0.48383855087708,
184
+ "eval_recall": 0.5049180327868853,
185
+ "eval_runtime": 67.0004,
186
+ "eval_samples_per_second": 31.865,
187
+ "eval_steps_per_second": 1.0,
188
+ "step": 266
189
+ },
190
+ {
191
+ "epoch": 2.03,
192
+ "learning_rate": 9.376606683804627e-06,
193
+ "loss": 1.44,
194
+ "step": 270
195
+ },
196
+ {
197
+ "epoch": 2.1,
198
+ "learning_rate": 9.312339331619537e-06,
199
+ "loss": 1.3714,
200
+ "step": 280
201
+ },
202
+ {
203
+ "epoch": 2.18,
204
+ "learning_rate": 9.248071979434447e-06,
205
+ "loss": 1.3243,
206
+ "step": 290
207
+ },
208
+ {
209
+ "epoch": 2.25,
210
+ "learning_rate": 9.183804627249359e-06,
211
+ "loss": 1.3214,
212
+ "step": 300
213
+ },
214
+ {
215
+ "epoch": 2.33,
216
+ "learning_rate": 9.119537275064269e-06,
217
+ "loss": 1.2917,
218
+ "step": 310
219
+ },
220
+ {
221
+ "epoch": 2.4,
222
+ "learning_rate": 9.055269922879179e-06,
223
+ "loss": 1.3092,
224
+ "step": 320
225
+ },
226
+ {
227
+ "epoch": 2.48,
228
+ "learning_rate": 8.991002570694089e-06,
229
+ "loss": 1.2699,
230
+ "step": 330
231
+ },
232
+ {
233
+ "epoch": 2.55,
234
+ "learning_rate": 8.926735218508999e-06,
235
+ "loss": 1.2668,
236
+ "step": 340
237
+ },
238
+ {
239
+ "epoch": 2.63,
240
+ "learning_rate": 8.862467866323907e-06,
241
+ "loss": 1.2747,
242
+ "step": 350
243
+ },
244
+ {
245
+ "epoch": 2.7,
246
+ "learning_rate": 8.798200514138818e-06,
247
+ "loss": 1.2415,
248
+ "step": 360
249
+ },
250
+ {
251
+ "epoch": 2.78,
252
+ "learning_rate": 8.733933161953728e-06,
253
+ "loss": 1.2671,
254
+ "step": 370
255
+ },
256
+ {
257
+ "epoch": 2.85,
258
+ "learning_rate": 8.669665809768638e-06,
259
+ "loss": 1.2717,
260
+ "step": 380
261
+ },
262
+ {
263
+ "epoch": 2.93,
264
+ "learning_rate": 8.605398457583548e-06,
265
+ "loss": 1.2468,
266
+ "step": 390
267
+ },
268
+ {
269
+ "epoch": 3.0,
270
+ "eval_accuracy": 0.5466042154566745,
271
+ "eval_f1": 0.5114853306120848,
272
+ "eval_loss": 1.2475332021713257,
273
+ "eval_precision": 0.5451199399022204,
274
+ "eval_recall": 0.5466042154566745,
275
+ "eval_runtime": 66.2985,
276
+ "eval_samples_per_second": 32.203,
277
+ "eval_steps_per_second": 1.011,
278
+ "step": 399
279
+ },
280
+ {
281
+ "epoch": 3.01,
282
+ "learning_rate": 8.541131105398458e-06,
283
+ "loss": 1.2806,
284
+ "step": 400
285
+ },
286
+ {
287
+ "epoch": 3.08,
288
+ "learning_rate": 8.476863753213368e-06,
289
+ "loss": 1.2417,
290
+ "step": 410
291
+ },
292
+ {
293
+ "epoch": 3.16,
294
+ "learning_rate": 8.412596401028278e-06,
295
+ "loss": 1.2221,
296
+ "step": 420
297
+ },
298
+ {
299
+ "epoch": 3.23,
300
+ "learning_rate": 8.348329048843188e-06,
301
+ "loss": 1.1889,
302
+ "step": 430
303
+ },
304
+ {
305
+ "epoch": 3.31,
306
+ "learning_rate": 8.2840616966581e-06,
307
+ "loss": 1.1994,
308
+ "step": 440
309
+ },
310
+ {
311
+ "epoch": 3.38,
312
+ "learning_rate": 8.219794344473008e-06,
313
+ "loss": 1.196,
314
+ "step": 450
315
+ },
316
+ {
317
+ "epoch": 3.46,
318
+ "learning_rate": 8.155526992287918e-06,
319
+ "loss": 1.1762,
320
+ "step": 460
321
+ },
322
+ {
323
+ "epoch": 3.53,
324
+ "learning_rate": 8.091259640102828e-06,
325
+ "loss": 1.1714,
326
+ "step": 470
327
+ },
328
+ {
329
+ "epoch": 3.61,
330
+ "learning_rate": 8.026992287917738e-06,
331
+ "loss": 1.1773,
332
+ "step": 480
333
+ },
334
+ {
335
+ "epoch": 3.68,
336
+ "learning_rate": 7.962724935732648e-06,
337
+ "loss": 1.1672,
338
+ "step": 490
339
+ },
340
+ {
341
+ "epoch": 3.76,
342
+ "learning_rate": 7.898457583547558e-06,
343
+ "loss": 1.1577,
344
+ "step": 500
345
+ },
346
+ {
347
+ "epoch": 3.83,
348
+ "learning_rate": 7.834190231362468e-06,
349
+ "loss": 1.1438,
350
+ "step": 510
351
+ },
352
+ {
353
+ "epoch": 3.91,
354
+ "learning_rate": 7.769922879177378e-06,
355
+ "loss": 1.1594,
356
+ "step": 520
357
+ },
358
+ {
359
+ "epoch": 3.98,
360
+ "learning_rate": 7.705655526992289e-06,
361
+ "loss": 1.1527,
362
+ "step": 530
363
+ },
364
+ {
365
+ "epoch": 4.0,
366
+ "eval_accuracy": 0.5761124121779859,
367
+ "eval_f1": 0.5579541032814586,
368
+ "eval_loss": 1.1865291595458984,
369
+ "eval_precision": 0.561218961307198,
370
+ "eval_recall": 0.5761124121779859,
371
+ "eval_runtime": 65.9861,
372
+ "eval_samples_per_second": 32.355,
373
+ "eval_steps_per_second": 1.015,
374
+ "step": 532
375
+ },
376
+ {
377
+ "epoch": 4.06,
378
+ "learning_rate": 7.641388174807199e-06,
379
+ "loss": 1.2137,
380
+ "step": 540
381
+ },
382
+ {
383
+ "epoch": 4.13,
384
+ "learning_rate": 7.577120822622109e-06,
385
+ "loss": 1.1637,
386
+ "step": 550
387
+ },
388
+ {
389
+ "epoch": 4.21,
390
+ "learning_rate": 7.512853470437018e-06,
391
+ "loss": 1.108,
392
+ "step": 560
393
+ },
394
+ {
395
+ "epoch": 4.28,
396
+ "learning_rate": 7.448586118251928e-06,
397
+ "loss": 1.0764,
398
+ "step": 570
399
+ },
400
+ {
401
+ "epoch": 4.36,
402
+ "learning_rate": 7.384318766066839e-06,
403
+ "loss": 1.1141,
404
+ "step": 580
405
+ },
406
+ {
407
+ "epoch": 4.43,
408
+ "learning_rate": 7.320051413881749e-06,
409
+ "loss": 1.1119,
410
+ "step": 590
411
+ },
412
+ {
413
+ "epoch": 4.51,
414
+ "learning_rate": 7.255784061696659e-06,
415
+ "loss": 1.0904,
416
+ "step": 600
417
+ },
418
+ {
419
+ "epoch": 4.58,
420
+ "learning_rate": 7.191516709511569e-06,
421
+ "loss": 1.1055,
422
+ "step": 610
423
+ },
424
+ {
425
+ "epoch": 4.66,
426
+ "learning_rate": 7.127249357326479e-06,
427
+ "loss": 1.1115,
428
+ "step": 620
429
+ },
430
+ {
431
+ "epoch": 4.73,
432
+ "learning_rate": 7.062982005141389e-06,
433
+ "loss": 1.1029,
434
+ "step": 630
435
+ },
436
+ {
437
+ "epoch": 4.81,
438
+ "learning_rate": 6.998714652956299e-06,
439
+ "loss": 1.1017,
440
+ "step": 640
441
+ },
442
+ {
443
+ "epoch": 4.88,
444
+ "learning_rate": 6.934447300771208e-06,
445
+ "loss": 1.0723,
446
+ "step": 650
447
+ },
448
+ {
449
+ "epoch": 4.96,
450
+ "learning_rate": 6.870179948586118e-06,
451
+ "loss": 1.0862,
452
+ "step": 660
453
+ },
454
+ {
455
+ "epoch": 5.0,
456
+ "eval_accuracy": 0.5784543325526932,
457
+ "eval_f1": 0.5658700696765325,
458
+ "eval_loss": 1.1448348760604858,
459
+ "eval_precision": 0.5686897608132185,
460
+ "eval_recall": 0.5784543325526932,
461
+ "eval_runtime": 66.3078,
462
+ "eval_samples_per_second": 32.198,
463
+ "eval_steps_per_second": 1.01,
464
+ "step": 665
465
+ },
466
+ {
467
+ "epoch": 5.04,
468
+ "learning_rate": 6.8059125964010285e-06,
469
+ "loss": 1.1261,
470
+ "step": 670
471
+ },
472
+ {
473
+ "epoch": 5.11,
474
+ "learning_rate": 6.7416452442159385e-06,
475
+ "loss": 1.0695,
476
+ "step": 680
477
+ },
478
+ {
479
+ "epoch": 5.19,
480
+ "learning_rate": 6.677377892030849e-06,
481
+ "loss": 1.0273,
482
+ "step": 690
483
+ },
484
+ {
485
+ "epoch": 5.26,
486
+ "learning_rate": 6.613110539845759e-06,
487
+ "loss": 1.0905,
488
+ "step": 700
489
+ },
490
+ {
491
+ "epoch": 5.34,
492
+ "learning_rate": 6.548843187660669e-06,
493
+ "loss": 1.0512,
494
+ "step": 710
495
+ },
496
+ {
497
+ "epoch": 5.41,
498
+ "learning_rate": 6.48457583547558e-06,
499
+ "loss": 1.0329,
500
+ "step": 720
501
+ },
502
+ {
503
+ "epoch": 5.49,
504
+ "learning_rate": 6.42030848329049e-06,
505
+ "loss": 1.0337,
506
+ "step": 730
507
+ },
508
+ {
509
+ "epoch": 5.56,
510
+ "learning_rate": 6.3560411311054e-06,
511
+ "loss": 1.0676,
512
+ "step": 740
513
+ },
514
+ {
515
+ "epoch": 5.64,
516
+ "learning_rate": 6.291773778920309e-06,
517
+ "loss": 1.0171,
518
+ "step": 750
519
+ },
520
+ {
521
+ "epoch": 5.71,
522
+ "learning_rate": 6.227506426735219e-06,
523
+ "loss": 1.0644,
524
+ "step": 760
525
+ },
526
+ {
527
+ "epoch": 5.79,
528
+ "learning_rate": 6.163239074550129e-06,
529
+ "loss": 1.048,
530
+ "step": 770
531
+ },
532
+ {
533
+ "epoch": 5.86,
534
+ "learning_rate": 6.098971722365039e-06,
535
+ "loss": 1.0129,
536
+ "step": 780
537
+ },
538
+ {
539
+ "epoch": 5.94,
540
+ "learning_rate": 6.034704370179949e-06,
541
+ "loss": 1.064,
542
+ "step": 790
543
+ },
544
+ {
545
+ "epoch": 6.0,
546
+ "eval_accuracy": 0.5971896955503513,
547
+ "eval_f1": 0.5853025369018718,
548
+ "eval_loss": 1.1107723712921143,
549
+ "eval_precision": 0.5866546554640366,
550
+ "eval_recall": 0.5971896955503513,
551
+ "eval_runtime": 66.5889,
552
+ "eval_samples_per_second": 32.062,
553
+ "eval_steps_per_second": 1.006,
554
+ "step": 798
555
+ },
556
+ {
557
+ "epoch": 6.01,
558
+ "learning_rate": 5.970437017994859e-06,
559
+ "loss": 1.0766,
560
+ "step": 800
561
+ },
562
+ {
563
+ "epoch": 6.09,
564
+ "learning_rate": 5.906169665809769e-06,
565
+ "loss": 1.0108,
566
+ "step": 810
567
+ },
568
+ {
569
+ "epoch": 6.16,
570
+ "learning_rate": 5.8419023136246785e-06,
571
+ "loss": 1.0335,
572
+ "step": 820
573
+ },
574
+ {
575
+ "epoch": 6.24,
576
+ "learning_rate": 5.7776349614395885e-06,
577
+ "loss": 1.007,
578
+ "step": 830
579
+ },
580
+ {
581
+ "epoch": 6.31,
582
+ "learning_rate": 5.713367609254499e-06,
583
+ "loss": 1.0071,
584
+ "step": 840
585
+ },
586
+ {
587
+ "epoch": 6.39,
588
+ "learning_rate": 5.649100257069409e-06,
589
+ "loss": 1.0096,
590
+ "step": 850
591
+ },
592
+ {
593
+ "epoch": 6.46,
594
+ "learning_rate": 5.5848329048843196e-06,
595
+ "loss": 0.9766,
596
+ "step": 860
597
+ },
598
+ {
599
+ "epoch": 6.54,
600
+ "learning_rate": 5.52056555269923e-06,
601
+ "loss": 0.9753,
602
+ "step": 870
603
+ },
604
+ {
605
+ "epoch": 6.61,
606
+ "learning_rate": 5.45629820051414e-06,
607
+ "loss": 0.9808,
608
+ "step": 880
609
+ },
610
+ {
611
+ "epoch": 6.69,
612
+ "learning_rate": 5.39203084832905e-06,
613
+ "loss": 1.0024,
614
+ "step": 890
615
+ },
616
+ {
617
+ "epoch": 6.76,
618
+ "learning_rate": 5.32776349614396e-06,
619
+ "loss": 0.9615,
620
+ "step": 900
621
+ },
622
+ {
623
+ "epoch": 6.84,
624
+ "learning_rate": 5.26349614395887e-06,
625
+ "loss": 0.9698,
626
+ "step": 910
627
+ },
628
+ {
629
+ "epoch": 6.91,
630
+ "learning_rate": 5.199228791773779e-06,
631
+ "loss": 1.0006,
632
+ "step": 920
633
+ },
634
+ {
635
+ "epoch": 6.99,
636
+ "learning_rate": 5.134961439588689e-06,
637
+ "loss": 1.0037,
638
+ "step": 930
639
+ },
640
+ {
641
+ "epoch": 7.0,
642
+ "eval_accuracy": 0.6018735362997658,
643
+ "eval_f1": 0.5945820873638747,
644
+ "eval_loss": 1.0968866348266602,
645
+ "eval_precision": 0.5968403329145666,
646
+ "eval_recall": 0.6018735362997658,
647
+ "eval_runtime": 65.51,
648
+ "eval_samples_per_second": 32.59,
649
+ "eval_steps_per_second": 1.023,
650
+ "step": 931
651
+ },
652
+ {
653
+ "epoch": 7.07,
654
+ "learning_rate": 5.070694087403599e-06,
655
+ "loss": 0.9839,
656
+ "step": 940
657
+ },
658
+ {
659
+ "epoch": 7.14,
660
+ "learning_rate": 5.006426735218509e-06,
661
+ "loss": 0.9645,
662
+ "step": 950
663
+ },
664
+ {
665
+ "epoch": 7.22,
666
+ "learning_rate": 4.942159383033419e-06,
667
+ "loss": 0.9692,
668
+ "step": 960
669
+ },
670
+ {
671
+ "epoch": 7.29,
672
+ "learning_rate": 4.877892030848329e-06,
673
+ "loss": 0.9593,
674
+ "step": 970
675
+ },
676
+ {
677
+ "epoch": 7.37,
678
+ "learning_rate": 4.813624678663239e-06,
679
+ "loss": 1.0119,
680
+ "step": 980
681
+ },
682
+ {
683
+ "epoch": 7.44,
684
+ "learning_rate": 4.7493573264781495e-06,
685
+ "loss": 0.9453,
686
+ "step": 990
687
+ },
688
+ {
689
+ "epoch": 7.52,
690
+ "learning_rate": 4.6850899742930595e-06,
691
+ "loss": 0.933,
692
+ "step": 1000
693
+ },
694
+ {
695
+ "epoch": 7.59,
696
+ "learning_rate": 4.62082262210797e-06,
697
+ "loss": 0.9249,
698
+ "step": 1010
699
+ },
700
+ {
701
+ "epoch": 7.67,
702
+ "learning_rate": 4.55655526992288e-06,
703
+ "loss": 0.9538,
704
+ "step": 1020
705
+ },
706
+ {
707
+ "epoch": 7.74,
708
+ "learning_rate": 4.49228791773779e-06,
709
+ "loss": 0.9535,
710
+ "step": 1030
711
+ },
712
+ {
713
+ "epoch": 7.82,
714
+ "learning_rate": 4.4280205655527e-06,
715
+ "loss": 0.9383,
716
+ "step": 1040
717
+ },
718
+ {
719
+ "epoch": 7.89,
720
+ "learning_rate": 4.36375321336761e-06,
721
+ "loss": 0.9449,
722
+ "step": 1050
723
+ },
724
+ {
725
+ "epoch": 7.97,
726
+ "learning_rate": 4.29948586118252e-06,
727
+ "loss": 0.9533,
728
+ "step": 1060
729
+ },
730
+ {
731
+ "epoch": 8.0,
732
+ "eval_accuracy": 0.6126463700234192,
733
+ "eval_f1": 0.6046320973113557,
734
+ "eval_loss": 1.0763905048370361,
735
+ "eval_precision": 0.6033502517484726,
736
+ "eval_recall": 0.6126463700234192,
737
+ "eval_runtime": 66.8256,
738
+ "eval_samples_per_second": 31.949,
739
+ "eval_steps_per_second": 1.003,
740
+ "step": 1064
741
+ },
742
+ {
743
+ "epoch": 8.04,
744
+ "learning_rate": 4.23521850899743e-06,
745
+ "loss": 0.9792,
746
+ "step": 1070
747
+ },
748
+ {
749
+ "epoch": 8.12,
750
+ "learning_rate": 4.170951156812339e-06,
751
+ "loss": 0.9325,
752
+ "step": 1080
753
+ },
754
+ {
755
+ "epoch": 8.19,
756
+ "learning_rate": 4.106683804627249e-06,
757
+ "loss": 0.9394,
758
+ "step": 1090
759
+ },
760
+ {
761
+ "epoch": 8.27,
762
+ "learning_rate": 4.04241645244216e-06,
763
+ "loss": 0.8888,
764
+ "step": 1100
765
+ },
766
+ {
767
+ "epoch": 8.34,
768
+ "learning_rate": 3.97814910025707e-06,
769
+ "loss": 0.9058,
770
+ "step": 1110
771
+ },
772
+ {
773
+ "epoch": 8.42,
774
+ "learning_rate": 3.91388174807198e-06,
775
+ "loss": 0.9316,
776
+ "step": 1120
777
+ },
778
+ {
779
+ "epoch": 8.49,
780
+ "learning_rate": 3.849614395886889e-06,
781
+ "loss": 0.9179,
782
+ "step": 1130
783
+ },
784
+ {
785
+ "epoch": 8.57,
786
+ "learning_rate": 3.7853470437018e-06,
787
+ "loss": 0.885,
788
+ "step": 1140
789
+ },
790
+ {
791
+ "epoch": 8.64,
792
+ "learning_rate": 3.7210796915167095e-06,
793
+ "loss": 0.933,
794
+ "step": 1150
795
+ },
796
+ {
797
+ "epoch": 8.72,
798
+ "learning_rate": 3.6568123393316196e-06,
799
+ "loss": 0.9388,
800
+ "step": 1160
801
+ },
802
+ {
803
+ "epoch": 8.79,
804
+ "learning_rate": 3.59254498714653e-06,
805
+ "loss": 0.9112,
806
+ "step": 1170
807
+ },
808
+ {
809
+ "epoch": 8.87,
810
+ "learning_rate": 3.52827763496144e-06,
811
+ "loss": 0.9207,
812
+ "step": 1180
813
+ },
814
+ {
815
+ "epoch": 8.94,
816
+ "learning_rate": 3.46401028277635e-06,
817
+ "loss": 0.9063,
818
+ "step": 1190
819
+ },
820
+ {
821
+ "epoch": 9.0,
822
+ "eval_accuracy": 0.615456674473068,
823
+ "eval_f1": 0.6046961104183071,
824
+ "eval_loss": 1.071085810661316,
825
+ "eval_precision": 0.6035147338471722,
826
+ "eval_recall": 0.615456674473068,
827
+ "eval_runtime": 66.7376,
828
+ "eval_samples_per_second": 31.991,
829
+ "eval_steps_per_second": 1.004,
830
+ "step": 1197
831
+ },
832
+ {
833
+ "epoch": 9.02,
834
+ "learning_rate": 3.39974293059126e-06,
835
+ "loss": 0.9774,
836
+ "step": 1200
837
+ },
838
+ {
839
+ "epoch": 9.1,
840
+ "learning_rate": 3.33547557840617e-06,
841
+ "loss": 0.8738,
842
+ "step": 1210
843
+ },
844
+ {
845
+ "epoch": 9.17,
846
+ "learning_rate": 3.27120822622108e-06,
847
+ "loss": 0.8628,
848
+ "step": 1220
849
+ },
850
+ {
851
+ "epoch": 9.25,
852
+ "learning_rate": 3.2069408740359896e-06,
853
+ "loss": 0.8989,
854
+ "step": 1230
855
+ },
856
+ {
857
+ "epoch": 9.32,
858
+ "learning_rate": 3.1426735218509e-06,
859
+ "loss": 0.8994,
860
+ "step": 1240
861
+ },
862
+ {
863
+ "epoch": 9.4,
864
+ "learning_rate": 3.07840616966581e-06,
865
+ "loss": 0.8749,
866
+ "step": 1250
867
+ },
868
+ {
869
+ "epoch": 9.47,
870
+ "learning_rate": 3.01413881748072e-06,
871
+ "loss": 0.8741,
872
+ "step": 1260
873
+ },
874
+ {
875
+ "epoch": 9.55,
876
+ "learning_rate": 2.9498714652956302e-06,
877
+ "loss": 0.8934,
878
+ "step": 1270
879
+ },
880
+ {
881
+ "epoch": 9.62,
882
+ "learning_rate": 2.88560411311054e-06,
883
+ "loss": 0.8906,
884
+ "step": 1280
885
+ },
886
+ {
887
+ "epoch": 9.7,
888
+ "learning_rate": 2.82133676092545e-06,
889
+ "loss": 0.9228,
890
+ "step": 1290
891
+ },
892
+ {
893
+ "epoch": 9.77,
894
+ "learning_rate": 2.75706940874036e-06,
895
+ "loss": 0.867,
896
+ "step": 1300
897
+ },
898
+ {
899
+ "epoch": 9.85,
900
+ "learning_rate": 2.6928020565552705e-06,
901
+ "loss": 0.9146,
902
+ "step": 1310
903
+ },
904
+ {
905
+ "epoch": 9.92,
906
+ "learning_rate": 2.6285347043701805e-06,
907
+ "loss": 0.9205,
908
+ "step": 1320
909
+ },
910
+ {
911
+ "epoch": 10.0,
912
+ "learning_rate": 2.56426735218509e-06,
913
+ "loss": 0.8666,
914
+ "step": 1330
915
+ },
916
+ {
917
+ "epoch": 10.0,
918
+ "eval_accuracy": 0.6173302107728337,
919
+ "eval_f1": 0.61075488056197,
920
+ "eval_loss": 1.0589306354522705,
921
+ "eval_precision": 0.6106645901461774,
922
+ "eval_recall": 0.6173302107728337,
923
+ "eval_runtime": 66.4667,
924
+ "eval_samples_per_second": 32.121,
925
+ "eval_steps_per_second": 1.008,
926
+ "step": 1330
927
+ },
928
+ {
929
+ "epoch": 10.07,
930
+ "learning_rate": 2.5e-06,
931
+ "loss": 0.8981,
932
+ "step": 1340
933
+ },
934
+ {
935
+ "epoch": 10.15,
936
+ "learning_rate": 2.4357326478149103e-06,
937
+ "loss": 0.8835,
938
+ "step": 1350
939
+ },
940
+ {
941
+ "epoch": 10.22,
942
+ "learning_rate": 2.3714652956298203e-06,
943
+ "loss": 0.8641,
944
+ "step": 1360
945
+ },
946
+ {
947
+ "epoch": 10.3,
948
+ "learning_rate": 2.3071979434447304e-06,
949
+ "loss": 0.8887,
950
+ "step": 1370
951
+ },
952
+ {
953
+ "epoch": 10.37,
954
+ "learning_rate": 2.2429305912596404e-06,
955
+ "loss": 0.878,
956
+ "step": 1380
957
+ },
958
+ {
959
+ "epoch": 10.45,
960
+ "learning_rate": 2.17866323907455e-06,
961
+ "loss": 0.8524,
962
+ "step": 1390
963
+ },
964
+ {
965
+ "epoch": 10.52,
966
+ "learning_rate": 2.1143958868894605e-06,
967
+ "loss": 0.8656,
968
+ "step": 1400
969
+ },
970
+ {
971
+ "epoch": 10.6,
972
+ "learning_rate": 2.05012853470437e-06,
973
+ "loss": 0.8966,
974
+ "step": 1410
975
+ },
976
+ {
977
+ "epoch": 10.67,
978
+ "learning_rate": 1.9858611825192802e-06,
979
+ "loss": 0.8858,
980
+ "step": 1420
981
+ },
982
+ {
983
+ "epoch": 10.75,
984
+ "learning_rate": 1.9215938303341903e-06,
985
+ "loss": 0.8647,
986
+ "step": 1430
987
+ },
988
+ {
989
+ "epoch": 10.82,
990
+ "learning_rate": 1.8573264781491004e-06,
991
+ "loss": 0.8278,
992
+ "step": 1440
993
+ },
994
+ {
995
+ "epoch": 10.9,
996
+ "learning_rate": 1.7930591259640104e-06,
997
+ "loss": 0.8571,
998
+ "step": 1450
999
+ },
1000
+ {
1001
+ "epoch": 10.97,
1002
+ "learning_rate": 1.7287917737789203e-06,
1003
+ "loss": 0.8364,
1004
+ "step": 1460
1005
+ },
1006
+ {
1007
+ "epoch": 11.0,
1008
+ "eval_accuracy": 0.6177985948477752,
1009
+ "eval_f1": 0.6108344054636982,
1010
+ "eval_loss": 1.0555986166000366,
1011
+ "eval_precision": 0.6110212240412639,
1012
+ "eval_recall": 0.6177985948477752,
1013
+ "eval_runtime": 66.1391,
1014
+ "eval_samples_per_second": 32.28,
1015
+ "eval_steps_per_second": 1.013,
1016
+ "step": 1463
1017
+ },
1018
+ {
1019
+ "epoch": 11.05,
1020
+ "learning_rate": 1.6645244215938305e-06,
1021
+ "loss": 0.8808,
1022
+ "step": 1470
1023
+ },
1024
+ {
1025
+ "epoch": 11.13,
1026
+ "learning_rate": 1.6002570694087406e-06,
1027
+ "loss": 0.8383,
1028
+ "step": 1480
1029
+ },
1030
+ {
1031
+ "epoch": 11.2,
1032
+ "learning_rate": 1.5359897172236504e-06,
1033
+ "loss": 0.863,
1034
+ "step": 1490
1035
+ },
1036
+ {
1037
+ "epoch": 11.28,
1038
+ "learning_rate": 1.4717223650385607e-06,
1039
+ "loss": 0.8312,
1040
+ "step": 1500
1041
+ },
1042
+ {
1043
+ "epoch": 11.35,
1044
+ "learning_rate": 1.4074550128534705e-06,
1045
+ "loss": 0.8933,
1046
+ "step": 1510
1047
+ },
1048
+ {
1049
+ "epoch": 11.43,
1050
+ "learning_rate": 1.3431876606683806e-06,
1051
+ "loss": 0.8563,
1052
+ "step": 1520
1053
+ },
1054
+ {
1055
+ "epoch": 11.5,
1056
+ "learning_rate": 1.2789203084832904e-06,
1057
+ "loss": 0.8588,
1058
+ "step": 1530
1059
+ },
1060
+ {
1061
+ "epoch": 11.58,
1062
+ "learning_rate": 1.2146529562982007e-06,
1063
+ "loss": 0.8611,
1064
+ "step": 1540
1065
+ },
1066
+ {
1067
+ "epoch": 11.65,
1068
+ "learning_rate": 1.1503856041131106e-06,
1069
+ "loss": 0.8356,
1070
+ "step": 1550
1071
+ },
1072
+ {
1073
+ "epoch": 11.73,
1074
+ "learning_rate": 1.0861182519280206e-06,
1075
+ "loss": 0.8371,
1076
+ "step": 1560
1077
+ },
1078
+ {
1079
+ "epoch": 11.8,
1080
+ "learning_rate": 1.0218508997429307e-06,
1081
+ "loss": 0.8177,
1082
+ "step": 1570
1083
+ },
1084
+ {
1085
+ "epoch": 11.88,
1086
+ "learning_rate": 9.575835475578407e-07,
1087
+ "loss": 0.8477,
1088
+ "step": 1580
1089
+ },
1090
+ {
1091
+ "epoch": 11.95,
1092
+ "learning_rate": 8.933161953727507e-07,
1093
+ "loss": 0.8659,
1094
+ "step": 1590
1095
+ },
1096
+ {
1097
+ "epoch": 12.0,
1098
+ "eval_accuracy": 0.6196721311475409,
1099
+ "eval_f1": 0.6150725729241909,
1100
+ "eval_loss": 1.0520738363265991,
1101
+ "eval_precision": 0.6140597586468644,
1102
+ "eval_recall": 0.6196721311475409,
1103
+ "eval_runtime": 66.4481,
1104
+ "eval_samples_per_second": 32.13,
1105
+ "eval_steps_per_second": 1.008,
1106
+ "step": 1596
1107
+ },
1108
+ {
1109
+ "epoch": 12.03,
1110
+ "learning_rate": 8.290488431876607e-07,
1111
+ "loss": 0.9068,
1112
+ "step": 1600
1113
+ },
1114
+ {
1115
+ "epoch": 12.1,
1116
+ "learning_rate": 7.647814910025708e-07,
1117
+ "loss": 0.8314,
1118
+ "step": 1610
1119
+ },
1120
+ {
1121
+ "epoch": 12.18,
1122
+ "learning_rate": 7.005141388174808e-07,
1123
+ "loss": 0.8761,
1124
+ "step": 1620
1125
+ },
1126
+ {
1127
+ "epoch": 12.25,
1128
+ "learning_rate": 6.362467866323908e-07,
1129
+ "loss": 0.8346,
1130
+ "step": 1630
1131
+ },
1132
+ {
1133
+ "epoch": 12.33,
1134
+ "learning_rate": 5.719794344473009e-07,
1135
+ "loss": 0.8345,
1136
+ "step": 1640
1137
+ },
1138
+ {
1139
+ "epoch": 12.4,
1140
+ "learning_rate": 5.077120822622108e-07,
1141
+ "loss": 0.8641,
1142
+ "step": 1650
1143
+ },
1144
+ {
1145
+ "epoch": 12.48,
1146
+ "learning_rate": 4.434447300771209e-07,
1147
+ "loss": 0.8213,
1148
+ "step": 1660
1149
+ },
1150
+ {
1151
+ "epoch": 12.55,
1152
+ "learning_rate": 3.791773778920309e-07,
1153
+ "loss": 0.8584,
1154
+ "step": 1670
1155
+ },
1156
+ {
1157
+ "epoch": 12.63,
1158
+ "learning_rate": 3.149100257069409e-07,
1159
+ "loss": 0.8402,
1160
+ "step": 1680
1161
+ },
1162
+ {
1163
+ "epoch": 12.7,
1164
+ "learning_rate": 2.5064267352185095e-07,
1165
+ "loss": 0.8239,
1166
+ "step": 1690
1167
+ },
1168
+ {
1169
+ "epoch": 12.78,
1170
+ "learning_rate": 1.8637532133676095e-07,
1171
+ "loss": 0.7957,
1172
+ "step": 1700
1173
+ },
1174
+ {
1175
+ "epoch": 12.85,
1176
+ "learning_rate": 1.2210796915167095e-07,
1177
+ "loss": 0.8504,
1178
+ "step": 1710
1179
+ },
1180
+ {
1181
+ "epoch": 12.93,
1182
+ "learning_rate": 5.784061696658098e-08,
1183
+ "loss": 0.8383,
1184
+ "step": 1720
1185
+ },
1186
+ {
1187
+ "epoch": 13.0,
1188
+ "eval_accuracy": 0.6224824355971897,
1189
+ "eval_f1": 0.6166583378418276,
1190
+ "eval_loss": 1.0525678396224976,
1191
+ "eval_precision": 0.6161446612041342,
1192
+ "eval_recall": 0.6224824355971897,
1193
+ "eval_runtime": 65.747,
1194
+ "eval_samples_per_second": 32.473,
1195
+ "eval_steps_per_second": 1.019,
1196
+ "step": 1729
1197
+ },
1198
+ {
1199
+ "epoch": 13.0,
1200
+ "step": 1729,
1201
+ "total_flos": 1.7197173028930535e+19,
1202
+ "train_loss": 1.109088444861461,
1203
+ "train_runtime": 12754.1331,
1204
+ "train_samples_per_second": 17.403,
1205
+ "train_steps_per_second": 0.136
1206
+ }
1207
+ ],
1208
+ "max_steps": 1729,
1209
+ "num_train_epochs": 13,
1210
+ "total_flos": 1.7197173028930535e+19,
1211
+ "trial_name": null,
1212
+ "trial_params": null
1213
+ }