End of training
Browse files- all_results.json +10 -10
- eval_results.json +5 -5
- runs/Jul03_03-14-06_db0a38fd66c3/events.out.tfevents.1719976718.db0a38fd66c3.3870.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +171 -332
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
-
"total_flos": 1.
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 1.
|
13 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9691358024691358,
|
4 |
+
"eval_loss": 0.08843858540058136,
|
5 |
+
"eval_runtime": 4.5137,
|
6 |
+
"eval_samples_per_second": 358.903,
|
7 |
+
"eval_steps_per_second": 11.299,
|
8 |
+
"total_flos": 1.0874439903456461e+18,
|
9 |
+
"train_loss": 0.41079250579340415,
|
10 |
+
"train_runtime": 248.8888,
|
11 |
+
"train_samples_per_second": 175.741,
|
12 |
+
"train_steps_per_second": 1.374
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.9691358024691358,
|
4 |
+
"eval_loss": 0.08843858540058136,
|
5 |
+
"eval_runtime": 4.5137,
|
6 |
+
"eval_samples_per_second": 358.903,
|
7 |
+
"eval_steps_per_second": 11.299
|
8 |
}
|
runs/Jul03_03-14-06_db0a38fd66c3/events.out.tfevents.1719976718.db0a38fd66c3.3870.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1855b093188c0e5e244cb1962cd00400154d241fc684e93b16bd2bcaab298ed6
|
3 |
+
size 411
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"total_flos": 1.
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 1.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"total_flos": 1.0874439903456461e+18,
|
4 |
+
"train_loss": 0.41079250579340415,
|
5 |
+
"train_runtime": 248.8888,
|
6 |
+
"train_samples_per_second": 175.741,
|
7 |
+
"train_steps_per_second": 1.374
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,451 +1,290 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
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.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate":
|
15 |
-
"loss": 2.
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate":
|
22 |
-
"loss":
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
-
"epoch": 0.
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate":
|
29 |
-
"loss": 1.
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
-
"epoch": 0.
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss":
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
-
"epoch": 0.
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate": 4.
|
43 |
-
"loss": 0.
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
-
"epoch": 0.
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate": 4.
|
50 |
-
"loss": 0.
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
-
"epoch": 0.
|
55 |
-
"grad_norm":
|
56 |
-
"learning_rate": 4.
|
57 |
-
"loss": 0.
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
-
"epoch": 0.
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate": 4.
|
64 |
-
"loss": 0.
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
-
"epoch": 0.
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate": 4.
|
71 |
-
"loss": 0.
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
-
"epoch": 0.
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate":
|
78 |
-
"loss": 0.
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
-
"epoch": 0.
|
83 |
-
"grad_norm":
|
84 |
-
"learning_rate":
|
85 |
-
"loss": 0.
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
-
"epoch": 0
|
90 |
-
"
|
91 |
-
"
|
92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
93 |
"step": 120
|
94 |
},
|
95 |
{
|
96 |
-
"epoch":
|
97 |
-
"grad_norm":
|
98 |
-
"learning_rate":
|
99 |
-
"loss": 0.
|
100 |
"step": 130
|
101 |
},
|
102 |
{
|
103 |
-
"epoch":
|
104 |
-
"grad_norm":
|
105 |
-
"learning_rate":
|
106 |
-
"loss": 0.
|
107 |
"step": 140
|
108 |
},
|
109 |
{
|
110 |
-
"epoch":
|
111 |
-
"grad_norm":
|
112 |
-
"learning_rate":
|
113 |
-
"loss": 0.
|
114 |
"step": 150
|
115 |
},
|
116 |
{
|
117 |
-
"epoch":
|
118 |
-
"grad_norm":
|
119 |
-
"learning_rate":
|
120 |
-
"loss": 0.
|
121 |
"step": 160
|
122 |
},
|
123 |
{
|
124 |
-
"epoch":
|
125 |
-
"grad_norm":
|
126 |
-
"learning_rate":
|
127 |
-
"loss": 0.
|
128 |
"step": 170
|
129 |
},
|
130 |
{
|
131 |
-
"epoch":
|
132 |
-
"grad_norm":
|
133 |
-
"learning_rate":
|
134 |
-
"loss": 0.
|
135 |
"step": 180
|
136 |
},
|
137 |
{
|
138 |
-
"epoch": 1.
|
139 |
-
"grad_norm":
|
140 |
-
"learning_rate":
|
141 |
-
"loss": 0.
|
142 |
-
"step": 190
|
143 |
-
},
|
144 |
-
{
|
145 |
-
"epoch": 1.0,
|
146 |
-
"eval_accuracy": 0.9622222222222222,
|
147 |
-
"eval_loss": 0.10585666447877884,
|
148 |
-
"eval_runtime": 6.6908,
|
149 |
-
"eval_samples_per_second": 403.536,
|
150 |
-
"eval_steps_per_second": 12.704,
|
151 |
"step": 190
|
152 |
},
|
153 |
{
|
154 |
-
"epoch": 1.
|
155 |
-
"grad_norm":
|
156 |
-
"learning_rate":
|
157 |
-
"loss": 0.
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
-
"epoch": 1.
|
162 |
-
"grad_norm":
|
163 |
-
"learning_rate":
|
164 |
-
"loss": 0.
|
165 |
"step": 210
|
166 |
},
|
167 |
{
|
168 |
-
"epoch": 1.
|
169 |
-
"grad_norm":
|
170 |
-
"learning_rate":
|
171 |
-
"loss": 0.
|
172 |
"step": 220
|
173 |
},
|
174 |
{
|
175 |
-
"epoch":
|
176 |
-
"
|
177 |
-
"
|
178 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
179 |
"step": 230
|
180 |
},
|
181 |
{
|
182 |
-
"epoch":
|
183 |
-
"grad_norm":
|
184 |
-
"learning_rate":
|
185 |
-
"loss": 0.
|
186 |
"step": 240
|
187 |
},
|
188 |
{
|
189 |
-
"epoch":
|
190 |
-
"grad_norm":
|
191 |
-
"learning_rate":
|
192 |
-
"loss": 0.
|
193 |
"step": 250
|
194 |
},
|
195 |
{
|
196 |
-
"epoch":
|
197 |
-
"grad_norm":
|
198 |
-
"learning_rate":
|
199 |
-
"loss": 0.
|
200 |
"step": 260
|
201 |
},
|
202 |
{
|
203 |
-
"epoch":
|
204 |
-
"grad_norm":
|
205 |
-
"learning_rate":
|
206 |
-
"loss": 0.
|
207 |
"step": 270
|
208 |
},
|
209 |
{
|
210 |
-
"epoch":
|
211 |
-
"grad_norm":
|
212 |
-
"learning_rate":
|
213 |
-
"loss": 0.
|
214 |
"step": 280
|
215 |
},
|
216 |
{
|
217 |
-
"epoch":
|
218 |
-
"grad_norm":
|
219 |
-
"learning_rate":
|
220 |
-
"loss": 0.
|
221 |
"step": 290
|
222 |
},
|
223 |
{
|
224 |
-
"epoch":
|
225 |
-
"grad_norm":
|
226 |
-
"learning_rate":
|
227 |
-
"loss": 0.
|
228 |
"step": 300
|
229 |
},
|
230 |
{
|
231 |
-
"epoch":
|
232 |
-
"grad_norm":
|
233 |
-
"learning_rate":
|
234 |
-
"loss": 0.
|
235 |
"step": 310
|
236 |
},
|
237 |
{
|
238 |
-
"epoch":
|
239 |
-
"grad_norm":
|
240 |
-
"learning_rate":
|
241 |
-
"loss": 0.
|
242 |
"step": 320
|
243 |
},
|
244 |
{
|
245 |
-
"epoch":
|
246 |
-
"grad_norm":
|
247 |
-
"learning_rate":
|
248 |
-
"loss": 0.
|
249 |
"step": 330
|
250 |
},
|
251 |
{
|
252 |
-
"epoch":
|
253 |
-
"grad_norm":
|
254 |
-
"learning_rate":
|
255 |
-
"loss": 0.
|
256 |
"step": 340
|
257 |
},
|
258 |
-
{
|
259 |
-
"epoch": 1.8421052631578947,
|
260 |
-
"grad_norm": 8.70970344543457,
|
261 |
-
"learning_rate": 2.1442495126705653e-05,
|
262 |
-
"loss": 0.2318,
|
263 |
-
"step": 350
|
264 |
-
},
|
265 |
-
{
|
266 |
-
"epoch": 1.8947368421052633,
|
267 |
-
"grad_norm": 7.852567672729492,
|
268 |
-
"learning_rate": 2.046783625730994e-05,
|
269 |
-
"loss": 0.1734,
|
270 |
-
"step": 360
|
271 |
-
},
|
272 |
-
{
|
273 |
-
"epoch": 1.9473684210526314,
|
274 |
-
"grad_norm": 4.875802993774414,
|
275 |
-
"learning_rate": 1.949317738791423e-05,
|
276 |
-
"loss": 0.1841,
|
277 |
-
"step": 370
|
278 |
-
},
|
279 |
-
{
|
280 |
-
"epoch": 2.0,
|
281 |
-
"grad_norm": 15.681636810302734,
|
282 |
-
"learning_rate": 1.8518518518518518e-05,
|
283 |
-
"loss": 0.1941,
|
284 |
-
"step": 380
|
285 |
-
},
|
286 |
-
{
|
287 |
-
"epoch": 2.0,
|
288 |
-
"eval_accuracy": 0.9714814814814815,
|
289 |
-
"eval_loss": 0.07911109179258347,
|
290 |
-
"eval_runtime": 6.6371,
|
291 |
-
"eval_samples_per_second": 406.804,
|
292 |
-
"eval_steps_per_second": 12.807,
|
293 |
-
"step": 380
|
294 |
-
},
|
295 |
-
{
|
296 |
-
"epoch": 2.0526315789473686,
|
297 |
-
"grad_norm": 13.168564796447754,
|
298 |
-
"learning_rate": 1.7543859649122806e-05,
|
299 |
-
"loss": 0.213,
|
300 |
-
"step": 390
|
301 |
-
},
|
302 |
-
{
|
303 |
-
"epoch": 2.1052631578947367,
|
304 |
-
"grad_norm": 6.1030802726745605,
|
305 |
-
"learning_rate": 1.6569200779727097e-05,
|
306 |
-
"loss": 0.1842,
|
307 |
-
"step": 400
|
308 |
-
},
|
309 |
-
{
|
310 |
-
"epoch": 2.1578947368421053,
|
311 |
-
"grad_norm": 4.817559242248535,
|
312 |
-
"learning_rate": 1.5594541910331384e-05,
|
313 |
-
"loss": 0.1902,
|
314 |
-
"step": 410
|
315 |
-
},
|
316 |
-
{
|
317 |
-
"epoch": 2.2105263157894735,
|
318 |
-
"grad_norm": 7.957576751708984,
|
319 |
-
"learning_rate": 1.4619883040935673e-05,
|
320 |
-
"loss": 0.1605,
|
321 |
-
"step": 420
|
322 |
-
},
|
323 |
-
{
|
324 |
-
"epoch": 2.263157894736842,
|
325 |
-
"grad_norm": 9.295196533203125,
|
326 |
-
"learning_rate": 1.364522417153996e-05,
|
327 |
-
"loss": 0.1542,
|
328 |
-
"step": 430
|
329 |
-
},
|
330 |
-
{
|
331 |
-
"epoch": 2.3157894736842106,
|
332 |
-
"grad_norm": 5.581339359283447,
|
333 |
-
"learning_rate": 1.267056530214425e-05,
|
334 |
-
"loss": 0.1878,
|
335 |
-
"step": 440
|
336 |
-
},
|
337 |
-
{
|
338 |
-
"epoch": 2.3684210526315788,
|
339 |
-
"grad_norm": 7.6587347984313965,
|
340 |
-
"learning_rate": 1.1695906432748537e-05,
|
341 |
-
"loss": 0.1537,
|
342 |
-
"step": 450
|
343 |
-
},
|
344 |
-
{
|
345 |
-
"epoch": 2.4210526315789473,
|
346 |
-
"grad_norm": 4.522464752197266,
|
347 |
-
"learning_rate": 1.0721247563352826e-05,
|
348 |
-
"loss": 0.1474,
|
349 |
-
"step": 460
|
350 |
-
},
|
351 |
-
{
|
352 |
-
"epoch": 2.473684210526316,
|
353 |
-
"grad_norm": 8.627243995666504,
|
354 |
-
"learning_rate": 9.746588693957115e-06,
|
355 |
-
"loss": 0.1679,
|
356 |
-
"step": 470
|
357 |
-
},
|
358 |
-
{
|
359 |
-
"epoch": 2.526315789473684,
|
360 |
-
"grad_norm": 8.504610061645508,
|
361 |
-
"learning_rate": 8.771929824561403e-06,
|
362 |
-
"loss": 0.1582,
|
363 |
-
"step": 480
|
364 |
-
},
|
365 |
-
{
|
366 |
-
"epoch": 2.5789473684210527,
|
367 |
-
"grad_norm": 7.775728225708008,
|
368 |
-
"learning_rate": 7.797270955165692e-06,
|
369 |
-
"loss": 0.1412,
|
370 |
-
"step": 490
|
371 |
-
},
|
372 |
-
{
|
373 |
-
"epoch": 2.6315789473684212,
|
374 |
-
"grad_norm": 7.758059024810791,
|
375 |
-
"learning_rate": 6.82261208576998e-06,
|
376 |
-
"loss": 0.1704,
|
377 |
-
"step": 500
|
378 |
-
},
|
379 |
-
{
|
380 |
-
"epoch": 2.6842105263157894,
|
381 |
-
"grad_norm": 8.407027244567871,
|
382 |
-
"learning_rate": 5.8479532163742686e-06,
|
383 |
-
"loss": 0.1622,
|
384 |
-
"step": 510
|
385 |
-
},
|
386 |
-
{
|
387 |
-
"epoch": 2.736842105263158,
|
388 |
-
"grad_norm": 5.691274166107178,
|
389 |
-
"learning_rate": 4.873294346978558e-06,
|
390 |
-
"loss": 0.1503,
|
391 |
-
"step": 520
|
392 |
-
},
|
393 |
-
{
|
394 |
-
"epoch": 2.7894736842105265,
|
395 |
-
"grad_norm": 10.755898475646973,
|
396 |
-
"learning_rate": 3.898635477582846e-06,
|
397 |
-
"loss": 0.1477,
|
398 |
-
"step": 530
|
399 |
-
},
|
400 |
-
{
|
401 |
-
"epoch": 2.8421052631578947,
|
402 |
-
"grad_norm": 6.478678226470947,
|
403 |
-
"learning_rate": 2.9239766081871343e-06,
|
404 |
-
"loss": 0.1284,
|
405 |
-
"step": 540
|
406 |
-
},
|
407 |
-
{
|
408 |
-
"epoch": 2.8947368421052633,
|
409 |
-
"grad_norm": 8.324136734008789,
|
410 |
-
"learning_rate": 1.949317738791423e-06,
|
411 |
-
"loss": 0.1452,
|
412 |
-
"step": 550
|
413 |
-
},
|
414 |
-
{
|
415 |
-
"epoch": 2.9473684210526314,
|
416 |
-
"grad_norm": 10.353252410888672,
|
417 |
-
"learning_rate": 9.746588693957115e-07,
|
418 |
-
"loss": 0.1188,
|
419 |
-
"step": 560
|
420 |
-
},
|
421 |
-
{
|
422 |
-
"epoch": 3.0,
|
423 |
-
"grad_norm": 6.009654998779297,
|
424 |
-
"learning_rate": 0.0,
|
425 |
-
"loss": 0.1551,
|
426 |
-
"step": 570
|
427 |
-
},
|
428 |
{
|
429 |
"epoch": 3.0,
|
430 |
-
"eval_accuracy": 0.
|
431 |
-
"eval_loss": 0.
|
432 |
-
"eval_runtime":
|
433 |
-
"eval_samples_per_second":
|
434 |
-
"eval_steps_per_second":
|
435 |
-
"step":
|
436 |
},
|
437 |
{
|
438 |
"epoch": 3.0,
|
439 |
-
"step":
|
440 |
-
"total_flos": 1.
|
441 |
-
"train_loss": 0.
|
442 |
-
"train_runtime":
|
443 |
-
"train_samples_per_second":
|
444 |
-
"train_steps_per_second": 1.
|
445 |
}
|
446 |
],
|
447 |
"logging_steps": 10,
|
448 |
-
"max_steps":
|
449 |
"num_input_tokens_seen": 0,
|
450 |
"num_train_epochs": 3,
|
451 |
"save_steps": 500,
|
@@ -461,7 +300,7 @@
|
|
461 |
"attributes": {}
|
462 |
}
|
463 |
},
|
464 |
-
"total_flos": 1.
|
465 |
"train_batch_size": 32,
|
466 |
"trial_name": null,
|
467 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9691358024691358,
|
3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-342",
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 342,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.08771929824561403,
|
13 |
+
"grad_norm": 4.784806251525879,
|
14 |
+
"learning_rate": 1.4285714285714285e-05,
|
15 |
+
"loss": 2.2537,
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.17543859649122806,
|
20 |
+
"grad_norm": 7.325711727142334,
|
21 |
+
"learning_rate": 2.857142857142857e-05,
|
22 |
+
"loss": 1.9328,
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
+
"epoch": 0.2631578947368421,
|
27 |
+
"grad_norm": 12.205706596374512,
|
28 |
+
"learning_rate": 4.2857142857142856e-05,
|
29 |
+
"loss": 1.3033,
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
+
"epoch": 0.3508771929824561,
|
34 |
+
"grad_norm": 21.546316146850586,
|
35 |
+
"learning_rate": 4.9185667752443e-05,
|
36 |
+
"loss": 0.7435,
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
+
"epoch": 0.43859649122807015,
|
41 |
+
"grad_norm": 11.416431427001953,
|
42 |
+
"learning_rate": 4.755700325732899e-05,
|
43 |
+
"loss": 0.5723,
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 0.5263157894736842,
|
48 |
+
"grad_norm": 13.339532852172852,
|
49 |
+
"learning_rate": 4.592833876221499e-05,
|
50 |
+
"loss": 0.4688,
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
+
"epoch": 0.6140350877192983,
|
55 |
+
"grad_norm": 18.226308822631836,
|
56 |
+
"learning_rate": 4.429967426710098e-05,
|
57 |
+
"loss": 0.4296,
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
+
"epoch": 0.7017543859649122,
|
62 |
+
"grad_norm": 24.142580032348633,
|
63 |
+
"learning_rate": 4.2671009771986977e-05,
|
64 |
+
"loss": 0.3557,
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 0.7894736842105263,
|
69 |
+
"grad_norm": 18.747182846069336,
|
70 |
+
"learning_rate": 4.104234527687297e-05,
|
71 |
+
"loss": 0.3364,
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
+
"epoch": 0.8771929824561403,
|
76 |
+
"grad_norm": 9.604598999023438,
|
77 |
+
"learning_rate": 3.941368078175896e-05,
|
78 |
+
"loss": 0.379,
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
+
"epoch": 0.9649122807017544,
|
83 |
+
"grad_norm": 9.6405668258667,
|
84 |
+
"learning_rate": 3.778501628664495e-05,
|
85 |
+
"loss": 0.2967,
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
+
"epoch": 1.0,
|
90 |
+
"eval_accuracy": 0.9271604938271605,
|
91 |
+
"eval_loss": 0.20734398066997528,
|
92 |
+
"eval_runtime": 4.5343,
|
93 |
+
"eval_samples_per_second": 357.28,
|
94 |
+
"eval_steps_per_second": 11.248,
|
95 |
+
"step": 114
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 1.0526315789473684,
|
99 |
+
"grad_norm": 10.606453895568848,
|
100 |
+
"learning_rate": 3.615635179153095e-05,
|
101 |
+
"loss": 0.2506,
|
102 |
"step": 120
|
103 |
},
|
104 |
{
|
105 |
+
"epoch": 1.1403508771929824,
|
106 |
+
"grad_norm": 13.219770431518555,
|
107 |
+
"learning_rate": 3.452768729641694e-05,
|
108 |
+
"loss": 0.2646,
|
109 |
"step": 130
|
110 |
},
|
111 |
{
|
112 |
+
"epoch": 1.2280701754385965,
|
113 |
+
"grad_norm": 14.424461364746094,
|
114 |
+
"learning_rate": 3.289902280130293e-05,
|
115 |
+
"loss": 0.3161,
|
116 |
"step": 140
|
117 |
},
|
118 |
{
|
119 |
+
"epoch": 1.3157894736842106,
|
120 |
+
"grad_norm": 16.468141555786133,
|
121 |
+
"learning_rate": 3.127035830618892e-05,
|
122 |
+
"loss": 0.2296,
|
123 |
"step": 150
|
124 |
},
|
125 |
{
|
126 |
+
"epoch": 1.4035087719298245,
|
127 |
+
"grad_norm": 18.474872589111328,
|
128 |
+
"learning_rate": 2.9641693811074923e-05,
|
129 |
+
"loss": 0.252,
|
130 |
"step": 160
|
131 |
},
|
132 |
{
|
133 |
+
"epoch": 1.4912280701754386,
|
134 |
+
"grad_norm": 8.512451171875,
|
135 |
+
"learning_rate": 2.8013029315960915e-05,
|
136 |
+
"loss": 0.2526,
|
137 |
"step": 170
|
138 |
},
|
139 |
{
|
140 |
+
"epoch": 1.5789473684210527,
|
141 |
+
"grad_norm": 10.639461517333984,
|
142 |
+
"learning_rate": 2.6384364820846906e-05,
|
143 |
+
"loss": 0.2218,
|
144 |
"step": 180
|
145 |
},
|
146 |
{
|
147 |
+
"epoch": 1.6666666666666665,
|
148 |
+
"grad_norm": 10.84925651550293,
|
149 |
+
"learning_rate": 2.47557003257329e-05,
|
150 |
+
"loss": 0.2246,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
151 |
"step": 190
|
152 |
},
|
153 |
{
|
154 |
+
"epoch": 1.7543859649122808,
|
155 |
+
"grad_norm": 7.055878162384033,
|
156 |
+
"learning_rate": 2.3127035830618892e-05,
|
157 |
+
"loss": 0.2181,
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
+
"epoch": 1.8421052631578947,
|
162 |
+
"grad_norm": 10.151970863342285,
|
163 |
+
"learning_rate": 2.149837133550489e-05,
|
164 |
+
"loss": 0.2336,
|
165 |
"step": 210
|
166 |
},
|
167 |
{
|
168 |
+
"epoch": 1.9298245614035088,
|
169 |
+
"grad_norm": 10.90556812286377,
|
170 |
+
"learning_rate": 1.986970684039088e-05,
|
171 |
+
"loss": 0.2175,
|
172 |
"step": 220
|
173 |
},
|
174 |
{
|
175 |
+
"epoch": 2.0,
|
176 |
+
"eval_accuracy": 0.9617283950617284,
|
177 |
+
"eval_loss": 0.1142377108335495,
|
178 |
+
"eval_runtime": 4.4649,
|
179 |
+
"eval_samples_per_second": 362.828,
|
180 |
+
"eval_steps_per_second": 11.422,
|
181 |
+
"step": 228
|
182 |
+
},
|
183 |
+
{
|
184 |
+
"epoch": 2.017543859649123,
|
185 |
+
"grad_norm": 7.411689281463623,
|
186 |
+
"learning_rate": 1.8241042345276872e-05,
|
187 |
+
"loss": 0.2493,
|
188 |
"step": 230
|
189 |
},
|
190 |
{
|
191 |
+
"epoch": 2.1052631578947367,
|
192 |
+
"grad_norm": 6.981323719024658,
|
193 |
+
"learning_rate": 1.6612377850162867e-05,
|
194 |
+
"loss": 0.1832,
|
195 |
"step": 240
|
196 |
},
|
197 |
{
|
198 |
+
"epoch": 2.192982456140351,
|
199 |
+
"grad_norm": 17.770326614379883,
|
200 |
+
"learning_rate": 1.4983713355048862e-05,
|
201 |
+
"loss": 0.1862,
|
202 |
"step": 250
|
203 |
},
|
204 |
{
|
205 |
+
"epoch": 2.280701754385965,
|
206 |
+
"grad_norm": 14.688613891601562,
|
207 |
+
"learning_rate": 1.3355048859934855e-05,
|
208 |
+
"loss": 0.1725,
|
209 |
"step": 260
|
210 |
},
|
211 |
{
|
212 |
+
"epoch": 2.3684210526315788,
|
213 |
+
"grad_norm": 9.546713829040527,
|
214 |
+
"learning_rate": 1.1726384364820847e-05,
|
215 |
+
"loss": 0.2187,
|
216 |
"step": 270
|
217 |
},
|
218 |
{
|
219 |
+
"epoch": 2.456140350877193,
|
220 |
+
"grad_norm": 13.742387771606445,
|
221 |
+
"learning_rate": 1.009771986970684e-05,
|
222 |
+
"loss": 0.211,
|
223 |
"step": 280
|
224 |
},
|
225 |
{
|
226 |
+
"epoch": 2.543859649122807,
|
227 |
+
"grad_norm": 10.309969902038574,
|
228 |
+
"learning_rate": 8.469055374592833e-06,
|
229 |
+
"loss": 0.1776,
|
230 |
"step": 290
|
231 |
},
|
232 |
{
|
233 |
+
"epoch": 2.6315789473684212,
|
234 |
+
"grad_norm": 11.343367576599121,
|
235 |
+
"learning_rate": 6.840390879478828e-06,
|
236 |
+
"loss": 0.1828,
|
237 |
"step": 300
|
238 |
},
|
239 |
{
|
240 |
+
"epoch": 2.719298245614035,
|
241 |
+
"grad_norm": 12.15509033203125,
|
242 |
+
"learning_rate": 5.211726384364822e-06,
|
243 |
+
"loss": 0.1565,
|
244 |
"step": 310
|
245 |
},
|
246 |
{
|
247 |
+
"epoch": 2.807017543859649,
|
248 |
+
"grad_norm": 13.258170127868652,
|
249 |
+
"learning_rate": 3.5830618892508147e-06,
|
250 |
+
"loss": 0.1601,
|
251 |
"step": 320
|
252 |
},
|
253 |
{
|
254 |
+
"epoch": 2.8947368421052633,
|
255 |
+
"grad_norm": 13.760292053222656,
|
256 |
+
"learning_rate": 1.9543973941368076e-06,
|
257 |
+
"loss": 0.1777,
|
258 |
"step": 330
|
259 |
},
|
260 |
{
|
261 |
+
"epoch": 2.982456140350877,
|
262 |
+
"grad_norm": 10.578232765197754,
|
263 |
+
"learning_rate": 3.2573289902280136e-07,
|
264 |
+
"loss": 0.1881,
|
265 |
"step": 340
|
266 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
267 |
{
|
268 |
"epoch": 3.0,
|
269 |
+
"eval_accuracy": 0.9691358024691358,
|
270 |
+
"eval_loss": 0.08843858540058136,
|
271 |
+
"eval_runtime": 4.4533,
|
272 |
+
"eval_samples_per_second": 363.776,
|
273 |
+
"eval_steps_per_second": 11.452,
|
274 |
+
"step": 342
|
275 |
},
|
276 |
{
|
277 |
"epoch": 3.0,
|
278 |
+
"step": 342,
|
279 |
+
"total_flos": 1.0874439903456461e+18,
|
280 |
+
"train_loss": 0.41079250579340415,
|
281 |
+
"train_runtime": 248.8888,
|
282 |
+
"train_samples_per_second": 175.741,
|
283 |
+
"train_steps_per_second": 1.374
|
284 |
}
|
285 |
],
|
286 |
"logging_steps": 10,
|
287 |
+
"max_steps": 342,
|
288 |
"num_input_tokens_seen": 0,
|
289 |
"num_train_epochs": 3,
|
290 |
"save_steps": 500,
|
|
|
300 |
"attributes": {}
|
301 |
}
|
302 |
},
|
303 |
+
"total_flos": 1.0874439903456461e+18,
|
304 |
"train_batch_size": 32,
|
305 |
"trial_name": null,
|
306 |
"trial_params": null
|