ALM-AHME commited on
Commit
9b3c681
·
1 Parent(s): bf0e037

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "eval_accuracy": 0.9869324473975637,
4
+ "eval_loss": 0.04178118333220482,
5
+ "eval_runtime": 208.5606,
6
+ "eval_samples_per_second": 21.648,
7
+ "eval_steps_per_second": 2.709,
8
+ "total_flos": 6.883415802275955e+18,
9
+ "train_loss": 0.24478418331413984,
10
+ "train_runtime": 10756.8608,
11
+ "train_samples_per_second": 2.782,
12
+ "train_steps_per_second": 0.087
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "eval_accuracy": 0.9869324473975637,
4
+ "eval_loss": 0.04178118333220482,
5
+ "eval_runtime": 208.5606,
6
+ "eval_samples_per_second": 21.648,
7
+ "eval_steps_per_second": 2.709
8
+ }
runs/Jul11_23-28-17_f79ca8bc464c/events.out.tfevents.1689129175.f79ca8bc464c.1684.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bade5bc621f8ce9e1d828370f77f7d514e7c932fe97a0e3387daf7840c3ad9fc
3
+ size 779
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.99,
3
+ "total_flos": 6.883415802275955e+18,
4
+ "train_loss": 0.24478418331413984,
5
+ "train_runtime": 10756.8608,
6
+ "train_samples_per_second": 2.782,
7
+ "train_steps_per_second": 0.087
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1192 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9869324473975637,
3
+ "best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-LungCancer-LC25000-AH-40-30-30/checkpoint-935",
4
+ "epoch": 4.993324432576769,
5
+ "global_step": 935,
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": 5.341880341880342e-06,
13
+ "loss": 1.0848,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.05,
18
+ "learning_rate": 1.0683760683760684e-05,
19
+ "loss": 0.8475,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.08,
24
+ "learning_rate": 1.6025641025641026e-05,
25
+ "loss": 0.5854,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.11,
30
+ "learning_rate": 2.1367521367521368e-05,
31
+ "loss": 0.4202,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.13,
36
+ "learning_rate": 2.670940170940171e-05,
37
+ "loss": 0.2353,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.16,
42
+ "learning_rate": 3.205128205128205e-05,
43
+ "loss": 0.3821,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.19,
48
+ "learning_rate": 3.739316239316239e-05,
49
+ "loss": 0.2081,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.21,
54
+ "learning_rate": 4.2735042735042735e-05,
55
+ "loss": 0.2178,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.24,
60
+ "learning_rate": 4.8076923076923084e-05,
61
+ "loss": 0.2564,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.27,
66
+ "learning_rate": 5.341880341880342e-05,
67
+ "loss": 0.3044,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.29,
72
+ "learning_rate": 5.876068376068376e-05,
73
+ "loss": 0.3078,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.32,
78
+ "learning_rate": 6.41025641025641e-05,
79
+ "loss": 0.2956,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.35,
84
+ "learning_rate": 6.944444444444444e-05,
85
+ "loss": 0.3315,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.37,
90
+ "learning_rate": 7.478632478632479e-05,
91
+ "loss": 0.2415,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.4,
96
+ "learning_rate": 8.012820512820514e-05,
97
+ "loss": 0.2106,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.43,
102
+ "learning_rate": 8.547008547008547e-05,
103
+ "loss": 0.3041,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.45,
108
+ "learning_rate": 9.081196581196581e-05,
109
+ "loss": 0.2804,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.48,
114
+ "learning_rate": 9.615384615384617e-05,
115
+ "loss": 0.2595,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.51,
120
+ "learning_rate": 0.0001014957264957265,
121
+ "loss": 0.2368,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.53,
126
+ "learning_rate": 0.00010683760683760684,
127
+ "loss": 0.2289,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.56,
132
+ "learning_rate": 0.00011217948717948718,
133
+ "loss": 0.2365,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.59,
138
+ "learning_rate": 0.00011752136752136752,
139
+ "loss": 0.1883,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.61,
144
+ "learning_rate": 0.00012286324786324785,
145
+ "loss": 0.1392,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.64,
150
+ "learning_rate": 0.0001282051282051282,
151
+ "loss": 0.1846,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.67,
156
+ "learning_rate": 0.00013354700854700856,
157
+ "loss": 0.1617,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.69,
162
+ "learning_rate": 0.0001388888888888889,
163
+ "loss": 0.175,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.72,
168
+ "learning_rate": 0.00014423076923076922,
169
+ "loss": 0.1804,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 0.75,
174
+ "learning_rate": 0.00014957264957264957,
175
+ "loss": 0.1856,
176
+ "step": 140
177
+ },
178
+ {
179
+ "epoch": 0.77,
180
+ "learning_rate": 0.00015491452991452993,
181
+ "loss": 0.3335,
182
+ "step": 145
183
+ },
184
+ {
185
+ "epoch": 0.8,
186
+ "learning_rate": 0.00016025641025641028,
187
+ "loss": 0.2025,
188
+ "step": 150
189
+ },
190
+ {
191
+ "epoch": 0.83,
192
+ "learning_rate": 0.00016559829059829058,
193
+ "loss": 0.2874,
194
+ "step": 155
195
+ },
196
+ {
197
+ "epoch": 0.85,
198
+ "learning_rate": 0.00017094017094017094,
199
+ "loss": 0.1968,
200
+ "step": 160
201
+ },
202
+ {
203
+ "epoch": 0.88,
204
+ "learning_rate": 0.0001762820512820513,
205
+ "loss": 0.2061,
206
+ "step": 165
207
+ },
208
+ {
209
+ "epoch": 0.91,
210
+ "learning_rate": 0.00018162393162393162,
211
+ "loss": 0.1617,
212
+ "step": 170
213
+ },
214
+ {
215
+ "epoch": 0.93,
216
+ "learning_rate": 0.00018696581196581198,
217
+ "loss": 0.3653,
218
+ "step": 175
219
+ },
220
+ {
221
+ "epoch": 0.96,
222
+ "learning_rate": 0.00019230769230769233,
223
+ "loss": 0.3449,
224
+ "step": 180
225
+ },
226
+ {
227
+ "epoch": 0.99,
228
+ "learning_rate": 0.00019764957264957264,
229
+ "loss": 0.3299,
230
+ "step": 185
231
+ },
232
+ {
233
+ "epoch": 1.0,
234
+ "eval_accuracy": 0.9218161683277962,
235
+ "eval_loss": 0.2117503136396408,
236
+ "eval_runtime": 2740.4947,
237
+ "eval_samples_per_second": 1.648,
238
+ "eval_steps_per_second": 0.206,
239
+ "step": 187
240
+ },
241
+ {
242
+ "epoch": 1.01,
243
+ "learning_rate": 0.000202991452991453,
244
+ "loss": 0.2588,
245
+ "step": 190
246
+ },
247
+ {
248
+ "epoch": 1.04,
249
+ "learning_rate": 0.00020833333333333335,
250
+ "loss": 0.2101,
251
+ "step": 195
252
+ },
253
+ {
254
+ "epoch": 1.07,
255
+ "learning_rate": 0.00021367521367521368,
256
+ "loss": 0.5778,
257
+ "step": 200
258
+ },
259
+ {
260
+ "epoch": 1.09,
261
+ "learning_rate": 0.00021901709401709403,
262
+ "loss": 0.1887,
263
+ "step": 205
264
+ },
265
+ {
266
+ "epoch": 1.12,
267
+ "learning_rate": 0.00022435897435897436,
268
+ "loss": 0.2366,
269
+ "step": 210
270
+ },
271
+ {
272
+ "epoch": 1.15,
273
+ "learning_rate": 0.0002297008547008547,
274
+ "loss": 0.1779,
275
+ "step": 215
276
+ },
277
+ {
278
+ "epoch": 1.17,
279
+ "learning_rate": 0.00023504273504273504,
280
+ "loss": 0.1902,
281
+ "step": 220
282
+ },
283
+ {
284
+ "epoch": 1.2,
285
+ "learning_rate": 0.0002403846153846154,
286
+ "loss": 0.4555,
287
+ "step": 225
288
+ },
289
+ {
290
+ "epoch": 1.23,
291
+ "learning_rate": 0.0002457264957264957,
292
+ "loss": 0.2295,
293
+ "step": 230
294
+ },
295
+ {
296
+ "epoch": 1.26,
297
+ "learning_rate": 0.00025106837606837605,
298
+ "loss": 0.2224,
299
+ "step": 235
300
+ },
301
+ {
302
+ "epoch": 1.28,
303
+ "learning_rate": 0.0002564102564102564,
304
+ "loss": 0.257,
305
+ "step": 240
306
+ },
307
+ {
308
+ "epoch": 1.31,
309
+ "learning_rate": 0.00026175213675213677,
310
+ "loss": 0.395,
311
+ "step": 245
312
+ },
313
+ {
314
+ "epoch": 1.34,
315
+ "learning_rate": 0.0002670940170940171,
316
+ "loss": 0.3104,
317
+ "step": 250
318
+ },
319
+ {
320
+ "epoch": 1.36,
321
+ "learning_rate": 0.0002724358974358974,
322
+ "loss": 0.1329,
323
+ "step": 255
324
+ },
325
+ {
326
+ "epoch": 1.39,
327
+ "learning_rate": 0.0002777777777777778,
328
+ "loss": 0.6173,
329
+ "step": 260
330
+ },
331
+ {
332
+ "epoch": 1.42,
333
+ "learning_rate": 0.00028311965811965813,
334
+ "loss": 0.2237,
335
+ "step": 265
336
+ },
337
+ {
338
+ "epoch": 1.44,
339
+ "learning_rate": 0.00028846153846153843,
340
+ "loss": 0.1947,
341
+ "step": 270
342
+ },
343
+ {
344
+ "epoch": 1.47,
345
+ "learning_rate": 0.00029380341880341884,
346
+ "loss": 0.5807,
347
+ "step": 275
348
+ },
349
+ {
350
+ "epoch": 1.5,
351
+ "learning_rate": 0.00029914529914529915,
352
+ "loss": 0.193,
353
+ "step": 280
354
+ },
355
+ {
356
+ "epoch": 1.52,
357
+ "learning_rate": 0.00030448717948717945,
358
+ "loss": 0.2717,
359
+ "step": 285
360
+ },
361
+ {
362
+ "epoch": 1.55,
363
+ "learning_rate": 0.00030982905982905986,
364
+ "loss": 0.2197,
365
+ "step": 290
366
+ },
367
+ {
368
+ "epoch": 1.58,
369
+ "learning_rate": 0.00031517094017094016,
370
+ "loss": 0.3151,
371
+ "step": 295
372
+ },
373
+ {
374
+ "epoch": 1.6,
375
+ "learning_rate": 0.00032051282051282057,
376
+ "loss": 0.2323,
377
+ "step": 300
378
+ },
379
+ {
380
+ "epoch": 1.63,
381
+ "learning_rate": 0.00032585470085470087,
382
+ "loss": 0.1102,
383
+ "step": 305
384
+ },
385
+ {
386
+ "epoch": 1.66,
387
+ "learning_rate": 0.00033119658119658117,
388
+ "loss": 0.1711,
389
+ "step": 310
390
+ },
391
+ {
392
+ "epoch": 1.68,
393
+ "learning_rate": 0.0003365384615384616,
394
+ "loss": 0.411,
395
+ "step": 315
396
+ },
397
+ {
398
+ "epoch": 1.71,
399
+ "learning_rate": 0.0003418803418803419,
400
+ "loss": 0.2767,
401
+ "step": 320
402
+ },
403
+ {
404
+ "epoch": 1.74,
405
+ "learning_rate": 0.00034722222222222224,
406
+ "loss": 0.3315,
407
+ "step": 325
408
+ },
409
+ {
410
+ "epoch": 1.76,
411
+ "learning_rate": 0.0003525641025641026,
412
+ "loss": 0.368,
413
+ "step": 330
414
+ },
415
+ {
416
+ "epoch": 1.79,
417
+ "learning_rate": 0.0003579059829059829,
418
+ "loss": 0.0977,
419
+ "step": 335
420
+ },
421
+ {
422
+ "epoch": 1.82,
423
+ "learning_rate": 0.00036324786324786325,
424
+ "loss": 0.346,
425
+ "step": 340
426
+ },
427
+ {
428
+ "epoch": 1.84,
429
+ "learning_rate": 0.0003685897435897436,
430
+ "loss": 0.6349,
431
+ "step": 345
432
+ },
433
+ {
434
+ "epoch": 1.87,
435
+ "learning_rate": 0.00037393162393162396,
436
+ "loss": 0.7342,
437
+ "step": 350
438
+ },
439
+ {
440
+ "epoch": 1.9,
441
+ "learning_rate": 0.00037927350427350426,
442
+ "loss": 0.3546,
443
+ "step": 355
444
+ },
445
+ {
446
+ "epoch": 1.92,
447
+ "learning_rate": 0.00038461538461538467,
448
+ "loss": 0.1764,
449
+ "step": 360
450
+ },
451
+ {
452
+ "epoch": 1.95,
453
+ "learning_rate": 0.00038995726495726497,
454
+ "loss": 0.2848,
455
+ "step": 365
456
+ },
457
+ {
458
+ "epoch": 1.98,
459
+ "learning_rate": 0.00039529914529914527,
460
+ "loss": 0.5922,
461
+ "step": 370
462
+ },
463
+ {
464
+ "epoch": 2.0,
465
+ "eval_accuracy": 0.8629014396456257,
466
+ "eval_loss": 0.3205777704715729,
467
+ "eval_runtime": 207.8751,
468
+ "eval_samples_per_second": 21.72,
469
+ "eval_steps_per_second": 2.718,
470
+ "step": 374
471
+ },
472
+ {
473
+ "epoch": 2.0,
474
+ "learning_rate": 0.0004006410256410257,
475
+ "loss": 0.4581,
476
+ "step": 375
477
+ },
478
+ {
479
+ "epoch": 2.03,
480
+ "learning_rate": 0.000405982905982906,
481
+ "loss": 0.3347,
482
+ "step": 380
483
+ },
484
+ {
485
+ "epoch": 2.06,
486
+ "learning_rate": 0.00041132478632478634,
487
+ "loss": 0.2635,
488
+ "step": 385
489
+ },
490
+ {
491
+ "epoch": 2.08,
492
+ "learning_rate": 0.0004166666666666667,
493
+ "loss": 0.2595,
494
+ "step": 390
495
+ },
496
+ {
497
+ "epoch": 2.11,
498
+ "learning_rate": 0.000422008547008547,
499
+ "loss": 0.3457,
500
+ "step": 395
501
+ },
502
+ {
503
+ "epoch": 2.14,
504
+ "learning_rate": 0.00042735042735042735,
505
+ "loss": 0.2144,
506
+ "step": 400
507
+ },
508
+ {
509
+ "epoch": 2.16,
510
+ "learning_rate": 0.0004326923076923077,
511
+ "loss": 0.3235,
512
+ "step": 405
513
+ },
514
+ {
515
+ "epoch": 2.19,
516
+ "learning_rate": 0.00043803418803418806,
517
+ "loss": 0.4917,
518
+ "step": 410
519
+ },
520
+ {
521
+ "epoch": 2.22,
522
+ "learning_rate": 0.00044337606837606836,
523
+ "loss": 0.4038,
524
+ "step": 415
525
+ },
526
+ {
527
+ "epoch": 2.24,
528
+ "learning_rate": 0.0004487179487179487,
529
+ "loss": 0.1941,
530
+ "step": 420
531
+ },
532
+ {
533
+ "epoch": 2.27,
534
+ "learning_rate": 0.0004540598290598291,
535
+ "loss": 0.2165,
536
+ "step": 425
537
+ },
538
+ {
539
+ "epoch": 2.3,
540
+ "learning_rate": 0.0004594017094017094,
541
+ "loss": 0.2612,
542
+ "step": 430
543
+ },
544
+ {
545
+ "epoch": 2.32,
546
+ "learning_rate": 0.0004647435897435898,
547
+ "loss": 0.3141,
548
+ "step": 435
549
+ },
550
+ {
551
+ "epoch": 2.35,
552
+ "learning_rate": 0.0004700854700854701,
553
+ "loss": 0.1971,
554
+ "step": 440
555
+ },
556
+ {
557
+ "epoch": 2.38,
558
+ "learning_rate": 0.00047542735042735044,
559
+ "loss": 0.3959,
560
+ "step": 445
561
+ },
562
+ {
563
+ "epoch": 2.4,
564
+ "learning_rate": 0.0004807692307692308,
565
+ "loss": 0.3421,
566
+ "step": 450
567
+ },
568
+ {
569
+ "epoch": 2.43,
570
+ "learning_rate": 0.0004861111111111111,
571
+ "loss": 0.697,
572
+ "step": 455
573
+ },
574
+ {
575
+ "epoch": 2.46,
576
+ "learning_rate": 0.0004914529914529914,
577
+ "loss": 0.4143,
578
+ "step": 460
579
+ },
580
+ {
581
+ "epoch": 2.48,
582
+ "learning_rate": 0.0004967948717948718,
583
+ "loss": 0.2158,
584
+ "step": 465
585
+ },
586
+ {
587
+ "epoch": 2.51,
588
+ "learning_rate": 0.0004978586723768737,
589
+ "loss": 0.307,
590
+ "step": 470
591
+ },
592
+ {
593
+ "epoch": 2.54,
594
+ "learning_rate": 0.0004925053533190578,
595
+ "loss": 0.3092,
596
+ "step": 475
597
+ },
598
+ {
599
+ "epoch": 2.56,
600
+ "learning_rate": 0.000487152034261242,
601
+ "loss": 0.2478,
602
+ "step": 480
603
+ },
604
+ {
605
+ "epoch": 2.59,
606
+ "learning_rate": 0.00048179871520342613,
607
+ "loss": 0.343,
608
+ "step": 485
609
+ },
610
+ {
611
+ "epoch": 2.62,
612
+ "learning_rate": 0.0004764453961456103,
613
+ "loss": 0.2089,
614
+ "step": 490
615
+ },
616
+ {
617
+ "epoch": 2.64,
618
+ "learning_rate": 0.0004710920770877944,
619
+ "loss": 0.3304,
620
+ "step": 495
621
+ },
622
+ {
623
+ "epoch": 2.67,
624
+ "learning_rate": 0.0004657387580299786,
625
+ "loss": 0.2251,
626
+ "step": 500
627
+ },
628
+ {
629
+ "epoch": 2.7,
630
+ "learning_rate": 0.0004603854389721627,
631
+ "loss": 0.6617,
632
+ "step": 505
633
+ },
634
+ {
635
+ "epoch": 2.72,
636
+ "learning_rate": 0.0004550321199143469,
637
+ "loss": 0.4365,
638
+ "step": 510
639
+ },
640
+ {
641
+ "epoch": 2.75,
642
+ "learning_rate": 0.00044967880085653106,
643
+ "loss": 0.3923,
644
+ "step": 515
645
+ },
646
+ {
647
+ "epoch": 2.78,
648
+ "learning_rate": 0.00044432548179871526,
649
+ "loss": 0.1877,
650
+ "step": 520
651
+ },
652
+ {
653
+ "epoch": 2.8,
654
+ "learning_rate": 0.00043897216274089935,
655
+ "loss": 0.2519,
656
+ "step": 525
657
+ },
658
+ {
659
+ "epoch": 2.83,
660
+ "learning_rate": 0.0004336188436830835,
661
+ "loss": 0.1869,
662
+ "step": 530
663
+ },
664
+ {
665
+ "epoch": 2.86,
666
+ "learning_rate": 0.0004282655246252677,
667
+ "loss": 0.2737,
668
+ "step": 535
669
+ },
670
+ {
671
+ "epoch": 2.88,
672
+ "learning_rate": 0.0004229122055674518,
673
+ "loss": 0.3389,
674
+ "step": 540
675
+ },
676
+ {
677
+ "epoch": 2.91,
678
+ "learning_rate": 0.000417558886509636,
679
+ "loss": 0.1682,
680
+ "step": 545
681
+ },
682
+ {
683
+ "epoch": 2.94,
684
+ "learning_rate": 0.00041220556745182014,
685
+ "loss": 0.1925,
686
+ "step": 550
687
+ },
688
+ {
689
+ "epoch": 2.96,
690
+ "learning_rate": 0.0004068522483940043,
691
+ "loss": 0.0948,
692
+ "step": 555
693
+ },
694
+ {
695
+ "epoch": 2.99,
696
+ "learning_rate": 0.00040149892933618843,
697
+ "loss": 0.1763,
698
+ "step": 560
699
+ },
700
+ {
701
+ "epoch": 3.0,
702
+ "eval_accuracy": 0.9127353266888151,
703
+ "eval_loss": 0.2446804940700531,
704
+ "eval_runtime": 208.6202,
705
+ "eval_samples_per_second": 21.642,
706
+ "eval_steps_per_second": 2.708,
707
+ "step": 561
708
+ },
709
+ {
710
+ "epoch": 3.02,
711
+ "learning_rate": 0.0003961456102783726,
712
+ "loss": 0.2367,
713
+ "step": 565
714
+ },
715
+ {
716
+ "epoch": 3.04,
717
+ "learning_rate": 0.0003907922912205568,
718
+ "loss": 0.2498,
719
+ "step": 570
720
+ },
721
+ {
722
+ "epoch": 3.07,
723
+ "learning_rate": 0.0003854389721627409,
724
+ "loss": 0.2699,
725
+ "step": 575
726
+ },
727
+ {
728
+ "epoch": 3.1,
729
+ "learning_rate": 0.000380085653104925,
730
+ "loss": 0.1427,
731
+ "step": 580
732
+ },
733
+ {
734
+ "epoch": 3.12,
735
+ "learning_rate": 0.0003747323340471092,
736
+ "loss": 0.2793,
737
+ "step": 585
738
+ },
739
+ {
740
+ "epoch": 3.15,
741
+ "learning_rate": 0.00036937901498929336,
742
+ "loss": 0.1657,
743
+ "step": 590
744
+ },
745
+ {
746
+ "epoch": 3.18,
747
+ "learning_rate": 0.0003640256959314775,
748
+ "loss": 0.1275,
749
+ "step": 595
750
+ },
751
+ {
752
+ "epoch": 3.2,
753
+ "learning_rate": 0.00035867237687366166,
754
+ "loss": 0.3048,
755
+ "step": 600
756
+ },
757
+ {
758
+ "epoch": 3.23,
759
+ "learning_rate": 0.00035331905781584586,
760
+ "loss": 0.1886,
761
+ "step": 605
762
+ },
763
+ {
764
+ "epoch": 3.26,
765
+ "learning_rate": 0.00034796573875803,
766
+ "loss": 0.1595,
767
+ "step": 610
768
+ },
769
+ {
770
+ "epoch": 3.28,
771
+ "learning_rate": 0.0003426124197002141,
772
+ "loss": 0.2196,
773
+ "step": 615
774
+ },
775
+ {
776
+ "epoch": 3.31,
777
+ "learning_rate": 0.0003372591006423983,
778
+ "loss": 0.3155,
779
+ "step": 620
780
+ },
781
+ {
782
+ "epoch": 3.34,
783
+ "learning_rate": 0.00033190578158458244,
784
+ "loss": 0.2957,
785
+ "step": 625
786
+ },
787
+ {
788
+ "epoch": 3.36,
789
+ "learning_rate": 0.00032655246252676664,
790
+ "loss": 0.2555,
791
+ "step": 630
792
+ },
793
+ {
794
+ "epoch": 3.39,
795
+ "learning_rate": 0.00032119914346895073,
796
+ "loss": 0.2318,
797
+ "step": 635
798
+ },
799
+ {
800
+ "epoch": 3.42,
801
+ "learning_rate": 0.0003158458244111349,
802
+ "loss": 0.271,
803
+ "step": 640
804
+ },
805
+ {
806
+ "epoch": 3.44,
807
+ "learning_rate": 0.0003104925053533191,
808
+ "loss": 0.1326,
809
+ "step": 645
810
+ },
811
+ {
812
+ "epoch": 3.47,
813
+ "learning_rate": 0.00030513918629550323,
814
+ "loss": 0.184,
815
+ "step": 650
816
+ },
817
+ {
818
+ "epoch": 3.5,
819
+ "learning_rate": 0.0002997858672376874,
820
+ "loss": 0.1722,
821
+ "step": 655
822
+ },
823
+ {
824
+ "epoch": 3.52,
825
+ "learning_rate": 0.0002944325481798715,
826
+ "loss": 0.1703,
827
+ "step": 660
828
+ },
829
+ {
830
+ "epoch": 3.55,
831
+ "learning_rate": 0.0002890792291220557,
832
+ "loss": 0.1832,
833
+ "step": 665
834
+ },
835
+ {
836
+ "epoch": 3.58,
837
+ "learning_rate": 0.0002837259100642398,
838
+ "loss": 0.0724,
839
+ "step": 670
840
+ },
841
+ {
842
+ "epoch": 3.6,
843
+ "learning_rate": 0.00027837259100642396,
844
+ "loss": 0.2362,
845
+ "step": 675
846
+ },
847
+ {
848
+ "epoch": 3.63,
849
+ "learning_rate": 0.00027301927194860816,
850
+ "loss": 0.1697,
851
+ "step": 680
852
+ },
853
+ {
854
+ "epoch": 3.66,
855
+ "learning_rate": 0.0002676659528907923,
856
+ "loss": 0.2273,
857
+ "step": 685
858
+ },
859
+ {
860
+ "epoch": 3.68,
861
+ "learning_rate": 0.00026231263383297645,
862
+ "loss": 0.1521,
863
+ "step": 690
864
+ },
865
+ {
866
+ "epoch": 3.71,
867
+ "learning_rate": 0.0002569593147751606,
868
+ "loss": 0.1556,
869
+ "step": 695
870
+ },
871
+ {
872
+ "epoch": 3.74,
873
+ "learning_rate": 0.00025160599571734475,
874
+ "loss": 0.2266,
875
+ "step": 700
876
+ },
877
+ {
878
+ "epoch": 3.77,
879
+ "learning_rate": 0.0002462526766595289,
880
+ "loss": 0.1827,
881
+ "step": 705
882
+ },
883
+ {
884
+ "epoch": 3.79,
885
+ "learning_rate": 0.00024089935760171306,
886
+ "loss": 0.083,
887
+ "step": 710
888
+ },
889
+ {
890
+ "epoch": 3.82,
891
+ "learning_rate": 0.0002355460385438972,
892
+ "loss": 0.1368,
893
+ "step": 715
894
+ },
895
+ {
896
+ "epoch": 3.85,
897
+ "learning_rate": 0.00023019271948608136,
898
+ "loss": 0.1649,
899
+ "step": 720
900
+ },
901
+ {
902
+ "epoch": 3.87,
903
+ "learning_rate": 0.00022483940042826553,
904
+ "loss": 0.1999,
905
+ "step": 725
906
+ },
907
+ {
908
+ "epoch": 3.9,
909
+ "learning_rate": 0.00021948608137044968,
910
+ "loss": 0.147,
911
+ "step": 730
912
+ },
913
+ {
914
+ "epoch": 3.93,
915
+ "learning_rate": 0.00021413276231263385,
916
+ "loss": 0.1472,
917
+ "step": 735
918
+ },
919
+ {
920
+ "epoch": 3.95,
921
+ "learning_rate": 0.000208779443254818,
922
+ "loss": 0.1193,
923
+ "step": 740
924
+ },
925
+ {
926
+ "epoch": 3.98,
927
+ "learning_rate": 0.00020342612419700214,
928
+ "loss": 0.1351,
929
+ "step": 745
930
+ },
931
+ {
932
+ "epoch": 4.0,
933
+ "eval_accuracy": 0.9563676633444075,
934
+ "eval_loss": 0.10276087373495102,
935
+ "eval_runtime": 207.7116,
936
+ "eval_samples_per_second": 21.737,
937
+ "eval_steps_per_second": 2.72,
938
+ "step": 749
939
+ },
940
+ {
941
+ "epoch": 4.01,
942
+ "learning_rate": 0.0001980728051391863,
943
+ "loss": 0.2576,
944
+ "step": 750
945
+ },
946
+ {
947
+ "epoch": 4.03,
948
+ "learning_rate": 0.00019271948608137046,
949
+ "loss": 0.1087,
950
+ "step": 755
951
+ },
952
+ {
953
+ "epoch": 4.06,
954
+ "learning_rate": 0.0001873661670235546,
955
+ "loss": 0.12,
956
+ "step": 760
957
+ },
958
+ {
959
+ "epoch": 4.09,
960
+ "learning_rate": 0.00018201284796573876,
961
+ "loss": 0.1427,
962
+ "step": 765
963
+ },
964
+ {
965
+ "epoch": 4.11,
966
+ "learning_rate": 0.00017665952890792293,
967
+ "loss": 0.1297,
968
+ "step": 770
969
+ },
970
+ {
971
+ "epoch": 4.14,
972
+ "learning_rate": 0.00017130620985010705,
973
+ "loss": 0.0887,
974
+ "step": 775
975
+ },
976
+ {
977
+ "epoch": 4.17,
978
+ "learning_rate": 0.00016595289079229122,
979
+ "loss": 0.0872,
980
+ "step": 780
981
+ },
982
+ {
983
+ "epoch": 4.19,
984
+ "learning_rate": 0.00016059957173447537,
985
+ "loss": 0.1259,
986
+ "step": 785
987
+ },
988
+ {
989
+ "epoch": 4.22,
990
+ "learning_rate": 0.00015524625267665954,
991
+ "loss": 0.1486,
992
+ "step": 790
993
+ },
994
+ {
995
+ "epoch": 4.25,
996
+ "learning_rate": 0.0001498929336188437,
997
+ "loss": 0.1123,
998
+ "step": 795
999
+ },
1000
+ {
1001
+ "epoch": 4.27,
1002
+ "learning_rate": 0.00014453961456102786,
1003
+ "loss": 0.1699,
1004
+ "step": 800
1005
+ },
1006
+ {
1007
+ "epoch": 4.3,
1008
+ "learning_rate": 0.00013918629550321198,
1009
+ "loss": 0.0964,
1010
+ "step": 805
1011
+ },
1012
+ {
1013
+ "epoch": 4.33,
1014
+ "learning_rate": 0.00013383297644539615,
1015
+ "loss": 0.1556,
1016
+ "step": 810
1017
+ },
1018
+ {
1019
+ "epoch": 4.35,
1020
+ "learning_rate": 0.0001284796573875803,
1021
+ "loss": 0.1347,
1022
+ "step": 815
1023
+ },
1024
+ {
1025
+ "epoch": 4.38,
1026
+ "learning_rate": 0.00012312633832976445,
1027
+ "loss": 0.1876,
1028
+ "step": 820
1029
+ },
1030
+ {
1031
+ "epoch": 4.41,
1032
+ "learning_rate": 0.0001177730192719486,
1033
+ "loss": 0.0863,
1034
+ "step": 825
1035
+ },
1036
+ {
1037
+ "epoch": 4.43,
1038
+ "learning_rate": 0.00011241970021413277,
1039
+ "loss": 0.1614,
1040
+ "step": 830
1041
+ },
1042
+ {
1043
+ "epoch": 4.46,
1044
+ "learning_rate": 0.00010706638115631693,
1045
+ "loss": 0.1407,
1046
+ "step": 835
1047
+ },
1048
+ {
1049
+ "epoch": 4.49,
1050
+ "learning_rate": 0.00010171306209850107,
1051
+ "loss": 0.0918,
1052
+ "step": 840
1053
+ },
1054
+ {
1055
+ "epoch": 4.51,
1056
+ "learning_rate": 9.635974304068523e-05,
1057
+ "loss": 0.0966,
1058
+ "step": 845
1059
+ },
1060
+ {
1061
+ "epoch": 4.54,
1062
+ "learning_rate": 9.100642398286938e-05,
1063
+ "loss": 0.0563,
1064
+ "step": 850
1065
+ },
1066
+ {
1067
+ "epoch": 4.57,
1068
+ "learning_rate": 8.565310492505352e-05,
1069
+ "loss": 0.0795,
1070
+ "step": 855
1071
+ },
1072
+ {
1073
+ "epoch": 4.59,
1074
+ "learning_rate": 8.029978586723768e-05,
1075
+ "loss": 0.0409,
1076
+ "step": 860
1077
+ },
1078
+ {
1079
+ "epoch": 4.62,
1080
+ "learning_rate": 7.494646680942184e-05,
1081
+ "loss": 0.0707,
1082
+ "step": 865
1083
+ },
1084
+ {
1085
+ "epoch": 4.65,
1086
+ "learning_rate": 6.959314775160599e-05,
1087
+ "loss": 0.1622,
1088
+ "step": 870
1089
+ },
1090
+ {
1091
+ "epoch": 4.67,
1092
+ "learning_rate": 6.423982869379015e-05,
1093
+ "loss": 0.0814,
1094
+ "step": 875
1095
+ },
1096
+ {
1097
+ "epoch": 4.7,
1098
+ "learning_rate": 5.88865096359743e-05,
1099
+ "loss": 0.1475,
1100
+ "step": 880
1101
+ },
1102
+ {
1103
+ "epoch": 4.73,
1104
+ "learning_rate": 5.353319057815846e-05,
1105
+ "loss": 0.129,
1106
+ "step": 885
1107
+ },
1108
+ {
1109
+ "epoch": 4.75,
1110
+ "learning_rate": 4.8179871520342616e-05,
1111
+ "loss": 0.1013,
1112
+ "step": 890
1113
+ },
1114
+ {
1115
+ "epoch": 4.78,
1116
+ "learning_rate": 4.282655246252676e-05,
1117
+ "loss": 0.0852,
1118
+ "step": 895
1119
+ },
1120
+ {
1121
+ "epoch": 4.81,
1122
+ "learning_rate": 3.747323340471092e-05,
1123
+ "loss": 0.0799,
1124
+ "step": 900
1125
+ },
1126
+ {
1127
+ "epoch": 4.83,
1128
+ "learning_rate": 3.2119914346895075e-05,
1129
+ "loss": 0.1067,
1130
+ "step": 905
1131
+ },
1132
+ {
1133
+ "epoch": 4.86,
1134
+ "learning_rate": 2.676659528907923e-05,
1135
+ "loss": 0.0713,
1136
+ "step": 910
1137
+ },
1138
+ {
1139
+ "epoch": 4.89,
1140
+ "learning_rate": 2.141327623126338e-05,
1141
+ "loss": 0.0497,
1142
+ "step": 915
1143
+ },
1144
+ {
1145
+ "epoch": 4.91,
1146
+ "learning_rate": 1.6059957173447537e-05,
1147
+ "loss": 0.0876,
1148
+ "step": 920
1149
+ },
1150
+ {
1151
+ "epoch": 4.94,
1152
+ "learning_rate": 1.070663811563169e-05,
1153
+ "loss": 0.0796,
1154
+ "step": 925
1155
+ },
1156
+ {
1157
+ "epoch": 4.97,
1158
+ "learning_rate": 5.353319057815845e-06,
1159
+ "loss": 0.0755,
1160
+ "step": 930
1161
+ },
1162
+ {
1163
+ "epoch": 4.99,
1164
+ "learning_rate": 0.0,
1165
+ "loss": 0.142,
1166
+ "step": 935
1167
+ },
1168
+ {
1169
+ "epoch": 4.99,
1170
+ "eval_accuracy": 0.9869324473975637,
1171
+ "eval_loss": 0.04178118333220482,
1172
+ "eval_runtime": 207.692,
1173
+ "eval_samples_per_second": 21.739,
1174
+ "eval_steps_per_second": 2.72,
1175
+ "step": 935
1176
+ },
1177
+ {
1178
+ "epoch": 4.99,
1179
+ "step": 935,
1180
+ "total_flos": 6.883415802275955e+18,
1181
+ "train_loss": 0.24478418331413984,
1182
+ "train_runtime": 10756.8608,
1183
+ "train_samples_per_second": 2.782,
1184
+ "train_steps_per_second": 0.087
1185
+ }
1186
+ ],
1187
+ "max_steps": 935,
1188
+ "num_train_epochs": 5,
1189
+ "total_flos": 6.883415802275955e+18,
1190
+ "trial_name": null,
1191
+ "trial_params": null
1192
+ }