MF21377197 commited on
Commit
c07108c
1 Parent(s): 4a6268b

End of training

Browse files
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "eval_accuracy": 0.9614,
4
+ "eval_loss": 0.12246201187372208,
5
+ "eval_runtime": 77.055,
6
+ "eval_samples_per_second": 64.889,
7
+ "eval_steps_per_second": 2.038,
8
+ "total_flos": 1.0638717415807058e+19,
9
+ "train_loss": 0.6938335181754313,
10
+ "train_runtime": 1591.9984,
11
+ "train_samples_per_second": 84.799,
12
+ "train_steps_per_second": 0.661
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "eval_accuracy": 0.9614,
4
+ "eval_loss": 0.12246201187372208,
5
+ "eval_runtime": 77.055,
6
+ "eval_samples_per_second": 64.889,
7
+ "eval_steps_per_second": 2.038
8
+ }
runs/Apr06_11-50-14_65ab92936a95/events.out.tfevents.1712407979.65ab92936a95.344.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7603f40c136d1398cd71f37d894350d63e9999eba1729c5eccf7ea3d7b84dc70
3
+ size 411
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "total_flos": 1.0638717415807058e+19,
4
+ "train_loss": 0.6938335181754313,
5
+ "train_runtime": 1591.9984,
6
+ "train_samples_per_second": 84.799,
7
+ "train_steps_per_second": 0.661
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,792 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9614,
3
+ "best_model_checkpoint": "clip-vit-base-patch32-finetuned-eurosat/checkpoint-1053",
4
+ "epoch": 2.9936034115138592,
5
+ "eval_steps": 500,
6
+ "global_step": 1053,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.03,
13
+ "grad_norm": 1.7241759300231934,
14
+ "learning_rate": 4.716981132075472e-06,
15
+ "loss": 2.3024,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.06,
20
+ "grad_norm": 3.8202145099639893,
21
+ "learning_rate": 9.433962264150944e-06,
22
+ "loss": 2.2975,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.09,
27
+ "grad_norm": 8.925788879394531,
28
+ "learning_rate": 1.4150943396226415e-05,
29
+ "loss": 2.2628,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.11,
34
+ "grad_norm": 16.57902717590332,
35
+ "learning_rate": 1.8867924528301888e-05,
36
+ "loss": 2.1487,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.14,
41
+ "grad_norm": 22.54959487915039,
42
+ "learning_rate": 2.358490566037736e-05,
43
+ "loss": 1.9168,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.17,
48
+ "grad_norm": 20.70589256286621,
49
+ "learning_rate": 2.830188679245283e-05,
50
+ "loss": 1.7142,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.2,
55
+ "grad_norm": 25.339296340942383,
56
+ "learning_rate": 3.30188679245283e-05,
57
+ "loss": 1.5649,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.23,
62
+ "grad_norm": 30.290817260742188,
63
+ "learning_rate": 3.7735849056603776e-05,
64
+ "loss": 1.4656,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.26,
69
+ "grad_norm": 36.66164016723633,
70
+ "learning_rate": 4.245283018867925e-05,
71
+ "loss": 1.2988,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.28,
76
+ "grad_norm": 37.965667724609375,
77
+ "learning_rate": 4.716981132075472e-05,
78
+ "loss": 1.3997,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.31,
83
+ "grad_norm": 22.961999893188477,
84
+ "learning_rate": 4.978880675818374e-05,
85
+ "loss": 1.1785,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.34,
90
+ "grad_norm": 17.519174575805664,
91
+ "learning_rate": 4.9260823653643085e-05,
92
+ "loss": 1.1307,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.37,
97
+ "grad_norm": 22.371517181396484,
98
+ "learning_rate": 4.8732840549102435e-05,
99
+ "loss": 1.1619,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.4,
104
+ "grad_norm": 26.505937576293945,
105
+ "learning_rate": 4.820485744456177e-05,
106
+ "loss": 1.0016,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.43,
111
+ "grad_norm": 26.786195755004883,
112
+ "learning_rate": 4.767687434002112e-05,
113
+ "loss": 0.9419,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.45,
118
+ "grad_norm": 27.246932983398438,
119
+ "learning_rate": 4.7148891235480466e-05,
120
+ "loss": 0.9388,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.48,
125
+ "grad_norm": 26.55625343322754,
126
+ "learning_rate": 4.662090813093981e-05,
127
+ "loss": 0.9833,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.51,
132
+ "grad_norm": 31.99274253845215,
133
+ "learning_rate": 4.609292502639916e-05,
134
+ "loss": 0.9239,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.54,
139
+ "grad_norm": 29.839174270629883,
140
+ "learning_rate": 4.55649419218585e-05,
141
+ "loss": 0.8898,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.57,
146
+ "grad_norm": 30.187131881713867,
147
+ "learning_rate": 4.503695881731785e-05,
148
+ "loss": 0.8662,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.6,
153
+ "grad_norm": 26.78498649597168,
154
+ "learning_rate": 4.45089757127772e-05,
155
+ "loss": 0.8236,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.63,
160
+ "grad_norm": 21.103544235229492,
161
+ "learning_rate": 4.398099260823654e-05,
162
+ "loss": 0.8185,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.65,
167
+ "grad_norm": 19.34087371826172,
168
+ "learning_rate": 4.3453009503695884e-05,
169
+ "loss": 0.8716,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.68,
174
+ "grad_norm": 14.883440017700195,
175
+ "learning_rate": 4.292502639915523e-05,
176
+ "loss": 0.8187,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.71,
181
+ "grad_norm": 16.855531692504883,
182
+ "learning_rate": 4.239704329461457e-05,
183
+ "loss": 0.7924,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.74,
188
+ "grad_norm": 14.5891695022583,
189
+ "learning_rate": 4.186906019007392e-05,
190
+ "loss": 0.7738,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.77,
195
+ "grad_norm": 17.91446876525879,
196
+ "learning_rate": 4.1341077085533265e-05,
197
+ "loss": 0.7254,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.8,
202
+ "grad_norm": 16.657079696655273,
203
+ "learning_rate": 4.081309398099261e-05,
204
+ "loss": 0.7443,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.82,
209
+ "grad_norm": 28.285350799560547,
210
+ "learning_rate": 4.028511087645195e-05,
211
+ "loss": 0.7667,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.85,
216
+ "grad_norm": 15.021693229675293,
217
+ "learning_rate": 3.97571277719113e-05,
218
+ "loss": 0.7379,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.88,
223
+ "grad_norm": 24.571500778198242,
224
+ "learning_rate": 3.9229144667370646e-05,
225
+ "loss": 0.7032,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.91,
230
+ "grad_norm": 18.02920150756836,
231
+ "learning_rate": 3.870116156282999e-05,
232
+ "loss": 0.7563,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.94,
237
+ "grad_norm": 24.86216926574707,
238
+ "learning_rate": 3.817317845828934e-05,
239
+ "loss": 0.7057,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.97,
244
+ "grad_norm": 21.064966201782227,
245
+ "learning_rate": 3.764519535374868e-05,
246
+ "loss": 0.704,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.0,
251
+ "grad_norm": 17.003034591674805,
252
+ "learning_rate": 3.711721224920803e-05,
253
+ "loss": 0.6844,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.0,
258
+ "eval_accuracy": 0.8944,
259
+ "eval_loss": 0.32426366209983826,
260
+ "eval_runtime": 45.5941,
261
+ "eval_samples_per_second": 109.663,
262
+ "eval_steps_per_second": 3.443,
263
+ "step": 351
264
+ },
265
+ {
266
+ "epoch": 1.02,
267
+ "grad_norm": 16.356178283691406,
268
+ "learning_rate": 3.658922914466738e-05,
269
+ "loss": 0.6784,
270
+ "step": 360
271
+ },
272
+ {
273
+ "epoch": 1.05,
274
+ "grad_norm": 17.41063690185547,
275
+ "learning_rate": 3.6061246040126714e-05,
276
+ "loss": 0.6746,
277
+ "step": 370
278
+ },
279
+ {
280
+ "epoch": 1.08,
281
+ "grad_norm": 22.446226119995117,
282
+ "learning_rate": 3.5533262935586064e-05,
283
+ "loss": 0.7191,
284
+ "step": 380
285
+ },
286
+ {
287
+ "epoch": 1.11,
288
+ "grad_norm": 14.2515230178833,
289
+ "learning_rate": 3.500527983104541e-05,
290
+ "loss": 0.6181,
291
+ "step": 390
292
+ },
293
+ {
294
+ "epoch": 1.14,
295
+ "grad_norm": 21.2375545501709,
296
+ "learning_rate": 3.447729672650475e-05,
297
+ "loss": 0.6146,
298
+ "step": 400
299
+ },
300
+ {
301
+ "epoch": 1.17,
302
+ "grad_norm": 30.0297794342041,
303
+ "learning_rate": 3.3949313621964095e-05,
304
+ "loss": 0.5986,
305
+ "step": 410
306
+ },
307
+ {
308
+ "epoch": 1.19,
309
+ "grad_norm": 18.535289764404297,
310
+ "learning_rate": 3.3421330517423445e-05,
311
+ "loss": 0.584,
312
+ "step": 420
313
+ },
314
+ {
315
+ "epoch": 1.22,
316
+ "grad_norm": 20.00425910949707,
317
+ "learning_rate": 3.289334741288279e-05,
318
+ "loss": 0.6802,
319
+ "step": 430
320
+ },
321
+ {
322
+ "epoch": 1.25,
323
+ "grad_norm": 16.380542755126953,
324
+ "learning_rate": 3.236536430834213e-05,
325
+ "loss": 0.6259,
326
+ "step": 440
327
+ },
328
+ {
329
+ "epoch": 1.28,
330
+ "grad_norm": 14.940071105957031,
331
+ "learning_rate": 3.183738120380148e-05,
332
+ "loss": 0.6096,
333
+ "step": 450
334
+ },
335
+ {
336
+ "epoch": 1.31,
337
+ "grad_norm": 15.746271133422852,
338
+ "learning_rate": 3.130939809926082e-05,
339
+ "loss": 0.6153,
340
+ "step": 460
341
+ },
342
+ {
343
+ "epoch": 1.34,
344
+ "grad_norm": 23.13232421875,
345
+ "learning_rate": 3.078141499472017e-05,
346
+ "loss": 0.6158,
347
+ "step": 470
348
+ },
349
+ {
350
+ "epoch": 1.36,
351
+ "grad_norm": 16.764955520629883,
352
+ "learning_rate": 3.0253431890179517e-05,
353
+ "loss": 0.5553,
354
+ "step": 480
355
+ },
356
+ {
357
+ "epoch": 1.39,
358
+ "grad_norm": 14.932197570800781,
359
+ "learning_rate": 2.972544878563886e-05,
360
+ "loss": 0.5601,
361
+ "step": 490
362
+ },
363
+ {
364
+ "epoch": 1.42,
365
+ "grad_norm": 13.885882377624512,
366
+ "learning_rate": 2.9197465681098207e-05,
367
+ "loss": 0.5535,
368
+ "step": 500
369
+ },
370
+ {
371
+ "epoch": 1.45,
372
+ "grad_norm": 17.77952003479004,
373
+ "learning_rate": 2.8669482576557548e-05,
374
+ "loss": 0.6082,
375
+ "step": 510
376
+ },
377
+ {
378
+ "epoch": 1.48,
379
+ "grad_norm": 15.429793357849121,
380
+ "learning_rate": 2.8141499472016898e-05,
381
+ "loss": 0.5267,
382
+ "step": 520
383
+ },
384
+ {
385
+ "epoch": 1.51,
386
+ "grad_norm": 16.568571090698242,
387
+ "learning_rate": 2.7613516367476245e-05,
388
+ "loss": 0.5515,
389
+ "step": 530
390
+ },
391
+ {
392
+ "epoch": 1.54,
393
+ "grad_norm": 17.248546600341797,
394
+ "learning_rate": 2.7085533262935585e-05,
395
+ "loss": 0.5345,
396
+ "step": 540
397
+ },
398
+ {
399
+ "epoch": 1.56,
400
+ "grad_norm": 12.159417152404785,
401
+ "learning_rate": 2.6557550158394935e-05,
402
+ "loss": 0.4961,
403
+ "step": 550
404
+ },
405
+ {
406
+ "epoch": 1.59,
407
+ "grad_norm": 14.413479804992676,
408
+ "learning_rate": 2.6029567053854276e-05,
409
+ "loss": 0.5178,
410
+ "step": 560
411
+ },
412
+ {
413
+ "epoch": 1.62,
414
+ "grad_norm": 14.496917724609375,
415
+ "learning_rate": 2.5501583949313622e-05,
416
+ "loss": 0.5092,
417
+ "step": 570
418
+ },
419
+ {
420
+ "epoch": 1.65,
421
+ "grad_norm": 14.64006519317627,
422
+ "learning_rate": 2.497360084477297e-05,
423
+ "loss": 0.5421,
424
+ "step": 580
425
+ },
426
+ {
427
+ "epoch": 1.68,
428
+ "grad_norm": 16.412023544311523,
429
+ "learning_rate": 2.4445617740232313e-05,
430
+ "loss": 0.5297,
431
+ "step": 590
432
+ },
433
+ {
434
+ "epoch": 1.71,
435
+ "grad_norm": 13.8049898147583,
436
+ "learning_rate": 2.391763463569166e-05,
437
+ "loss": 0.5268,
438
+ "step": 600
439
+ },
440
+ {
441
+ "epoch": 1.73,
442
+ "grad_norm": 18.511188507080078,
443
+ "learning_rate": 2.3389651531151003e-05,
444
+ "loss": 0.5105,
445
+ "step": 610
446
+ },
447
+ {
448
+ "epoch": 1.76,
449
+ "grad_norm": 14.077324867248535,
450
+ "learning_rate": 2.286166842661035e-05,
451
+ "loss": 0.4611,
452
+ "step": 620
453
+ },
454
+ {
455
+ "epoch": 1.79,
456
+ "grad_norm": 13.796318054199219,
457
+ "learning_rate": 2.2333685322069694e-05,
458
+ "loss": 0.5079,
459
+ "step": 630
460
+ },
461
+ {
462
+ "epoch": 1.82,
463
+ "grad_norm": 12.892987251281738,
464
+ "learning_rate": 2.180570221752904e-05,
465
+ "loss": 0.5148,
466
+ "step": 640
467
+ },
468
+ {
469
+ "epoch": 1.85,
470
+ "grad_norm": 16.096128463745117,
471
+ "learning_rate": 2.1277719112988384e-05,
472
+ "loss": 0.5426,
473
+ "step": 650
474
+ },
475
+ {
476
+ "epoch": 1.88,
477
+ "grad_norm": 13.577393531799316,
478
+ "learning_rate": 2.074973600844773e-05,
479
+ "loss": 0.5145,
480
+ "step": 660
481
+ },
482
+ {
483
+ "epoch": 1.9,
484
+ "grad_norm": 12.556589126586914,
485
+ "learning_rate": 2.0221752903907075e-05,
486
+ "loss": 0.472,
487
+ "step": 670
488
+ },
489
+ {
490
+ "epoch": 1.93,
491
+ "grad_norm": 21.511499404907227,
492
+ "learning_rate": 1.9693769799366422e-05,
493
+ "loss": 0.5104,
494
+ "step": 680
495
+ },
496
+ {
497
+ "epoch": 1.96,
498
+ "grad_norm": 19.831825256347656,
499
+ "learning_rate": 1.9165786694825765e-05,
500
+ "loss": 0.4633,
501
+ "step": 690
502
+ },
503
+ {
504
+ "epoch": 1.99,
505
+ "grad_norm": 17.47673797607422,
506
+ "learning_rate": 1.863780359028511e-05,
507
+ "loss": 0.4737,
508
+ "step": 700
509
+ },
510
+ {
511
+ "epoch": 2.0,
512
+ "eval_accuracy": 0.9384,
513
+ "eval_loss": 0.17878523468971252,
514
+ "eval_runtime": 44.7121,
515
+ "eval_samples_per_second": 111.827,
516
+ "eval_steps_per_second": 3.511,
517
+ "step": 703
518
+ },
519
+ {
520
+ "epoch": 2.02,
521
+ "grad_norm": 13.325546264648438,
522
+ "learning_rate": 1.810982048574446e-05,
523
+ "loss": 0.4421,
524
+ "step": 710
525
+ },
526
+ {
527
+ "epoch": 2.05,
528
+ "grad_norm": 12.169154167175293,
529
+ "learning_rate": 1.7581837381203803e-05,
530
+ "loss": 0.4291,
531
+ "step": 720
532
+ },
533
+ {
534
+ "epoch": 2.08,
535
+ "grad_norm": 15.889957427978516,
536
+ "learning_rate": 1.7053854276663146e-05,
537
+ "loss": 0.4336,
538
+ "step": 730
539
+ },
540
+ {
541
+ "epoch": 2.1,
542
+ "grad_norm": 16.29996681213379,
543
+ "learning_rate": 1.6525871172122493e-05,
544
+ "loss": 0.4241,
545
+ "step": 740
546
+ },
547
+ {
548
+ "epoch": 2.13,
549
+ "grad_norm": 14.88168716430664,
550
+ "learning_rate": 1.5997888067581837e-05,
551
+ "loss": 0.3791,
552
+ "step": 750
553
+ },
554
+ {
555
+ "epoch": 2.16,
556
+ "grad_norm": 16.35209083557129,
557
+ "learning_rate": 1.5469904963041184e-05,
558
+ "loss": 0.4707,
559
+ "step": 760
560
+ },
561
+ {
562
+ "epoch": 2.19,
563
+ "grad_norm": 14.11000919342041,
564
+ "learning_rate": 1.4941921858500529e-05,
565
+ "loss": 0.405,
566
+ "step": 770
567
+ },
568
+ {
569
+ "epoch": 2.22,
570
+ "grad_norm": 12.48878288269043,
571
+ "learning_rate": 1.4413938753959874e-05,
572
+ "loss": 0.4214,
573
+ "step": 780
574
+ },
575
+ {
576
+ "epoch": 2.25,
577
+ "grad_norm": 13.443764686584473,
578
+ "learning_rate": 1.388595564941922e-05,
579
+ "loss": 0.3816,
580
+ "step": 790
581
+ },
582
+ {
583
+ "epoch": 2.27,
584
+ "grad_norm": 11.2753267288208,
585
+ "learning_rate": 1.3357972544878563e-05,
586
+ "loss": 0.4273,
587
+ "step": 800
588
+ },
589
+ {
590
+ "epoch": 2.3,
591
+ "grad_norm": 9.510777473449707,
592
+ "learning_rate": 1.2829989440337912e-05,
593
+ "loss": 0.4543,
594
+ "step": 810
595
+ },
596
+ {
597
+ "epoch": 2.33,
598
+ "grad_norm": 15.178093910217285,
599
+ "learning_rate": 1.2302006335797255e-05,
600
+ "loss": 0.4011,
601
+ "step": 820
602
+ },
603
+ {
604
+ "epoch": 2.36,
605
+ "grad_norm": 12.959372520446777,
606
+ "learning_rate": 1.17740232312566e-05,
607
+ "loss": 0.3952,
608
+ "step": 830
609
+ },
610
+ {
611
+ "epoch": 2.39,
612
+ "grad_norm": 13.760416030883789,
613
+ "learning_rate": 1.1246040126715946e-05,
614
+ "loss": 0.4096,
615
+ "step": 840
616
+ },
617
+ {
618
+ "epoch": 2.42,
619
+ "grad_norm": 14.852051734924316,
620
+ "learning_rate": 1.0718057022175291e-05,
621
+ "loss": 0.4518,
622
+ "step": 850
623
+ },
624
+ {
625
+ "epoch": 2.44,
626
+ "grad_norm": 15.75942325592041,
627
+ "learning_rate": 1.0190073917634636e-05,
628
+ "loss": 0.4095,
629
+ "step": 860
630
+ },
631
+ {
632
+ "epoch": 2.47,
633
+ "grad_norm": 12.429645538330078,
634
+ "learning_rate": 9.662090813093982e-06,
635
+ "loss": 0.3606,
636
+ "step": 870
637
+ },
638
+ {
639
+ "epoch": 2.5,
640
+ "grad_norm": 14.525588035583496,
641
+ "learning_rate": 9.134107708553327e-06,
642
+ "loss": 0.386,
643
+ "step": 880
644
+ },
645
+ {
646
+ "epoch": 2.53,
647
+ "grad_norm": 8.88588809967041,
648
+ "learning_rate": 8.606124604012672e-06,
649
+ "loss": 0.3748,
650
+ "step": 890
651
+ },
652
+ {
653
+ "epoch": 2.56,
654
+ "grad_norm": 13.352007865905762,
655
+ "learning_rate": 8.078141499472017e-06,
656
+ "loss": 0.364,
657
+ "step": 900
658
+ },
659
+ {
660
+ "epoch": 2.59,
661
+ "grad_norm": 13.020881652832031,
662
+ "learning_rate": 7.5501583949313625e-06,
663
+ "loss": 0.371,
664
+ "step": 910
665
+ },
666
+ {
667
+ "epoch": 2.62,
668
+ "grad_norm": 10.579428672790527,
669
+ "learning_rate": 7.022175290390708e-06,
670
+ "loss": 0.3952,
671
+ "step": 920
672
+ },
673
+ {
674
+ "epoch": 2.64,
675
+ "grad_norm": 14.498714447021484,
676
+ "learning_rate": 6.494192185850054e-06,
677
+ "loss": 0.3954,
678
+ "step": 930
679
+ },
680
+ {
681
+ "epoch": 2.67,
682
+ "grad_norm": 11.845131874084473,
683
+ "learning_rate": 5.966209081309398e-06,
684
+ "loss": 0.3882,
685
+ "step": 940
686
+ },
687
+ {
688
+ "epoch": 2.7,
689
+ "grad_norm": 9.537396430969238,
690
+ "learning_rate": 5.438225976768744e-06,
691
+ "loss": 0.3686,
692
+ "step": 950
693
+ },
694
+ {
695
+ "epoch": 2.73,
696
+ "grad_norm": 11.347996711730957,
697
+ "learning_rate": 4.910242872228089e-06,
698
+ "loss": 0.362,
699
+ "step": 960
700
+ },
701
+ {
702
+ "epoch": 2.76,
703
+ "grad_norm": 13.484134674072266,
704
+ "learning_rate": 4.382259767687434e-06,
705
+ "loss": 0.3536,
706
+ "step": 970
707
+ },
708
+ {
709
+ "epoch": 2.79,
710
+ "grad_norm": 12.782801628112793,
711
+ "learning_rate": 3.854276663146779e-06,
712
+ "loss": 0.3648,
713
+ "step": 980
714
+ },
715
+ {
716
+ "epoch": 2.81,
717
+ "grad_norm": 11.670198440551758,
718
+ "learning_rate": 3.326293558606125e-06,
719
+ "loss": 0.3361,
720
+ "step": 990
721
+ },
722
+ {
723
+ "epoch": 2.84,
724
+ "grad_norm": 14.617063522338867,
725
+ "learning_rate": 2.79831045406547e-06,
726
+ "loss": 0.3345,
727
+ "step": 1000
728
+ },
729
+ {
730
+ "epoch": 2.87,
731
+ "grad_norm": 10.081778526306152,
732
+ "learning_rate": 2.2703273495248154e-06,
733
+ "loss": 0.3359,
734
+ "step": 1010
735
+ },
736
+ {
737
+ "epoch": 2.9,
738
+ "grad_norm": 13.262909889221191,
739
+ "learning_rate": 1.7423442449841606e-06,
740
+ "loss": 0.3221,
741
+ "step": 1020
742
+ },
743
+ {
744
+ "epoch": 2.93,
745
+ "grad_norm": 13.598889350891113,
746
+ "learning_rate": 1.2143611404435059e-06,
747
+ "loss": 0.3582,
748
+ "step": 1030
749
+ },
750
+ {
751
+ "epoch": 2.96,
752
+ "grad_norm": 11.878829002380371,
753
+ "learning_rate": 6.863780359028511e-07,
754
+ "loss": 0.3085,
755
+ "step": 1040
756
+ },
757
+ {
758
+ "epoch": 2.99,
759
+ "grad_norm": 13.71689224243164,
760
+ "learning_rate": 1.5839493136219642e-07,
761
+ "loss": 0.3252,
762
+ "step": 1050
763
+ },
764
+ {
765
+ "epoch": 2.99,
766
+ "eval_accuracy": 0.9614,
767
+ "eval_loss": 0.12246201187372208,
768
+ "eval_runtime": 45.5034,
769
+ "eval_samples_per_second": 109.882,
770
+ "eval_steps_per_second": 3.45,
771
+ "step": 1053
772
+ },
773
+ {
774
+ "epoch": 2.99,
775
+ "step": 1053,
776
+ "total_flos": 1.0638717415807058e+19,
777
+ "train_loss": 0.6938335181754313,
778
+ "train_runtime": 1591.9984,
779
+ "train_samples_per_second": 84.799,
780
+ "train_steps_per_second": 0.661
781
+ }
782
+ ],
783
+ "logging_steps": 10,
784
+ "max_steps": 1053,
785
+ "num_input_tokens_seen": 0,
786
+ "num_train_epochs": 3,
787
+ "save_steps": 500,
788
+ "total_flos": 1.0638717415807058e+19,
789
+ "train_batch_size": 32,
790
+ "trial_name": null,
791
+ "trial_params": null
792
+ }