Abhiram4 commited on
Commit
f298e42
1 Parent(s): 7f23e22

End of training

Browse files
Files changed (4) hide show
  1. all_results.json +13 -0
  2. eval_results.json +8 -0
  3. train_results.json +8 -0
  4. trainer_state.json +678 -0
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9959458017710445,
4
+ "eval_loss": 0.18922357261180878,
5
+ "eval_runtime": 135.9847,
6
+ "eval_samples_per_second": 68.927,
7
+ "eval_steps_per_second": 1.081,
8
+ "total_flos": 1.7439237705052348e+19,
9
+ "train_loss": 0.789499712071728,
10
+ "train_runtime": 9726.3121,
11
+ "train_samples_per_second": 23.129,
12
+ "train_steps_per_second": 0.09
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_accuracy": 0.9959458017710445,
4
+ "eval_loss": 0.18922357261180878,
5
+ "eval_runtime": 135.9847,
6
+ "eval_samples_per_second": 68.927,
7
+ "eval_steps_per_second": 1.081
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 1.7439237705052348e+19,
4
+ "train_loss": 0.789499712071728,
5
+ "train_runtime": 9726.3121,
6
+ "train_samples_per_second": 23.129,
7
+ "train_steps_per_second": 0.09
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,678 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.9959458017710445,
3
+ "best_model_checkpoint": "PlantDiseaseDetectorVit2/checkpoint-879",
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 879,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.034129692832764506,
13
+ "grad_norm": 0.6686996817588806,
14
+ "learning_rate": 5.681818181818182e-06,
15
+ "loss": 3.7946,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.06825938566552901,
20
+ "grad_norm": 0.6528059244155884,
21
+ "learning_rate": 1.1363636363636365e-05,
22
+ "loss": 3.7634,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.10238907849829351,
27
+ "grad_norm": 0.7110876441001892,
28
+ "learning_rate": 1.7045454545454546e-05,
29
+ "loss": 3.6945,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.13651877133105803,
34
+ "grad_norm": 0.7198315858840942,
35
+ "learning_rate": 2.272727272727273e-05,
36
+ "loss": 3.5857,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.17064846416382254,
41
+ "grad_norm": 0.8180440068244934,
42
+ "learning_rate": 2.8409090909090912e-05,
43
+ "loss": 3.4298,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.20477815699658702,
48
+ "grad_norm": 0.8139916658401489,
49
+ "learning_rate": 3.409090909090909e-05,
50
+ "loss": 3.2427,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.23890784982935154,
55
+ "grad_norm": 0.8880642056465149,
56
+ "learning_rate": 3.9772727272727275e-05,
57
+ "loss": 3.013,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.27303754266211605,
62
+ "grad_norm": 0.8926830291748047,
63
+ "learning_rate": 4.545454545454546e-05,
64
+ "loss": 2.7552,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.30716723549488056,
69
+ "grad_norm": 1.002000093460083,
70
+ "learning_rate": 4.9873577749683945e-05,
71
+ "loss": 2.5021,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.3412969283276451,
76
+ "grad_norm": 0.9452126622200012,
77
+ "learning_rate": 4.924146649810367e-05,
78
+ "loss": 2.2347,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.37542662116040953,
83
+ "grad_norm": 1.039056420326233,
84
+ "learning_rate": 4.860935524652339e-05,
85
+ "loss": 2.0152,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.40955631399317405,
90
+ "grad_norm": 0.9824926853179932,
91
+ "learning_rate": 4.797724399494311e-05,
92
+ "loss": 1.8409,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.44368600682593856,
97
+ "grad_norm": 1.0370389223098755,
98
+ "learning_rate": 4.734513274336283e-05,
99
+ "loss": 1.6624,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.4778156996587031,
104
+ "grad_norm": 0.9628086090087891,
105
+ "learning_rate": 4.6713021491782554e-05,
106
+ "loss": 1.5434,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.5119453924914675,
111
+ "grad_norm": 1.0051034688949585,
112
+ "learning_rate": 4.608091024020228e-05,
113
+ "loss": 1.4022,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.5460750853242321,
118
+ "grad_norm": 0.9724010825157166,
119
+ "learning_rate": 4.5448798988622e-05,
120
+ "loss": 1.2981,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.5802047781569966,
125
+ "grad_norm": 1.2477946281433105,
126
+ "learning_rate": 4.4816687737041726e-05,
127
+ "loss": 1.2025,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.6143344709897611,
132
+ "grad_norm": 0.9471719861030579,
133
+ "learning_rate": 4.418457648546144e-05,
134
+ "loss": 1.1202,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.6484641638225256,
139
+ "grad_norm": 0.9492740631103516,
140
+ "learning_rate": 4.355246523388117e-05,
141
+ "loss": 1.053,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.6825938566552902,
146
+ "grad_norm": 0.9678471088409424,
147
+ "learning_rate": 4.2920353982300885e-05,
148
+ "loss": 0.9739,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.7167235494880546,
153
+ "grad_norm": 0.902233362197876,
154
+ "learning_rate": 4.2288242730720607e-05,
155
+ "loss": 0.9053,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.7508532423208191,
160
+ "grad_norm": 1.190834879875183,
161
+ "learning_rate": 4.165613147914033e-05,
162
+ "loss": 0.8599,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.7849829351535836,
167
+ "grad_norm": 1.094313144683838,
168
+ "learning_rate": 4.102402022756005e-05,
169
+ "loss": 0.8202,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.8191126279863481,
174
+ "grad_norm": 0.8496561050415039,
175
+ "learning_rate": 4.039190897597978e-05,
176
+ "loss": 0.7651,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.8532423208191127,
181
+ "grad_norm": 0.8702057003974915,
182
+ "learning_rate": 3.9759797724399494e-05,
183
+ "loss": 0.7268,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.8873720136518771,
188
+ "grad_norm": 0.9215129017829895,
189
+ "learning_rate": 3.912768647281922e-05,
190
+ "loss": 0.6796,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.9215017064846417,
195
+ "grad_norm": 1.1348512172698975,
196
+ "learning_rate": 3.849557522123894e-05,
197
+ "loss": 0.6446,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.9556313993174061,
202
+ "grad_norm": 0.8226602077484131,
203
+ "learning_rate": 3.7863463969658666e-05,
204
+ "loss": 0.6234,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.9897610921501706,
209
+ "grad_norm": 0.9013721942901611,
210
+ "learning_rate": 3.723135271807838e-05,
211
+ "loss": 0.5919,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.0,
216
+ "eval_accuracy": 0.9869838898965112,
217
+ "eval_loss": 0.5612290501594543,
218
+ "eval_runtime": 136.7393,
219
+ "eval_samples_per_second": 68.546,
220
+ "eval_steps_per_second": 1.075,
221
+ "step": 293
222
+ },
223
+ {
224
+ "epoch": 1.023890784982935,
225
+ "grad_norm": 0.7781492471694946,
226
+ "learning_rate": 3.659924146649811e-05,
227
+ "loss": 0.5557,
228
+ "step": 300
229
+ },
230
+ {
231
+ "epoch": 1.0580204778156996,
232
+ "grad_norm": 0.8166685104370117,
233
+ "learning_rate": 3.5967130214917824e-05,
234
+ "loss": 0.5387,
235
+ "step": 310
236
+ },
237
+ {
238
+ "epoch": 1.0921501706484642,
239
+ "grad_norm": 0.9886695742607117,
240
+ "learning_rate": 3.533501896333755e-05,
241
+ "loss": 0.5075,
242
+ "step": 320
243
+ },
244
+ {
245
+ "epoch": 1.1262798634812285,
246
+ "grad_norm": 0.90567946434021,
247
+ "learning_rate": 3.470290771175727e-05,
248
+ "loss": 0.4924,
249
+ "step": 330
250
+ },
251
+ {
252
+ "epoch": 1.1604095563139931,
253
+ "grad_norm": 0.720587432384491,
254
+ "learning_rate": 3.407079646017699e-05,
255
+ "loss": 0.4753,
256
+ "step": 340
257
+ },
258
+ {
259
+ "epoch": 1.1945392491467577,
260
+ "grad_norm": 0.8734786510467529,
261
+ "learning_rate": 3.343868520859672e-05,
262
+ "loss": 0.4488,
263
+ "step": 350
264
+ },
265
+ {
266
+ "epoch": 1.2286689419795223,
267
+ "grad_norm": 1.006568431854248,
268
+ "learning_rate": 3.280657395701643e-05,
269
+ "loss": 0.4436,
270
+ "step": 360
271
+ },
272
+ {
273
+ "epoch": 1.2627986348122868,
274
+ "grad_norm": 0.6803867816925049,
275
+ "learning_rate": 3.217446270543616e-05,
276
+ "loss": 0.4152,
277
+ "step": 370
278
+ },
279
+ {
280
+ "epoch": 1.2969283276450512,
281
+ "grad_norm": 0.6060165166854858,
282
+ "learning_rate": 3.1542351453855877e-05,
283
+ "loss": 0.4105,
284
+ "step": 380
285
+ },
286
+ {
287
+ "epoch": 1.3310580204778157,
288
+ "grad_norm": 0.6753475069999695,
289
+ "learning_rate": 3.0910240202275605e-05,
290
+ "loss": 0.3933,
291
+ "step": 390
292
+ },
293
+ {
294
+ "epoch": 1.36518771331058,
295
+ "grad_norm": 0.6579934358596802,
296
+ "learning_rate": 3.0278128950695323e-05,
297
+ "loss": 0.3822,
298
+ "step": 400
299
+ },
300
+ {
301
+ "epoch": 1.3993174061433447,
302
+ "grad_norm": 0.511236310005188,
303
+ "learning_rate": 2.964601769911505e-05,
304
+ "loss": 0.3665,
305
+ "step": 410
306
+ },
307
+ {
308
+ "epoch": 1.4334470989761092,
309
+ "grad_norm": 0.6013672947883606,
310
+ "learning_rate": 2.9013906447534767e-05,
311
+ "loss": 0.3581,
312
+ "step": 420
313
+ },
314
+ {
315
+ "epoch": 1.4675767918088738,
316
+ "grad_norm": 0.9311569929122925,
317
+ "learning_rate": 2.8381795195954492e-05,
318
+ "loss": 0.3556,
319
+ "step": 430
320
+ },
321
+ {
322
+ "epoch": 1.5017064846416384,
323
+ "grad_norm": 0.6122246980667114,
324
+ "learning_rate": 2.774968394437421e-05,
325
+ "loss": 0.3421,
326
+ "step": 440
327
+ },
328
+ {
329
+ "epoch": 1.5358361774744027,
330
+ "grad_norm": 0.5536355376243591,
331
+ "learning_rate": 2.7117572692793936e-05,
332
+ "loss": 0.327,
333
+ "step": 450
334
+ },
335
+ {
336
+ "epoch": 1.5699658703071673,
337
+ "grad_norm": 0.5841261744499207,
338
+ "learning_rate": 2.6485461441213654e-05,
339
+ "loss": 0.3227,
340
+ "step": 460
341
+ },
342
+ {
343
+ "epoch": 1.6040955631399316,
344
+ "grad_norm": 1.0431301593780518,
345
+ "learning_rate": 2.5853350189633372e-05,
346
+ "loss": 0.3152,
347
+ "step": 470
348
+ },
349
+ {
350
+ "epoch": 1.6382252559726962,
351
+ "grad_norm": 0.8714063167572021,
352
+ "learning_rate": 2.5221238938053098e-05,
353
+ "loss": 0.3142,
354
+ "step": 480
355
+ },
356
+ {
357
+ "epoch": 1.6723549488054608,
358
+ "grad_norm": 0.7647233605384827,
359
+ "learning_rate": 2.458912768647282e-05,
360
+ "loss": 0.3049,
361
+ "step": 490
362
+ },
363
+ {
364
+ "epoch": 1.7064846416382253,
365
+ "grad_norm": 1.1779247522354126,
366
+ "learning_rate": 2.3957016434892544e-05,
367
+ "loss": 0.2965,
368
+ "step": 500
369
+ },
370
+ {
371
+ "epoch": 1.74061433447099,
372
+ "grad_norm": 1.051600694656372,
373
+ "learning_rate": 2.3324905183312266e-05,
374
+ "loss": 0.29,
375
+ "step": 510
376
+ },
377
+ {
378
+ "epoch": 1.7747440273037542,
379
+ "grad_norm": 0.7569793462753296,
380
+ "learning_rate": 2.2692793931731988e-05,
381
+ "loss": 0.2962,
382
+ "step": 520
383
+ },
384
+ {
385
+ "epoch": 1.8088737201365188,
386
+ "grad_norm": 0.6314426064491272,
387
+ "learning_rate": 2.206068268015171e-05,
388
+ "loss": 0.2763,
389
+ "step": 530
390
+ },
391
+ {
392
+ "epoch": 1.8430034129692832,
393
+ "grad_norm": 0.6302042603492737,
394
+ "learning_rate": 2.1428571428571428e-05,
395
+ "loss": 0.2653,
396
+ "step": 540
397
+ },
398
+ {
399
+ "epoch": 1.8771331058020477,
400
+ "grad_norm": 0.8586634993553162,
401
+ "learning_rate": 2.079646017699115e-05,
402
+ "loss": 0.2651,
403
+ "step": 550
404
+ },
405
+ {
406
+ "epoch": 1.9112627986348123,
407
+ "grad_norm": 0.5459042191505432,
408
+ "learning_rate": 2.016434892541087e-05,
409
+ "loss": 0.256,
410
+ "step": 560
411
+ },
412
+ {
413
+ "epoch": 1.9453924914675769,
414
+ "grad_norm": 0.4690670669078827,
415
+ "learning_rate": 1.9532237673830593e-05,
416
+ "loss": 0.2542,
417
+ "step": 570
418
+ },
419
+ {
420
+ "epoch": 1.9795221843003414,
421
+ "grad_norm": 0.6143674850463867,
422
+ "learning_rate": 1.8900126422250315e-05,
423
+ "loss": 0.2529,
424
+ "step": 580
425
+ },
426
+ {
427
+ "epoch": 2.0,
428
+ "eval_accuracy": 0.9951989757815001,
429
+ "eval_loss": 0.24359659850597382,
430
+ "eval_runtime": 134.2967,
431
+ "eval_samples_per_second": 69.793,
432
+ "eval_steps_per_second": 1.095,
433
+ "step": 586
434
+ },
435
+ {
436
+ "epoch": 2.013651877133106,
437
+ "grad_norm": 0.40023496747016907,
438
+ "learning_rate": 1.8268015170670037e-05,
439
+ "loss": 0.2421,
440
+ "step": 590
441
+ },
442
+ {
443
+ "epoch": 2.04778156996587,
444
+ "grad_norm": 0.5167109370231628,
445
+ "learning_rate": 1.7635903919089762e-05,
446
+ "loss": 0.2371,
447
+ "step": 600
448
+ },
449
+ {
450
+ "epoch": 2.0819112627986347,
451
+ "grad_norm": 0.5056605935096741,
452
+ "learning_rate": 1.7003792667509484e-05,
453
+ "loss": 0.232,
454
+ "step": 610
455
+ },
456
+ {
457
+ "epoch": 2.1160409556313993,
458
+ "grad_norm": 0.8010804057121277,
459
+ "learning_rate": 1.6371681415929206e-05,
460
+ "loss": 0.2306,
461
+ "step": 620
462
+ },
463
+ {
464
+ "epoch": 2.150170648464164,
465
+ "grad_norm": 0.47784796357154846,
466
+ "learning_rate": 1.5739570164348927e-05,
467
+ "loss": 0.2229,
468
+ "step": 630
469
+ },
470
+ {
471
+ "epoch": 2.1843003412969284,
472
+ "grad_norm": 0.5715758204460144,
473
+ "learning_rate": 1.510745891276865e-05,
474
+ "loss": 0.2255,
475
+ "step": 640
476
+ },
477
+ {
478
+ "epoch": 2.218430034129693,
479
+ "grad_norm": 0.3998388350009918,
480
+ "learning_rate": 1.4475347661188371e-05,
481
+ "loss": 0.219,
482
+ "step": 650
483
+ },
484
+ {
485
+ "epoch": 2.252559726962457,
486
+ "grad_norm": 0.6502085328102112,
487
+ "learning_rate": 1.3843236409608093e-05,
488
+ "loss": 0.2155,
489
+ "step": 660
490
+ },
491
+ {
492
+ "epoch": 2.2866894197952217,
493
+ "grad_norm": 0.5454604625701904,
494
+ "learning_rate": 1.3211125158027813e-05,
495
+ "loss": 0.2113,
496
+ "step": 670
497
+ },
498
+ {
499
+ "epoch": 2.3208191126279862,
500
+ "grad_norm": 0.669148862361908,
501
+ "learning_rate": 1.2579013906447535e-05,
502
+ "loss": 0.2124,
503
+ "step": 680
504
+ },
505
+ {
506
+ "epoch": 2.354948805460751,
507
+ "grad_norm": 0.4781040847301483,
508
+ "learning_rate": 1.1946902654867258e-05,
509
+ "loss": 0.2116,
510
+ "step": 690
511
+ },
512
+ {
513
+ "epoch": 2.3890784982935154,
514
+ "grad_norm": 0.2730717360973358,
515
+ "learning_rate": 1.1314791403286978e-05,
516
+ "loss": 0.2064,
517
+ "step": 700
518
+ },
519
+ {
520
+ "epoch": 2.42320819112628,
521
+ "grad_norm": 0.30644622445106506,
522
+ "learning_rate": 1.06826801517067e-05,
523
+ "loss": 0.2022,
524
+ "step": 710
525
+ },
526
+ {
527
+ "epoch": 2.4573378839590445,
528
+ "grad_norm": 0.4491395354270935,
529
+ "learning_rate": 1.0050568900126423e-05,
530
+ "loss": 0.2083,
531
+ "step": 720
532
+ },
533
+ {
534
+ "epoch": 2.491467576791809,
535
+ "grad_norm": 0.3245304524898529,
536
+ "learning_rate": 9.418457648546145e-06,
537
+ "loss": 0.1992,
538
+ "step": 730
539
+ },
540
+ {
541
+ "epoch": 2.5255972696245736,
542
+ "grad_norm": 0.4015863537788391,
543
+ "learning_rate": 8.786346396965867e-06,
544
+ "loss": 0.1993,
545
+ "step": 740
546
+ },
547
+ {
548
+ "epoch": 2.5597269624573378,
549
+ "grad_norm": 0.7087588310241699,
550
+ "learning_rate": 8.154235145385589e-06,
551
+ "loss": 0.2034,
552
+ "step": 750
553
+ },
554
+ {
555
+ "epoch": 2.5938566552901023,
556
+ "grad_norm": 0.27920275926589966,
557
+ "learning_rate": 7.52212389380531e-06,
558
+ "loss": 0.1951,
559
+ "step": 760
560
+ },
561
+ {
562
+ "epoch": 2.627986348122867,
563
+ "grad_norm": 0.686769425868988,
564
+ "learning_rate": 6.890012642225031e-06,
565
+ "loss": 0.1956,
566
+ "step": 770
567
+ },
568
+ {
569
+ "epoch": 2.6621160409556315,
570
+ "grad_norm": 0.293285995721817,
571
+ "learning_rate": 6.257901390644753e-06,
572
+ "loss": 0.1898,
573
+ "step": 780
574
+ },
575
+ {
576
+ "epoch": 2.696245733788396,
577
+ "grad_norm": 0.3410426080226898,
578
+ "learning_rate": 5.625790139064476e-06,
579
+ "loss": 0.1905,
580
+ "step": 790
581
+ },
582
+ {
583
+ "epoch": 2.73037542662116,
584
+ "grad_norm": 0.5243946313858032,
585
+ "learning_rate": 4.993678887484197e-06,
586
+ "loss": 0.1892,
587
+ "step": 800
588
+ },
589
+ {
590
+ "epoch": 2.7645051194539247,
591
+ "grad_norm": 0.39842483401298523,
592
+ "learning_rate": 4.361567635903919e-06,
593
+ "loss": 0.1902,
594
+ "step": 810
595
+ },
596
+ {
597
+ "epoch": 2.7986348122866893,
598
+ "grad_norm": 0.26469987630844116,
599
+ "learning_rate": 3.729456384323641e-06,
600
+ "loss": 0.1906,
601
+ "step": 820
602
+ },
603
+ {
604
+ "epoch": 2.832764505119454,
605
+ "grad_norm": 0.33791306614875793,
606
+ "learning_rate": 3.097345132743363e-06,
607
+ "loss": 0.1864,
608
+ "step": 830
609
+ },
610
+ {
611
+ "epoch": 2.8668941979522184,
612
+ "grad_norm": 0.6516376733779907,
613
+ "learning_rate": 2.465233881163085e-06,
614
+ "loss": 0.1858,
615
+ "step": 840
616
+ },
617
+ {
618
+ "epoch": 2.901023890784983,
619
+ "grad_norm": 0.28408199548721313,
620
+ "learning_rate": 1.8331226295828066e-06,
621
+ "loss": 0.1933,
622
+ "step": 850
623
+ },
624
+ {
625
+ "epoch": 2.9351535836177476,
626
+ "grad_norm": 0.2557958960533142,
627
+ "learning_rate": 1.2010113780025286e-06,
628
+ "loss": 0.1877,
629
+ "step": 860
630
+ },
631
+ {
632
+ "epoch": 2.969283276450512,
633
+ "grad_norm": 0.4732515811920166,
634
+ "learning_rate": 5.689001264222504e-07,
635
+ "loss": 0.1886,
636
+ "step": 870
637
+ },
638
+ {
639
+ "epoch": 3.0,
640
+ "eval_accuracy": 0.9959458017710445,
641
+ "eval_loss": 0.18922357261180878,
642
+ "eval_runtime": 134.1896,
643
+ "eval_samples_per_second": 69.849,
644
+ "eval_steps_per_second": 1.095,
645
+ "step": 879
646
+ },
647
+ {
648
+ "epoch": 3.0,
649
+ "step": 879,
650
+ "total_flos": 1.7439237705052348e+19,
651
+ "train_loss": 0.789499712071728,
652
+ "train_runtime": 9726.3121,
653
+ "train_samples_per_second": 23.129,
654
+ "train_steps_per_second": 0.09
655
+ }
656
+ ],
657
+ "logging_steps": 10,
658
+ "max_steps": 879,
659
+ "num_input_tokens_seen": 0,
660
+ "num_train_epochs": 3,
661
+ "save_steps": 500,
662
+ "stateful_callbacks": {
663
+ "TrainerControl": {
664
+ "args": {
665
+ "should_epoch_stop": false,
666
+ "should_evaluate": false,
667
+ "should_log": false,
668
+ "should_save": true,
669
+ "should_training_stop": true
670
+ },
671
+ "attributes": {}
672
+ }
673
+ },
674
+ "total_flos": 1.7439237705052348e+19,
675
+ "train_batch_size": 64,
676
+ "trial_name": null,
677
+ "trial_params": null
678
+ }