ealdaz commited on
Commit
c0d5fe7
1 Parent(s): 5528f8a

End of training

Browse files
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: apache-2.0
3
  base_model: microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft
4
  tags:
 
 
5
  - generated_from_trainer
6
  metrics:
7
  - accuracy
@@ -15,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # swin-windturbine
17
 
18
- This model is a fine-tuned version of [microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft](https://huggingface.co/microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.0472
21
- - Accuracy: 0.9964
22
 
23
  ## Model description
24
 
 
2
  license: apache-2.0
3
  base_model: microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft
4
  tags:
5
+ - image-classification
6
+ - vision
7
  - generated_from_trainer
8
  metrics:
9
  - accuracy
 
17
 
18
  # swin-windturbine
19
 
20
+ This model is a fine-tuned version of [microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft](https://huggingface.co/microsoft/swinv2-large-patch4-window12to16-192to256-22kto1k-ft) on the sees-ml-skunkworks/windturbinedataset-cleaned dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.0337
23
+ - Accuracy: 0.9857
24
 
25
  ## Model description
26
 
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.985663082437276,
4
+ "eval_loss": 0.03369523584842682,
5
+ "eval_runtime": 18.1452,
6
+ "eval_samples_per_second": 15.376,
7
+ "eval_steps_per_second": 1.929,
8
+ "total_flos": 1.814619221089321e+18,
9
+ "train_loss": 0.15722184067514908,
10
+ "train_runtime": 1696.7849,
11
+ "train_samples_per_second": 4.644,
12
+ "train_steps_per_second": 0.581
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.985663082437276,
4
+ "eval_loss": 0.03369523584842682,
5
+ "eval_runtime": 18.1452,
6
+ "eval_samples_per_second": 15.376,
7
+ "eval_steps_per_second": 1.929
8
+ }
runs/May23_22-05-26_Edus-MacBook-Pro.local/events.out.tfevents.1716501343.Edus-MacBook-Pro.local ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dce4760cb4dec45b4817c16db14cc9c1e3f5c3c559b56dca322883ba390aefc
3
+ size 40
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "total_flos": 1.814619221089321e+18,
4
+ "train_loss": 0.15722184067514908,
5
+ "train_runtime": 1696.7849,
6
+ "train_samples_per_second": 4.644,
7
+ "train_steps_per_second": 0.581
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,773 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.03369523584842682,
3
+ "best_model_checkpoint": "./windturbine_outputs/checkpoint-394",
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 985,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.050761421319796954,
13
+ "grad_norm": 46.4831657409668,
14
+ "learning_rate": 1.9796954314720812e-05,
15
+ "loss": 1.233,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.10152284263959391,
20
+ "grad_norm": 181.1588592529297,
21
+ "learning_rate": 1.9593908629441626e-05,
22
+ "loss": 0.7105,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.15228426395939088,
27
+ "grad_norm": 23.9494686126709,
28
+ "learning_rate": 1.939086294416244e-05,
29
+ "loss": 0.499,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.20304568527918782,
34
+ "grad_norm": 51.11117935180664,
35
+ "learning_rate": 1.918781725888325e-05,
36
+ "loss": 0.3099,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.25380710659898476,
41
+ "grad_norm": 47.91529083251953,
42
+ "learning_rate": 1.8984771573604063e-05,
43
+ "loss": 0.3527,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.30456852791878175,
48
+ "grad_norm": 655.8473510742188,
49
+ "learning_rate": 1.8781725888324877e-05,
50
+ "loss": 0.4137,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.3553299492385787,
55
+ "grad_norm": 1.7267569303512573,
56
+ "learning_rate": 1.8578680203045687e-05,
57
+ "loss": 0.3445,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.40609137055837563,
62
+ "grad_norm": 11.01191234588623,
63
+ "learning_rate": 1.8375634517766498e-05,
64
+ "loss": 0.3102,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.45685279187817257,
69
+ "grad_norm": 169.2871551513672,
70
+ "learning_rate": 1.817258883248731e-05,
71
+ "loss": 0.3288,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.5076142131979695,
76
+ "grad_norm": 31.255569458007812,
77
+ "learning_rate": 1.7969543147208125e-05,
78
+ "loss": 0.1398,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.5583756345177665,
83
+ "grad_norm": 33.036746978759766,
84
+ "learning_rate": 1.7766497461928935e-05,
85
+ "loss": 0.1228,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.6091370558375635,
90
+ "grad_norm": 0.6117672920227051,
91
+ "learning_rate": 1.7563451776649745e-05,
92
+ "loss": 0.1126,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.6598984771573604,
97
+ "grad_norm": 1.1897255182266235,
98
+ "learning_rate": 1.736040609137056e-05,
99
+ "loss": 0.0152,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.7106598984771574,
104
+ "grad_norm": 438.8088073730469,
105
+ "learning_rate": 1.7157360406091373e-05,
106
+ "loss": 0.279,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.7614213197969543,
111
+ "grad_norm": 492.2366638183594,
112
+ "learning_rate": 1.6954314720812183e-05,
113
+ "loss": 0.5028,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.8121827411167513,
118
+ "grad_norm": 0.6856859922409058,
119
+ "learning_rate": 1.6751269035532997e-05,
120
+ "loss": 0.0962,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.8629441624365483,
125
+ "grad_norm": 46.26337814331055,
126
+ "learning_rate": 1.654822335025381e-05,
127
+ "loss": 0.0995,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.9137055837563451,
132
+ "grad_norm": 52.820518493652344,
133
+ "learning_rate": 1.634517766497462e-05,
134
+ "loss": 0.2824,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.9644670050761421,
139
+ "grad_norm": 0.23557256162166595,
140
+ "learning_rate": 1.614213197969543e-05,
141
+ "loss": 0.0604,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 1.0,
146
+ "eval_accuracy": 0.989247311827957,
147
+ "eval_loss": 0.0458565279841423,
148
+ "eval_runtime": 18.211,
149
+ "eval_samples_per_second": 15.32,
150
+ "eval_steps_per_second": 1.922,
151
+ "step": 197
152
+ },
153
+ {
154
+ "epoch": 1.015228426395939,
155
+ "grad_norm": 4.585972309112549,
156
+ "learning_rate": 1.5939086294416245e-05,
157
+ "loss": 0.1812,
158
+ "step": 200
159
+ },
160
+ {
161
+ "epoch": 1.0659898477157361,
162
+ "grad_norm": 0.8356128931045532,
163
+ "learning_rate": 1.573604060913706e-05,
164
+ "loss": 0.0418,
165
+ "step": 210
166
+ },
167
+ {
168
+ "epoch": 1.116751269035533,
169
+ "grad_norm": 249.66070556640625,
170
+ "learning_rate": 1.553299492385787e-05,
171
+ "loss": 0.1217,
172
+ "step": 220
173
+ },
174
+ {
175
+ "epoch": 1.16751269035533,
176
+ "grad_norm": 131.45016479492188,
177
+ "learning_rate": 1.5329949238578682e-05,
178
+ "loss": 0.2738,
179
+ "step": 230
180
+ },
181
+ {
182
+ "epoch": 1.218274111675127,
183
+ "grad_norm": 0.0076106940396130085,
184
+ "learning_rate": 1.5126903553299494e-05,
185
+ "loss": 0.0528,
186
+ "step": 240
187
+ },
188
+ {
189
+ "epoch": 1.2690355329949239,
190
+ "grad_norm": 191.85903930664062,
191
+ "learning_rate": 1.4923857868020306e-05,
192
+ "loss": 0.096,
193
+ "step": 250
194
+ },
195
+ {
196
+ "epoch": 1.3197969543147208,
197
+ "grad_norm": 43.89326477050781,
198
+ "learning_rate": 1.4720812182741118e-05,
199
+ "loss": 0.1814,
200
+ "step": 260
201
+ },
202
+ {
203
+ "epoch": 1.3705583756345177,
204
+ "grad_norm": 0.0087745301425457,
205
+ "learning_rate": 1.4517766497461929e-05,
206
+ "loss": 0.1263,
207
+ "step": 270
208
+ },
209
+ {
210
+ "epoch": 1.4213197969543148,
211
+ "grad_norm": 342.1986083984375,
212
+ "learning_rate": 1.4314720812182742e-05,
213
+ "loss": 0.3518,
214
+ "step": 280
215
+ },
216
+ {
217
+ "epoch": 1.4720812182741116,
218
+ "grad_norm": 0.011494699865579605,
219
+ "learning_rate": 1.4111675126903554e-05,
220
+ "loss": 0.0469,
221
+ "step": 290
222
+ },
223
+ {
224
+ "epoch": 1.5228426395939088,
225
+ "grad_norm": 0.02110651694238186,
226
+ "learning_rate": 1.3908629441624366e-05,
227
+ "loss": 0.0432,
228
+ "step": 300
229
+ },
230
+ {
231
+ "epoch": 1.5736040609137056,
232
+ "grad_norm": 16.259416580200195,
233
+ "learning_rate": 1.3705583756345178e-05,
234
+ "loss": 0.0037,
235
+ "step": 310
236
+ },
237
+ {
238
+ "epoch": 1.6243654822335025,
239
+ "grad_norm": 1.5435930490493774,
240
+ "learning_rate": 1.3502538071065992e-05,
241
+ "loss": 0.0055,
242
+ "step": 320
243
+ },
244
+ {
245
+ "epoch": 1.6751269035532994,
246
+ "grad_norm": 0.022236082702875137,
247
+ "learning_rate": 1.3299492385786802e-05,
248
+ "loss": 0.0251,
249
+ "step": 330
250
+ },
251
+ {
252
+ "epoch": 1.7258883248730963,
253
+ "grad_norm": 0.002686608349904418,
254
+ "learning_rate": 1.3096446700507614e-05,
255
+ "loss": 0.0089,
256
+ "step": 340
257
+ },
258
+ {
259
+ "epoch": 1.7766497461928934,
260
+ "grad_norm": 0.48235443234443665,
261
+ "learning_rate": 1.2893401015228428e-05,
262
+ "loss": 0.067,
263
+ "step": 350
264
+ },
265
+ {
266
+ "epoch": 1.8274111675126905,
267
+ "grad_norm": 0.0026509715244174004,
268
+ "learning_rate": 1.269035532994924e-05,
269
+ "loss": 0.0734,
270
+ "step": 360
271
+ },
272
+ {
273
+ "epoch": 1.8781725888324874,
274
+ "grad_norm": 0.0006726986030116677,
275
+ "learning_rate": 1.2487309644670052e-05,
276
+ "loss": 0.1744,
277
+ "step": 370
278
+ },
279
+ {
280
+ "epoch": 1.9289340101522843,
281
+ "grad_norm": 0.00010891614510910586,
282
+ "learning_rate": 1.2284263959390864e-05,
283
+ "loss": 0.0148,
284
+ "step": 380
285
+ },
286
+ {
287
+ "epoch": 1.9796954314720812,
288
+ "grad_norm": 0.12294577062129974,
289
+ "learning_rate": 1.2081218274111678e-05,
290
+ "loss": 0.1309,
291
+ "step": 390
292
+ },
293
+ {
294
+ "epoch": 2.0,
295
+ "eval_accuracy": 0.985663082437276,
296
+ "eval_loss": 0.03369523584842682,
297
+ "eval_runtime": 17.3849,
298
+ "eval_samples_per_second": 16.048,
299
+ "eval_steps_per_second": 2.013,
300
+ "step": 394
301
+ },
302
+ {
303
+ "epoch": 2.030456852791878,
304
+ "grad_norm": 7.998709201812744,
305
+ "learning_rate": 1.1878172588832488e-05,
306
+ "loss": 0.2421,
307
+ "step": 400
308
+ },
309
+ {
310
+ "epoch": 2.081218274111675,
311
+ "grad_norm": 24.342329025268555,
312
+ "learning_rate": 1.16751269035533e-05,
313
+ "loss": 0.2092,
314
+ "step": 410
315
+ },
316
+ {
317
+ "epoch": 2.1319796954314723,
318
+ "grad_norm": 0.0385918989777565,
319
+ "learning_rate": 1.1472081218274113e-05,
320
+ "loss": 0.1426,
321
+ "step": 420
322
+ },
323
+ {
324
+ "epoch": 2.182741116751269,
325
+ "grad_norm": 0.007073443848639727,
326
+ "learning_rate": 1.1269035532994925e-05,
327
+ "loss": 0.1668,
328
+ "step": 430
329
+ },
330
+ {
331
+ "epoch": 2.233502538071066,
332
+ "grad_norm": 0.11695539951324463,
333
+ "learning_rate": 1.1065989847715737e-05,
334
+ "loss": 0.0358,
335
+ "step": 440
336
+ },
337
+ {
338
+ "epoch": 2.284263959390863,
339
+ "grad_norm": 10.85909652709961,
340
+ "learning_rate": 1.0862944162436548e-05,
341
+ "loss": 0.0129,
342
+ "step": 450
343
+ },
344
+ {
345
+ "epoch": 2.33502538071066,
346
+ "grad_norm": 83.56989288330078,
347
+ "learning_rate": 1.0659898477157361e-05,
348
+ "loss": 0.3374,
349
+ "step": 460
350
+ },
351
+ {
352
+ "epoch": 2.3857868020304567,
353
+ "grad_norm": 0.001134849968366325,
354
+ "learning_rate": 1.0456852791878173e-05,
355
+ "loss": 0.2406,
356
+ "step": 470
357
+ },
358
+ {
359
+ "epoch": 2.436548223350254,
360
+ "grad_norm": 0.303134560585022,
361
+ "learning_rate": 1.0253807106598985e-05,
362
+ "loss": 0.0009,
363
+ "step": 480
364
+ },
365
+ {
366
+ "epoch": 2.487309644670051,
367
+ "grad_norm": 1.2032800912857056,
368
+ "learning_rate": 1.0050761421319797e-05,
369
+ "loss": 0.0209,
370
+ "step": 490
371
+ },
372
+ {
373
+ "epoch": 2.5380710659898478,
374
+ "grad_norm": 0.6638129949569702,
375
+ "learning_rate": 9.84771573604061e-06,
376
+ "loss": 0.0677,
377
+ "step": 500
378
+ },
379
+ {
380
+ "epoch": 2.5888324873096447,
381
+ "grad_norm": 0.6139811277389526,
382
+ "learning_rate": 9.644670050761421e-06,
383
+ "loss": 0.2713,
384
+ "step": 510
385
+ },
386
+ {
387
+ "epoch": 2.6395939086294415,
388
+ "grad_norm": 0.042041704058647156,
389
+ "learning_rate": 9.441624365482235e-06,
390
+ "loss": 0.0177,
391
+ "step": 520
392
+ },
393
+ {
394
+ "epoch": 2.6903553299492384,
395
+ "grad_norm": 3578.884765625,
396
+ "learning_rate": 9.238578680203047e-06,
397
+ "loss": 0.1765,
398
+ "step": 530
399
+ },
400
+ {
401
+ "epoch": 2.7411167512690353,
402
+ "grad_norm": 0.800250232219696,
403
+ "learning_rate": 9.035532994923859e-06,
404
+ "loss": 0.0518,
405
+ "step": 540
406
+ },
407
+ {
408
+ "epoch": 2.7918781725888326,
409
+ "grad_norm": 0.12552772462368011,
410
+ "learning_rate": 8.832487309644671e-06,
411
+ "loss": 0.1098,
412
+ "step": 550
413
+ },
414
+ {
415
+ "epoch": 2.8426395939086295,
416
+ "grad_norm": 0.003844466060400009,
417
+ "learning_rate": 8.629441624365483e-06,
418
+ "loss": 0.1961,
419
+ "step": 560
420
+ },
421
+ {
422
+ "epoch": 2.8934010152284264,
423
+ "grad_norm": 0.0007168629672378302,
424
+ "learning_rate": 8.426395939086295e-06,
425
+ "loss": 0.0309,
426
+ "step": 570
427
+ },
428
+ {
429
+ "epoch": 2.9441624365482233,
430
+ "grad_norm": 0.20322203636169434,
431
+ "learning_rate": 8.223350253807107e-06,
432
+ "loss": 0.0921,
433
+ "step": 580
434
+ },
435
+ {
436
+ "epoch": 2.99492385786802,
437
+ "grad_norm": 0.010694732889533043,
438
+ "learning_rate": 8.020304568527919e-06,
439
+ "loss": 0.3207,
440
+ "step": 590
441
+ },
442
+ {
443
+ "epoch": 3.0,
444
+ "eval_accuracy": 0.978494623655914,
445
+ "eval_loss": 0.12610773742198944,
446
+ "eval_runtime": 16.6863,
447
+ "eval_samples_per_second": 16.72,
448
+ "eval_steps_per_second": 2.098,
449
+ "step": 591
450
+ },
451
+ {
452
+ "epoch": 3.045685279187817,
453
+ "grad_norm": 0.023461902514100075,
454
+ "learning_rate": 7.817258883248731e-06,
455
+ "loss": 0.2357,
456
+ "step": 600
457
+ },
458
+ {
459
+ "epoch": 3.0964467005076144,
460
+ "grad_norm": 0.19518044590950012,
461
+ "learning_rate": 7.614213197969543e-06,
462
+ "loss": 0.0022,
463
+ "step": 610
464
+ },
465
+ {
466
+ "epoch": 3.1472081218274113,
467
+ "grad_norm": 0.00014897708024363965,
468
+ "learning_rate": 7.411167512690356e-06,
469
+ "loss": 0.0005,
470
+ "step": 620
471
+ },
472
+ {
473
+ "epoch": 3.197969543147208,
474
+ "grad_norm": 294.190673828125,
475
+ "learning_rate": 7.208121827411169e-06,
476
+ "loss": 0.0252,
477
+ "step": 630
478
+ },
479
+ {
480
+ "epoch": 3.248730964467005,
481
+ "grad_norm": 0.00014258331793826073,
482
+ "learning_rate": 7.0050761421319806e-06,
483
+ "loss": 0.0231,
484
+ "step": 640
485
+ },
486
+ {
487
+ "epoch": 3.299492385786802,
488
+ "grad_norm": 0.006024135742336512,
489
+ "learning_rate": 6.8020304568527926e-06,
490
+ "loss": 0.1952,
491
+ "step": 650
492
+ },
493
+ {
494
+ "epoch": 3.350253807106599,
495
+ "grad_norm": 215.45094299316406,
496
+ "learning_rate": 6.5989847715736045e-06,
497
+ "loss": 0.175,
498
+ "step": 660
499
+ },
500
+ {
501
+ "epoch": 3.401015228426396,
502
+ "grad_norm": 0.15660379827022552,
503
+ "learning_rate": 6.395939086294417e-06,
504
+ "loss": 0.113,
505
+ "step": 670
506
+ },
507
+ {
508
+ "epoch": 3.451776649746193,
509
+ "grad_norm": 0.0003182363579981029,
510
+ "learning_rate": 6.1928934010152285e-06,
511
+ "loss": 0.3483,
512
+ "step": 680
513
+ },
514
+ {
515
+ "epoch": 3.50253807106599,
516
+ "grad_norm": 0.0007056689355522394,
517
+ "learning_rate": 5.989847715736041e-06,
518
+ "loss": 0.2458,
519
+ "step": 690
520
+ },
521
+ {
522
+ "epoch": 3.553299492385787,
523
+ "grad_norm": 313.6507263183594,
524
+ "learning_rate": 5.7868020304568525e-06,
525
+ "loss": 0.1121,
526
+ "step": 700
527
+ },
528
+ {
529
+ "epoch": 3.6040609137055837,
530
+ "grad_norm": 0.08964172750711441,
531
+ "learning_rate": 5.583756345177665e-06,
532
+ "loss": 0.0019,
533
+ "step": 710
534
+ },
535
+ {
536
+ "epoch": 3.6548223350253806,
537
+ "grad_norm": 0.05380258709192276,
538
+ "learning_rate": 5.380710659898477e-06,
539
+ "loss": 0.0163,
540
+ "step": 720
541
+ },
542
+ {
543
+ "epoch": 3.7055837563451774,
544
+ "grad_norm": 261.12548828125,
545
+ "learning_rate": 5.17766497461929e-06,
546
+ "loss": 0.1856,
547
+ "step": 730
548
+ },
549
+ {
550
+ "epoch": 3.7563451776649748,
551
+ "grad_norm": 0.021219402551651,
552
+ "learning_rate": 4.974619289340102e-06,
553
+ "loss": 0.1741,
554
+ "step": 740
555
+ },
556
+ {
557
+ "epoch": 3.8071065989847717,
558
+ "grad_norm": 0.006986899767071009,
559
+ "learning_rate": 4.771573604060914e-06,
560
+ "loss": 0.2666,
561
+ "step": 750
562
+ },
563
+ {
564
+ "epoch": 3.8578680203045685,
565
+ "grad_norm": 0.3382449746131897,
566
+ "learning_rate": 4.568527918781726e-06,
567
+ "loss": 0.0107,
568
+ "step": 760
569
+ },
570
+ {
571
+ "epoch": 3.9086294416243654,
572
+ "grad_norm": 6.089676026022062e-05,
573
+ "learning_rate": 4.365482233502538e-06,
574
+ "loss": 0.1188,
575
+ "step": 770
576
+ },
577
+ {
578
+ "epoch": 3.9593908629441623,
579
+ "grad_norm": 0.06970903277397156,
580
+ "learning_rate": 4.162436548223351e-06,
581
+ "loss": 0.2517,
582
+ "step": 780
583
+ },
584
+ {
585
+ "epoch": 4.0,
586
+ "eval_accuracy": 0.992831541218638,
587
+ "eval_loss": 0.04906105622649193,
588
+ "eval_runtime": 17.0919,
589
+ "eval_samples_per_second": 16.323,
590
+ "eval_steps_per_second": 2.048,
591
+ "step": 788
592
+ },
593
+ {
594
+ "epoch": 4.01015228426396,
595
+ "grad_norm": 0.7618089914321899,
596
+ "learning_rate": 3.959390862944163e-06,
597
+ "loss": 0.1249,
598
+ "step": 790
599
+ },
600
+ {
601
+ "epoch": 4.060913705583756,
602
+ "grad_norm": 0.11002158373594284,
603
+ "learning_rate": 3.756345177664975e-06,
604
+ "loss": 0.0001,
605
+ "step": 800
606
+ },
607
+ {
608
+ "epoch": 4.111675126903553,
609
+ "grad_norm": 0.009345640428364277,
610
+ "learning_rate": 3.5532994923857873e-06,
611
+ "loss": 0.0,
612
+ "step": 810
613
+ },
614
+ {
615
+ "epoch": 4.16243654822335,
616
+ "grad_norm": 0.0008610596996732056,
617
+ "learning_rate": 3.3502538071065993e-06,
618
+ "loss": 0.1807,
619
+ "step": 820
620
+ },
621
+ {
622
+ "epoch": 4.213197969543147,
623
+ "grad_norm": 0.0372774712741375,
624
+ "learning_rate": 3.1472081218274113e-06,
625
+ "loss": 0.0744,
626
+ "step": 830
627
+ },
628
+ {
629
+ "epoch": 4.2639593908629445,
630
+ "grad_norm": 0.026972278952598572,
631
+ "learning_rate": 2.9441624365482237e-06,
632
+ "loss": 0.0216,
633
+ "step": 840
634
+ },
635
+ {
636
+ "epoch": 4.314720812182741,
637
+ "grad_norm": 0.14636941254138947,
638
+ "learning_rate": 2.7411167512690357e-06,
639
+ "loss": 0.0692,
640
+ "step": 850
641
+ },
642
+ {
643
+ "epoch": 4.365482233502538,
644
+ "grad_norm": 225.23846435546875,
645
+ "learning_rate": 2.5380710659898476e-06,
646
+ "loss": 0.2541,
647
+ "step": 860
648
+ },
649
+ {
650
+ "epoch": 4.416243654822335,
651
+ "grad_norm": 0.1878076195716858,
652
+ "learning_rate": 2.33502538071066e-06,
653
+ "loss": 0.0001,
654
+ "step": 870
655
+ },
656
+ {
657
+ "epoch": 4.467005076142132,
658
+ "grad_norm": 0.011137721128761768,
659
+ "learning_rate": 2.1319796954314725e-06,
660
+ "loss": 0.0,
661
+ "step": 880
662
+ },
663
+ {
664
+ "epoch": 4.517766497461929,
665
+ "grad_norm": 8.737370808376e-05,
666
+ "learning_rate": 1.9289340101522844e-06,
667
+ "loss": 0.0004,
668
+ "step": 890
669
+ },
670
+ {
671
+ "epoch": 4.568527918781726,
672
+ "grad_norm": 0.0015903498278930783,
673
+ "learning_rate": 1.7258883248730964e-06,
674
+ "loss": 0.2335,
675
+ "step": 900
676
+ },
677
+ {
678
+ "epoch": 4.619289340101523,
679
+ "grad_norm": 0.02763758972287178,
680
+ "learning_rate": 1.5228426395939088e-06,
681
+ "loss": 0.0003,
682
+ "step": 910
683
+ },
684
+ {
685
+ "epoch": 4.67005076142132,
686
+ "grad_norm": 0.4715072214603424,
687
+ "learning_rate": 1.319796954314721e-06,
688
+ "loss": 0.2327,
689
+ "step": 920
690
+ },
691
+ {
692
+ "epoch": 4.720812182741117,
693
+ "grad_norm": 0.11949143558740616,
694
+ "learning_rate": 1.116751269035533e-06,
695
+ "loss": 0.0734,
696
+ "step": 930
697
+ },
698
+ {
699
+ "epoch": 4.771573604060913,
700
+ "grad_norm": 362.79876708984375,
701
+ "learning_rate": 9.137055837563452e-07,
702
+ "loss": 0.0735,
703
+ "step": 940
704
+ },
705
+ {
706
+ "epoch": 4.822335025380711,
707
+ "grad_norm": 0.0028335938695818186,
708
+ "learning_rate": 7.106598984771574e-07,
709
+ "loss": 0.2299,
710
+ "step": 950
711
+ },
712
+ {
713
+ "epoch": 4.873096446700508,
714
+ "grad_norm": 0.015159722417593002,
715
+ "learning_rate": 5.076142131979696e-07,
716
+ "loss": 0.0003,
717
+ "step": 960
718
+ },
719
+ {
720
+ "epoch": 4.9238578680203045,
721
+ "grad_norm": 0.07444437593221664,
722
+ "learning_rate": 3.0456852791878176e-07,
723
+ "loss": 0.0,
724
+ "step": 970
725
+ },
726
+ {
727
+ "epoch": 4.974619289340102,
728
+ "grad_norm": 0.0017945471918210387,
729
+ "learning_rate": 1.0152284263959391e-07,
730
+ "loss": 0.1549,
731
+ "step": 980
732
+ },
733
+ {
734
+ "epoch": 5.0,
735
+ "eval_accuracy": 0.996415770609319,
736
+ "eval_loss": 0.04717279225587845,
737
+ "eval_runtime": 17.0835,
738
+ "eval_samples_per_second": 16.332,
739
+ "eval_steps_per_second": 2.049,
740
+ "step": 985
741
+ },
742
+ {
743
+ "epoch": 5.0,
744
+ "step": 985,
745
+ "total_flos": 1.814619221089321e+18,
746
+ "train_loss": 0.15722184067514908,
747
+ "train_runtime": 1696.7849,
748
+ "train_samples_per_second": 4.644,
749
+ "train_steps_per_second": 0.581
750
+ }
751
+ ],
752
+ "logging_steps": 10,
753
+ "max_steps": 985,
754
+ "num_input_tokens_seen": 0,
755
+ "num_train_epochs": 5,
756
+ "save_steps": 500,
757
+ "stateful_callbacks": {
758
+ "TrainerControl": {
759
+ "args": {
760
+ "should_epoch_stop": false,
761
+ "should_evaluate": false,
762
+ "should_log": false,
763
+ "should_save": true,
764
+ "should_training_stop": true
765
+ },
766
+ "attributes": {}
767
+ }
768
+ },
769
+ "total_flos": 1.814619221089321e+18,
770
+ "train_batch_size": 8,
771
+ "trial_name": null,
772
+ "trial_params": null
773
+ }