Zheng Li
commited on
End of training
Browse files
README.md
CHANGED
@@ -3,6 +3,7 @@ library_name: transformers
|
|
3 |
license: apache-2.0
|
4 |
base_model: facebook/wav2vec2-base
|
5 |
tags:
|
|
|
6 |
- generated_from_trainer
|
7 |
datasets:
|
8 |
- superb
|
@@ -33,7 +34,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
33 |
|
34 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the superb dataset.
|
35 |
It achieves the following results on the evaluation set:
|
36 |
-
- Loss: 0.
|
37 |
- Accuracy: 0.9815
|
38 |
|
39 |
## Model description
|
|
|
3 |
license: apache-2.0
|
4 |
base_model: facebook/wav2vec2-base
|
5 |
tags:
|
6 |
+
- audio-classification
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- superb
|
|
|
34 |
|
35 |
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the superb dataset.
|
36 |
It achieves the following results on the evaluation set:
|
37 |
+
- Loss: 0.1244
|
38 |
- Accuracy: 0.9815
|
39 |
|
40 |
## Model description
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 5.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
"total_flos": 3.777723239743488e+18,
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 2.
|
13 |
}
|
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
+
"eval_accuracy": 0.9814651368049426,
|
4 |
+
"eval_loss": 0.12438357621431351,
|
5 |
+
"eval_runtime": 5.5465,
|
6 |
+
"eval_samples_per_second": 1225.644,
|
7 |
+
"eval_steps_per_second": 38.403,
|
8 |
"total_flos": 3.777723239743488e+18,
|
9 |
+
"train_loss": 0.726146658611058,
|
10 |
+
"train_runtime": 635.8846,
|
11 |
+
"train_samples_per_second": 642.808,
|
12 |
+
"train_steps_per_second": 2.504
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_runtime": 5.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
+
"eval_accuracy": 0.9814651368049426,
|
4 |
+
"eval_loss": 0.12438357621431351,
|
5 |
+
"eval_runtime": 5.5465,
|
6 |
+
"eval_samples_per_second": 1225.644,
|
7 |
+
"eval_steps_per_second": 38.403
|
8 |
}
|
runs/May15_00-55-30_cs-Precision-7960-Tower/events.out.tfevents.1747285605.cs-Precision-7960-Tower.141414.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:377b4458bfcf1e26e5f48fc363b72592e31a5d1563296edc5979737602f7404c
|
3 |
+
size 411
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
"total_flos": 3.777723239743488e+18,
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 2.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 7.996245306633291,
|
3 |
"total_flos": 3.777723239743488e+18,
|
4 |
+
"train_loss": 0.726146658611058,
|
5 |
+
"train_runtime": 635.8846,
|
6 |
+
"train_samples_per_second": 642.808,
|
7 |
+
"train_steps_per_second": 2.504
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "wav2vec2-base-ft-keyword-spotting/checkpoint-
|
4 |
"epoch": 7.996245306633291,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 1592,
|
@@ -10,1197 +10,1197 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.05006257822277847,
|
13 |
-
"grad_norm": 2.
|
14 |
-
"learning_rate":
|
15 |
-
"loss": 4.
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.10012515644555695,
|
20 |
-
"grad_norm": 3.
|
21 |
-
"learning_rate":
|
22 |
-
"loss": 4.
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.15018773466833543,
|
27 |
-
"grad_norm": 3.
|
28 |
-
"learning_rate":
|
29 |
-
"loss":
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.2002503128911139,
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss": 3.
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.2503128911138924,
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate":
|
43 |
-
"loss":
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.30037546933667086,
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate":
|
50 |
-
"loss":
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.3504380475594493,
|
55 |
-
"grad_norm":
|
56 |
-
"learning_rate":
|
57 |
-
"loss": 2.
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
"epoch": 0.4005006257822278,
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate":
|
64 |
-
"loss":
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
"epoch": 0.45056320400500627,
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate":
|
71 |
-
"loss":
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
"epoch": 0.5006257822277848,
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate":
|
78 |
-
"loss":
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
"epoch": 0.5506883604505632,
|
83 |
-
"grad_norm":
|
84 |
-
"learning_rate":
|
85 |
-
"loss":
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.6007509386733417,
|
90 |
-
"grad_norm":
|
91 |
-
"learning_rate":
|
92 |
-
"loss": 1.
|
93 |
"step": 120
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.6508135168961201,
|
97 |
-
"grad_norm":
|
98 |
-
"learning_rate":
|
99 |
-
"loss": 1.
|
100 |
"step": 130
|
101 |
},
|
102 |
{
|
103 |
"epoch": 0.7008760951188986,
|
104 |
-
"grad_norm": 2.
|
105 |
-
"learning_rate":
|
106 |
-
"loss": 1.
|
107 |
"step": 140
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.7509386733416771,
|
111 |
-
"grad_norm": 1.
|
112 |
-
"learning_rate":
|
113 |
-
"loss": 1.
|
114 |
"step": 150
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.8010012515644556,
|
118 |
-
"grad_norm":
|
119 |
-
"learning_rate":
|
120 |
-
"loss": 1.
|
121 |
"step": 160
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.851063829787234,
|
125 |
-
"grad_norm":
|
126 |
-
"learning_rate":
|
127 |
-
"loss": 1.
|
128 |
"step": 170
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.9011264080100125,
|
132 |
-
"grad_norm": 1.
|
133 |
-
"learning_rate":
|
134 |
-
"loss": 1.
|
135 |
"step": 180
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.951188986232791,
|
139 |
-
"grad_norm": 2.
|
140 |
-
"learning_rate":
|
141 |
-
"loss": 1.
|
142 |
"step": 190
|
143 |
},
|
144 |
{
|
145 |
"epoch": 0.9962453066332916,
|
146 |
-
"eval_accuracy": 0.
|
147 |
-
"eval_loss": 1.
|
148 |
-
"eval_runtime":
|
149 |
-
"eval_samples_per_second":
|
150 |
-
"eval_steps_per_second": 42.
|
151 |
"step": 199
|
152 |
},
|
153 |
{
|
154 |
"epoch": 1.0050062578222778,
|
155 |
-
"grad_norm":
|
156 |
-
"learning_rate":
|
157 |
-
"loss": 1.
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
"epoch": 1.0550688360450564,
|
162 |
-
"grad_norm":
|
163 |
-
"learning_rate":
|
164 |
-
"loss": 1.
|
165 |
"step": 210
|
166 |
},
|
167 |
{
|
168 |
"epoch": 1.1051314142678348,
|
169 |
-
"grad_norm":
|
170 |
-
"learning_rate":
|
171 |
-
"loss": 1.
|
172 |
"step": 220
|
173 |
},
|
174 |
{
|
175 |
"epoch": 1.1551939924906134,
|
176 |
-
"grad_norm":
|
177 |
-
"learning_rate":
|
178 |
-
"loss": 1.
|
179 |
"step": 230
|
180 |
},
|
181 |
{
|
182 |
"epoch": 1.2052565707133918,
|
183 |
-
"grad_norm":
|
184 |
-
"learning_rate":
|
185 |
-
"loss": 1.
|
186 |
"step": 240
|
187 |
},
|
188 |
{
|
189 |
"epoch": 1.2553191489361701,
|
190 |
-
"grad_norm":
|
191 |
-
"learning_rate":
|
192 |
-
"loss": 1.
|
193 |
"step": 250
|
194 |
},
|
195 |
{
|
196 |
"epoch": 1.3053817271589487,
|
197 |
-
"grad_norm":
|
198 |
-
"learning_rate":
|
199 |
-
"loss": 1.
|
200 |
"step": 260
|
201 |
},
|
202 |
{
|
203 |
"epoch": 1.355444305381727,
|
204 |
-
"grad_norm":
|
205 |
-
"learning_rate":
|
206 |
-
"loss": 1.
|
207 |
"step": 270
|
208 |
},
|
209 |
{
|
210 |
"epoch": 1.4055068836045057,
|
211 |
-
"grad_norm":
|
212 |
-
"learning_rate":
|
213 |
-
"loss": 1.
|
214 |
"step": 280
|
215 |
},
|
216 |
{
|
217 |
"epoch": 1.455569461827284,
|
218 |
-
"grad_norm":
|
219 |
-
"learning_rate":
|
220 |
-
"loss": 1.
|
221 |
"step": 290
|
222 |
},
|
223 |
{
|
224 |
"epoch": 1.5056320400500627,
|
225 |
-
"grad_norm":
|
226 |
-
"learning_rate":
|
227 |
-
"loss": 1.
|
228 |
"step": 300
|
229 |
},
|
230 |
{
|
231 |
"epoch": 1.555694618272841,
|
232 |
-
"grad_norm":
|
233 |
-
"learning_rate":
|
234 |
-
"loss": 1.
|
235 |
"step": 310
|
236 |
},
|
237 |
{
|
238 |
"epoch": 1.6057571964956194,
|
239 |
-
"grad_norm":
|
240 |
-
"learning_rate":
|
241 |
-
"loss": 1.
|
242 |
"step": 320
|
243 |
},
|
244 |
{
|
245 |
"epoch": 1.655819774718398,
|
246 |
-
"grad_norm":
|
247 |
-
"learning_rate":
|
248 |
-
"loss": 1.
|
249 |
"step": 330
|
250 |
},
|
251 |
{
|
252 |
"epoch": 1.7058823529411766,
|
253 |
-
"grad_norm":
|
254 |
-
"learning_rate":
|
255 |
-
"loss": 1.
|
256 |
"step": 340
|
257 |
},
|
258 |
{
|
259 |
"epoch": 1.7559449311639548,
|
260 |
-
"grad_norm": 3.
|
261 |
-
"learning_rate":
|
262 |
-
"loss":
|
263 |
"step": 350
|
264 |
},
|
265 |
{
|
266 |
"epoch": 1.8060075093867334,
|
267 |
-
"grad_norm": 2.
|
268 |
-
"learning_rate":
|
269 |
-
"loss":
|
270 |
"step": 360
|
271 |
},
|
272 |
{
|
273 |
"epoch": 1.856070087609512,
|
274 |
-
"grad_norm":
|
275 |
-
"learning_rate":
|
276 |
-
"loss":
|
277 |
"step": 370
|
278 |
},
|
279 |
{
|
280 |
"epoch": 1.9061326658322904,
|
281 |
-
"grad_norm":
|
282 |
-
"learning_rate":
|
283 |
-
"loss": 0.
|
284 |
"step": 380
|
285 |
},
|
286 |
{
|
287 |
"epoch": 1.9561952440550687,
|
288 |
-
"grad_norm":
|
289 |
-
"learning_rate":
|
290 |
-
"loss":
|
291 |
"step": 390
|
292 |
},
|
293 |
{
|
294 |
"epoch": 1.9962453066332917,
|
295 |
-
"eval_accuracy": 0.
|
296 |
-
"eval_loss": 0.
|
297 |
-
"eval_runtime":
|
298 |
-
"eval_samples_per_second":
|
299 |
-
"eval_steps_per_second":
|
300 |
"step": 398
|
301 |
},
|
302 |
{
|
303 |
"epoch": 2.0100125156445556,
|
304 |
-
"grad_norm":
|
305 |
-
"learning_rate":
|
306 |
-
"loss":
|
307 |
"step": 400
|
308 |
},
|
309 |
{
|
310 |
"epoch": 2.0600750938673342,
|
311 |
-
"grad_norm":
|
312 |
-
"learning_rate":
|
313 |
-
"loss": 0.
|
314 |
"step": 410
|
315 |
},
|
316 |
{
|
317 |
"epoch": 2.110137672090113,
|
318 |
-
"grad_norm":
|
319 |
-
"learning_rate":
|
320 |
-
"loss": 0.
|
321 |
"step": 420
|
322 |
},
|
323 |
{
|
324 |
"epoch": 2.160200250312891,
|
325 |
-
"grad_norm":
|
326 |
-
"learning_rate":
|
327 |
-
"loss": 0.
|
328 |
"step": 430
|
329 |
},
|
330 |
{
|
331 |
"epoch": 2.2102628285356696,
|
332 |
-
"grad_norm":
|
333 |
-
"learning_rate":
|
334 |
-
"loss": 0.
|
335 |
"step": 440
|
336 |
},
|
337 |
{
|
338 |
"epoch": 2.260325406758448,
|
339 |
-
"grad_norm":
|
340 |
-
"learning_rate":
|
341 |
-
"loss": 0.
|
342 |
"step": 450
|
343 |
},
|
344 |
{
|
345 |
"epoch": 2.3103879849812268,
|
346 |
-
"grad_norm":
|
347 |
-
"learning_rate":
|
348 |
-
"loss": 0.
|
349 |
"step": 460
|
350 |
},
|
351 |
{
|
352 |
"epoch": 2.360450563204005,
|
353 |
-
"grad_norm":
|
354 |
-
"learning_rate":
|
355 |
-
"loss": 0.
|
356 |
"step": 470
|
357 |
},
|
358 |
{
|
359 |
"epoch": 2.4105131414267835,
|
360 |
-
"grad_norm":
|
361 |
-
"learning_rate":
|
362 |
-
"loss": 0.
|
363 |
"step": 480
|
364 |
},
|
365 |
{
|
366 |
"epoch": 2.460575719649562,
|
367 |
-
"grad_norm": 2.
|
368 |
-
"learning_rate":
|
369 |
-
"loss": 0.
|
370 |
"step": 490
|
371 |
},
|
372 |
{
|
373 |
"epoch": 2.5106382978723403,
|
374 |
-
"grad_norm":
|
375 |
-
"learning_rate":
|
376 |
-
"loss": 0.
|
377 |
"step": 500
|
378 |
},
|
379 |
{
|
380 |
"epoch": 2.560700876095119,
|
381 |
-
"grad_norm":
|
382 |
-
"learning_rate":
|
383 |
-
"loss": 0.
|
384 |
"step": 510
|
385 |
},
|
386 |
{
|
387 |
"epoch": 2.6107634543178975,
|
388 |
-
"grad_norm":
|
389 |
-
"learning_rate":
|
390 |
-
"loss": 0.
|
391 |
"step": 520
|
392 |
},
|
393 |
{
|
394 |
"epoch": 2.660826032540676,
|
395 |
-
"grad_norm": 3.
|
396 |
-
"learning_rate":
|
397 |
-
"loss": 0.
|
398 |
"step": 530
|
399 |
},
|
400 |
{
|
401 |
"epoch": 2.710888610763454,
|
402 |
-
"grad_norm": 2.
|
403 |
-
"learning_rate":
|
404 |
-
"loss": 0.
|
405 |
"step": 540
|
406 |
},
|
407 |
{
|
408 |
"epoch": 2.760951188986233,
|
409 |
-
"grad_norm":
|
410 |
-
"learning_rate":
|
411 |
-
"loss": 0.
|
412 |
"step": 550
|
413 |
},
|
414 |
{
|
415 |
"epoch": 2.8110137672090114,
|
416 |
-
"grad_norm":
|
417 |
-
"learning_rate":
|
418 |
-
"loss": 0.
|
419 |
"step": 560
|
420 |
},
|
421 |
{
|
422 |
"epoch": 2.8610763454317896,
|
423 |
-
"grad_norm":
|
424 |
-
"learning_rate":
|
425 |
-
"loss": 0.
|
426 |
"step": 570
|
427 |
},
|
428 |
{
|
429 |
"epoch": 2.911138923654568,
|
430 |
-
"grad_norm": 3.
|
431 |
-
"learning_rate":
|
432 |
-
"loss": 0.
|
433 |
"step": 580
|
434 |
},
|
435 |
{
|
436 |
"epoch": 2.9612015018773468,
|
437 |
-
"grad_norm": 3.
|
438 |
-
"learning_rate":
|
439 |
-
"loss": 0.
|
440 |
"step": 590
|
441 |
},
|
442 |
{
|
443 |
"epoch": 2.9962453066332917,
|
444 |
-
"eval_accuracy": 0.
|
445 |
-
"eval_loss": 0.
|
446 |
-
"eval_runtime": 4.
|
447 |
-
"eval_samples_per_second":
|
448 |
-
"eval_steps_per_second": 43.
|
449 |
"step": 597
|
450 |
},
|
451 |
{
|
452 |
"epoch": 3.0150187734668337,
|
453 |
-
"grad_norm": 5.
|
454 |
-
"learning_rate":
|
455 |
-
"loss": 0.
|
456 |
"step": 600
|
457 |
},
|
458 |
{
|
459 |
"epoch": 3.065081351689612,
|
460 |
-
"grad_norm":
|
461 |
-
"learning_rate":
|
462 |
-
"loss": 0.
|
463 |
"step": 610
|
464 |
},
|
465 |
{
|
466 |
"epoch": 3.1151439299123904,
|
467 |
-
"grad_norm":
|
468 |
-
"learning_rate":
|
469 |
-
"loss": 0.
|
470 |
"step": 620
|
471 |
},
|
472 |
{
|
473 |
"epoch": 3.165206508135169,
|
474 |
-
"grad_norm": 3.
|
475 |
-
"learning_rate":
|
476 |
-
"loss": 0.
|
477 |
"step": 630
|
478 |
},
|
479 |
{
|
480 |
"epoch": 3.2152690863579476,
|
481 |
-
"grad_norm":
|
482 |
-
"learning_rate":
|
483 |
-
"loss": 0.
|
484 |
"step": 640
|
485 |
},
|
486 |
{
|
487 |
"epoch": 3.2653316645807258,
|
488 |
-
"grad_norm":
|
489 |
-
"learning_rate":
|
490 |
-
"loss": 0.
|
491 |
"step": 650
|
492 |
},
|
493 |
{
|
494 |
"epoch": 3.3153942428035044,
|
495 |
-
"grad_norm": 2.
|
496 |
-
"learning_rate":
|
497 |
-
"loss": 0.
|
498 |
"step": 660
|
499 |
},
|
500 |
{
|
501 |
"epoch": 3.365456821026283,
|
502 |
-
"grad_norm":
|
503 |
-
"learning_rate":
|
504 |
-
"loss": 0.
|
505 |
"step": 670
|
506 |
},
|
507 |
{
|
508 |
"epoch": 3.415519399249061,
|
509 |
-
"grad_norm":
|
510 |
-
"learning_rate":
|
511 |
-
"loss": 0.
|
512 |
"step": 680
|
513 |
},
|
514 |
{
|
515 |
"epoch": 3.4655819774718397,
|
516 |
-
"grad_norm": 2.
|
517 |
-
"learning_rate":
|
518 |
-
"loss": 0.
|
519 |
"step": 690
|
520 |
},
|
521 |
{
|
522 |
"epoch": 3.5156445556946183,
|
523 |
-
"grad_norm":
|
524 |
-
"learning_rate":
|
525 |
-
"loss": 0.
|
526 |
"step": 700
|
527 |
},
|
528 |
{
|
529 |
"epoch": 3.565707133917397,
|
530 |
-
"grad_norm":
|
531 |
-
"learning_rate":
|
532 |
-
"loss": 0.
|
533 |
"step": 710
|
534 |
},
|
535 |
{
|
536 |
"epoch": 3.615769712140175,
|
537 |
-
"grad_norm": 3.
|
538 |
-
"learning_rate":
|
539 |
-
"loss": 0.
|
540 |
"step": 720
|
541 |
},
|
542 |
{
|
543 |
"epoch": 3.6658322903629537,
|
544 |
-
"grad_norm":
|
545 |
-
"learning_rate":
|
546 |
-
"loss": 0.
|
547 |
"step": 730
|
548 |
},
|
549 |
{
|
550 |
"epoch": 3.7158948685857323,
|
551 |
-
"grad_norm": 3.
|
552 |
-
"learning_rate":
|
553 |
-
"loss": 0.
|
554 |
"step": 740
|
555 |
},
|
556 |
{
|
557 |
"epoch": 3.7659574468085104,
|
558 |
-
"grad_norm":
|
559 |
-
"learning_rate":
|
560 |
-
"loss": 0.
|
561 |
"step": 750
|
562 |
},
|
563 |
{
|
564 |
"epoch": 3.816020025031289,
|
565 |
-
"grad_norm":
|
566 |
-
"learning_rate":
|
567 |
-
"loss": 0.
|
568 |
"step": 760
|
569 |
},
|
570 |
{
|
571 |
"epoch": 3.8660826032540676,
|
572 |
-
"grad_norm":
|
573 |
-
"learning_rate":
|
574 |
-
"loss": 0.
|
575 |
"step": 770
|
576 |
},
|
577 |
{
|
578 |
"epoch": 3.916145181476846,
|
579 |
-
"grad_norm":
|
580 |
-
"learning_rate":
|
581 |
-
"loss": 0.
|
582 |
"step": 780
|
583 |
},
|
584 |
{
|
585 |
"epoch": 3.966207759699625,
|
586 |
-
"grad_norm": 3.
|
587 |
-
"learning_rate":
|
588 |
-
"loss": 0.
|
589 |
"step": 790
|
590 |
},
|
591 |
{
|
592 |
"epoch": 3.9962453066332917,
|
593 |
-
"eval_accuracy": 0.
|
594 |
-
"eval_loss": 0.
|
595 |
-
"eval_runtime": 4.
|
596 |
-
"eval_samples_per_second":
|
597 |
-
"eval_steps_per_second":
|
598 |
"step": 796
|
599 |
},
|
600 |
{
|
601 |
"epoch": 4.020025031289111,
|
602 |
-
"grad_norm":
|
603 |
-
"learning_rate":
|
604 |
-
"loss": 0.
|
605 |
"step": 800
|
606 |
},
|
607 |
{
|
608 |
"epoch": 4.07008760951189,
|
609 |
-
"grad_norm": 2.
|
610 |
-
"learning_rate":
|
611 |
-
"loss": 0.
|
612 |
"step": 810
|
613 |
},
|
614 |
{
|
615 |
"epoch": 4.1201501877346685,
|
616 |
-
"grad_norm":
|
617 |
-
"learning_rate":
|
618 |
-
"loss": 0.
|
619 |
"step": 820
|
620 |
},
|
621 |
{
|
622 |
"epoch": 4.170212765957447,
|
623 |
-
"grad_norm":
|
624 |
-
"learning_rate":
|
625 |
-
"loss": 0.
|
626 |
"step": 830
|
627 |
},
|
628 |
{
|
629 |
"epoch": 4.220275344180226,
|
630 |
-
"grad_norm":
|
631 |
-
"learning_rate":
|
632 |
-
"loss": 0.
|
633 |
"step": 840
|
634 |
},
|
635 |
{
|
636 |
"epoch": 4.270337922403003,
|
637 |
-
"grad_norm": 2.
|
638 |
-
"learning_rate":
|
639 |
-
"loss": 0.
|
640 |
"step": 850
|
641 |
},
|
642 |
{
|
643 |
"epoch": 4.320400500625782,
|
644 |
-
"grad_norm":
|
645 |
-
"learning_rate":
|
646 |
-
"loss": 0.
|
647 |
"step": 860
|
648 |
},
|
649 |
{
|
650 |
"epoch": 4.370463078848561,
|
651 |
-
"grad_norm":
|
652 |
-
"learning_rate":
|
653 |
-
"loss": 0.
|
654 |
"step": 870
|
655 |
},
|
656 |
{
|
657 |
"epoch": 4.420525657071339,
|
658 |
-
"grad_norm":
|
659 |
-
"learning_rate":
|
660 |
-
"loss": 0.
|
661 |
"step": 880
|
662 |
},
|
663 |
{
|
664 |
"epoch": 4.470588235294118,
|
665 |
-
"grad_norm": 2.
|
666 |
-
"learning_rate":
|
667 |
-
"loss": 0.
|
668 |
"step": 890
|
669 |
},
|
670 |
{
|
671 |
"epoch": 4.520650813516896,
|
672 |
-
"grad_norm": 1.
|
673 |
-
"learning_rate":
|
674 |
-
"loss": 0.
|
675 |
"step": 900
|
676 |
},
|
677 |
{
|
678 |
"epoch": 4.570713391739675,
|
679 |
-
"grad_norm":
|
680 |
-
"learning_rate":
|
681 |
-
"loss": 0.
|
682 |
"step": 910
|
683 |
},
|
684 |
{
|
685 |
"epoch": 4.6207759699624535,
|
686 |
-
"grad_norm":
|
687 |
-
"learning_rate":
|
688 |
-
"loss": 0.
|
689 |
"step": 920
|
690 |
},
|
691 |
{
|
692 |
"epoch": 4.670838548185231,
|
693 |
-
"grad_norm":
|
694 |
-
"learning_rate":
|
695 |
-
"loss": 0.
|
696 |
"step": 930
|
697 |
},
|
698 |
{
|
699 |
"epoch": 4.72090112640801,
|
700 |
-
"grad_norm":
|
701 |
-
"learning_rate":
|
702 |
-
"loss": 0.
|
703 |
"step": 940
|
704 |
},
|
705 |
{
|
706 |
"epoch": 4.7709637046307884,
|
707 |
-
"grad_norm": 3.
|
708 |
-
"learning_rate":
|
709 |
-
"loss": 0.
|
710 |
"step": 950
|
711 |
},
|
712 |
{
|
713 |
"epoch": 4.821026282853567,
|
714 |
-
"grad_norm": 3.
|
715 |
-
"learning_rate":
|
716 |
-
"loss": 0.
|
717 |
"step": 960
|
718 |
},
|
719 |
{
|
720 |
"epoch": 4.871088861076346,
|
721 |
-
"grad_norm":
|
722 |
-
"learning_rate":
|
723 |
-
"loss": 0.
|
724 |
"step": 970
|
725 |
},
|
726 |
{
|
727 |
"epoch": 4.921151439299124,
|
728 |
-
"grad_norm":
|
729 |
-
"learning_rate":
|
730 |
-
"loss": 0.
|
731 |
"step": 980
|
732 |
},
|
733 |
{
|
734 |
"epoch": 4.971214017521902,
|
735 |
-
"grad_norm": 3.
|
736 |
-
"learning_rate":
|
737 |
-
"loss": 0.
|
738 |
"step": 990
|
739 |
},
|
740 |
{
|
741 |
"epoch": 4.996245306633291,
|
742 |
-
"eval_accuracy": 0.
|
743 |
-
"eval_loss": 0.
|
744 |
-
"eval_runtime":
|
745 |
-
"eval_samples_per_second":
|
746 |
-
"eval_steps_per_second":
|
747 |
"step": 995
|
748 |
},
|
749 |
{
|
750 |
"epoch": 5.025031289111389,
|
751 |
-
"grad_norm":
|
752 |
-
"learning_rate":
|
753 |
-
"loss": 0.
|
754 |
"step": 1000
|
755 |
},
|
756 |
{
|
757 |
"epoch": 5.075093867334168,
|
758 |
-
"grad_norm":
|
759 |
-
"learning_rate":
|
760 |
-
"loss": 0.
|
761 |
"step": 1010
|
762 |
},
|
763 |
{
|
764 |
"epoch": 5.1251564455569465,
|
765 |
-
"grad_norm": 3.
|
766 |
-
"learning_rate":
|
767 |
-
"loss": 0.
|
768 |
"step": 1020
|
769 |
},
|
770 |
{
|
771 |
"epoch": 5.175219023779725,
|
772 |
-
"grad_norm":
|
773 |
-
"learning_rate":
|
774 |
-
"loss": 0.
|
775 |
"step": 1030
|
776 |
},
|
777 |
{
|
778 |
"epoch": 5.225281602002503,
|
779 |
-
"grad_norm":
|
780 |
-
"learning_rate":
|
781 |
-
"loss": 0.
|
782 |
"step": 1040
|
783 |
},
|
784 |
{
|
785 |
"epoch": 5.275344180225281,
|
786 |
-
"grad_norm": 2.
|
787 |
-
"learning_rate":
|
788 |
-
"loss": 0.
|
789 |
"step": 1050
|
790 |
},
|
791 |
{
|
792 |
"epoch": 5.32540675844806,
|
793 |
-
"grad_norm": 2.
|
794 |
-
"learning_rate":
|
795 |
-
"loss": 0.
|
796 |
"step": 1060
|
797 |
},
|
798 |
{
|
799 |
"epoch": 5.375469336670839,
|
800 |
-
"grad_norm":
|
801 |
-
"learning_rate":
|
802 |
-
"loss": 0.
|
803 |
"step": 1070
|
804 |
},
|
805 |
{
|
806 |
"epoch": 5.425531914893617,
|
807 |
-
"grad_norm": 2.
|
808 |
-
"learning_rate":
|
809 |
-
"loss": 0.
|
810 |
"step": 1080
|
811 |
},
|
812 |
{
|
813 |
"epoch": 5.475594493116396,
|
814 |
-
"grad_norm":
|
815 |
-
"learning_rate":
|
816 |
-
"loss": 0.
|
817 |
"step": 1090
|
818 |
},
|
819 |
{
|
820 |
"epoch": 5.5256570713391735,
|
821 |
-
"grad_norm":
|
822 |
-
"learning_rate":
|
823 |
-
"loss": 0.
|
824 |
"step": 1100
|
825 |
},
|
826 |
{
|
827 |
"epoch": 5.575719649561952,
|
828 |
-
"grad_norm": 2.
|
829 |
-
"learning_rate":
|
830 |
-
"loss": 0.
|
831 |
"step": 1110
|
832 |
},
|
833 |
{
|
834 |
"epoch": 5.625782227784731,
|
835 |
-
"grad_norm":
|
836 |
-
"learning_rate":
|
837 |
-
"loss": 0.
|
838 |
"step": 1120
|
839 |
},
|
840 |
{
|
841 |
"epoch": 5.675844806007509,
|
842 |
-
"grad_norm":
|
843 |
-
"learning_rate":
|
844 |
-
"loss": 0.
|
845 |
"step": 1130
|
846 |
},
|
847 |
{
|
848 |
"epoch": 5.725907384230288,
|
849 |
-
"grad_norm":
|
850 |
-
"learning_rate":
|
851 |
-
"loss": 0.
|
852 |
"step": 1140
|
853 |
},
|
854 |
{
|
855 |
"epoch": 5.7759699624530665,
|
856 |
-
"grad_norm": 3.
|
857 |
-
"learning_rate":
|
858 |
-
"loss": 0.
|
859 |
"step": 1150
|
860 |
},
|
861 |
{
|
862 |
"epoch": 5.826032540675845,
|
863 |
-
"grad_norm": 3.
|
864 |
-
"learning_rate":
|
865 |
-
"loss": 0.
|
866 |
"step": 1160
|
867 |
},
|
868 |
{
|
869 |
"epoch": 5.876095118898624,
|
870 |
-
"grad_norm": 2.
|
871 |
-
"learning_rate":
|
872 |
-
"loss": 0.
|
873 |
"step": 1170
|
874 |
},
|
875 |
{
|
876 |
"epoch": 5.926157697121401,
|
877 |
-
"grad_norm":
|
878 |
-
"learning_rate":
|
879 |
-
"loss": 0.
|
880 |
"step": 1180
|
881 |
},
|
882 |
{
|
883 |
"epoch": 5.97622027534418,
|
884 |
-
"grad_norm":
|
885 |
-
"learning_rate":
|
886 |
-
"loss": 0.
|
887 |
"step": 1190
|
888 |
},
|
889 |
{
|
890 |
"epoch": 5.996245306633291,
|
891 |
-
"eval_accuracy": 0.
|
892 |
-
"eval_loss": 0.
|
893 |
-
"eval_runtime":
|
894 |
-
"eval_samples_per_second":
|
895 |
-
"eval_steps_per_second":
|
896 |
"step": 1194
|
897 |
},
|
898 |
{
|
899 |
"epoch": 6.030037546933667,
|
900 |
-
"grad_norm":
|
901 |
-
"learning_rate":
|
902 |
-
"loss": 0.
|
903 |
"step": 1200
|
904 |
},
|
905 |
{
|
906 |
"epoch": 6.080100125156446,
|
907 |
-
"grad_norm": 2.
|
908 |
-
"learning_rate":
|
909 |
-
"loss": 0.
|
910 |
"step": 1210
|
911 |
},
|
912 |
{
|
913 |
"epoch": 6.130162703379224,
|
914 |
-
"grad_norm": 1.
|
915 |
-
"learning_rate":
|
916 |
-
"loss": 0.
|
917 |
"step": 1220
|
918 |
},
|
919 |
{
|
920 |
"epoch": 6.180225281602002,
|
921 |
-
"grad_norm":
|
922 |
-
"learning_rate":
|
923 |
-
"loss": 0.
|
924 |
"step": 1230
|
925 |
},
|
926 |
{
|
927 |
"epoch": 6.230287859824781,
|
928 |
-
"grad_norm": 2.
|
929 |
-
"learning_rate":
|
930 |
-
"loss": 0.
|
931 |
"step": 1240
|
932 |
},
|
933 |
{
|
934 |
"epoch": 6.280350438047559,
|
935 |
-
"grad_norm":
|
936 |
-
"learning_rate":
|
937 |
-
"loss": 0.
|
938 |
"step": 1250
|
939 |
},
|
940 |
{
|
941 |
"epoch": 6.330413016270338,
|
942 |
-
"grad_norm":
|
943 |
-
"learning_rate":
|
944 |
-
"loss": 0.
|
945 |
"step": 1260
|
946 |
},
|
947 |
{
|
948 |
"epoch": 6.380475594493117,
|
949 |
-
"grad_norm": 2.
|
950 |
-
"learning_rate":
|
951 |
-
"loss": 0.
|
952 |
"step": 1270
|
953 |
},
|
954 |
{
|
955 |
"epoch": 6.430538172715895,
|
956 |
-
"grad_norm": 2.
|
957 |
-
"learning_rate":
|
958 |
-
"loss": 0.
|
959 |
"step": 1280
|
960 |
},
|
961 |
{
|
962 |
"epoch": 6.480600750938673,
|
963 |
-
"grad_norm": 2.
|
964 |
-
"learning_rate":
|
965 |
-
"loss": 0.
|
966 |
"step": 1290
|
967 |
},
|
968 |
{
|
969 |
"epoch": 6.5306633291614515,
|
970 |
-
"grad_norm":
|
971 |
-
"learning_rate":
|
972 |
-
"loss": 0.
|
973 |
"step": 1300
|
974 |
},
|
975 |
{
|
976 |
"epoch": 6.58072590738423,
|
977 |
-
"grad_norm":
|
978 |
-
"learning_rate":
|
979 |
-
"loss": 0.
|
980 |
"step": 1310
|
981 |
},
|
982 |
{
|
983 |
"epoch": 6.630788485607009,
|
984 |
-
"grad_norm": 2.
|
985 |
-
"learning_rate":
|
986 |
-
"loss": 0.
|
987 |
"step": 1320
|
988 |
},
|
989 |
{
|
990 |
"epoch": 6.680851063829787,
|
991 |
-
"grad_norm": 2.
|
992 |
-
"learning_rate":
|
993 |
-
"loss": 0.
|
994 |
"step": 1330
|
995 |
},
|
996 |
{
|
997 |
"epoch": 6.730913642052566,
|
998 |
-
"grad_norm":
|
999 |
-
"learning_rate":
|
1000 |
-
"loss": 0.
|
1001 |
"step": 1340
|
1002 |
},
|
1003 |
{
|
1004 |
"epoch": 6.7809762202753445,
|
1005 |
-
"grad_norm":
|
1006 |
-
"learning_rate":
|
1007 |
-
"loss": 0.
|
1008 |
"step": 1350
|
1009 |
},
|
1010 |
{
|
1011 |
"epoch": 6.831038798498122,
|
1012 |
-
"grad_norm":
|
1013 |
-
"learning_rate":
|
1014 |
-
"loss": 0.
|
1015 |
"step": 1360
|
1016 |
},
|
1017 |
{
|
1018 |
"epoch": 6.881101376720901,
|
1019 |
-
"grad_norm": 2.
|
1020 |
-
"learning_rate":
|
1021 |
-
"loss": 0.
|
1022 |
"step": 1370
|
1023 |
},
|
1024 |
{
|
1025 |
"epoch": 6.931163954943679,
|
1026 |
-
"grad_norm": 2.
|
1027 |
-
"learning_rate":
|
1028 |
-
"loss": 0.
|
1029 |
"step": 1380
|
1030 |
},
|
1031 |
{
|
1032 |
"epoch": 6.981226533166458,
|
1033 |
-
"grad_norm":
|
1034 |
-
"learning_rate":
|
1035 |
-
"loss": 0.
|
1036 |
"step": 1390
|
1037 |
},
|
1038 |
{
|
1039 |
"epoch": 6.996245306633291,
|
1040 |
-
"eval_accuracy": 0.
|
1041 |
-
"eval_loss": 0.
|
1042 |
-
"eval_runtime":
|
1043 |
-
"eval_samples_per_second":
|
1044 |
-
"eval_steps_per_second":
|
1045 |
"step": 1393
|
1046 |
},
|
1047 |
{
|
1048 |
"epoch": 7.035043804755945,
|
1049 |
-
"grad_norm": 2.
|
1050 |
-
"learning_rate":
|
1051 |
-
"loss": 0.
|
1052 |
"step": 1400
|
1053 |
},
|
1054 |
{
|
1055 |
"epoch": 7.085106382978723,
|
1056 |
-
"grad_norm": 2.
|
1057 |
-
"learning_rate":
|
1058 |
-
"loss": 0.
|
1059 |
"step": 1410
|
1060 |
},
|
1061 |
{
|
1062 |
"epoch": 7.135168961201502,
|
1063 |
-
"grad_norm":
|
1064 |
-
"learning_rate":
|
1065 |
-
"loss": 0.
|
1066 |
"step": 1420
|
1067 |
},
|
1068 |
{
|
1069 |
"epoch": 7.18523153942428,
|
1070 |
-
"grad_norm":
|
1071 |
-
"learning_rate":
|
1072 |
-
"loss": 0.
|
1073 |
"step": 1430
|
1074 |
},
|
1075 |
{
|
1076 |
"epoch": 7.235294117647059,
|
1077 |
-
"grad_norm":
|
1078 |
-
"learning_rate":
|
1079 |
-
"loss": 0.
|
1080 |
"step": 1440
|
1081 |
},
|
1082 |
{
|
1083 |
"epoch": 7.2853566958698375,
|
1084 |
-
"grad_norm":
|
1085 |
-
"learning_rate":
|
1086 |
-
"loss": 0.
|
1087 |
"step": 1450
|
1088 |
},
|
1089 |
{
|
1090 |
"epoch": 7.335419274092616,
|
1091 |
-
"grad_norm":
|
1092 |
-
"learning_rate":
|
1093 |
-
"loss": 0.
|
1094 |
"step": 1460
|
1095 |
},
|
1096 |
{
|
1097 |
"epoch": 7.385481852315394,
|
1098 |
-
"grad_norm":
|
1099 |
-
"learning_rate":
|
1100 |
-
"loss": 0.
|
1101 |
"step": 1470
|
1102 |
},
|
1103 |
{
|
1104 |
"epoch": 7.435544430538172,
|
1105 |
-
"grad_norm": 2.
|
1106 |
-
"learning_rate":
|
1107 |
-
"loss": 0.
|
1108 |
"step": 1480
|
1109 |
},
|
1110 |
{
|
1111 |
"epoch": 7.485607008760951,
|
1112 |
-
"grad_norm":
|
1113 |
-
"learning_rate":
|
1114 |
-
"loss": 0.
|
1115 |
"step": 1490
|
1116 |
},
|
1117 |
{
|
1118 |
"epoch": 7.53566958698373,
|
1119 |
-
"grad_norm":
|
1120 |
-
"learning_rate":
|
1121 |
-
"loss": 0.
|
1122 |
"step": 1500
|
1123 |
},
|
1124 |
{
|
1125 |
"epoch": 7.585732165206508,
|
1126 |
-
"grad_norm":
|
1127 |
-
"learning_rate":
|
1128 |
-
"loss": 0.
|
1129 |
"step": 1510
|
1130 |
},
|
1131 |
{
|
1132 |
"epoch": 7.635794743429287,
|
1133 |
-
"grad_norm": 2.
|
1134 |
-
"learning_rate":
|
1135 |
-
"loss": 0.
|
1136 |
"step": 1520
|
1137 |
},
|
1138 |
{
|
1139 |
"epoch": 7.685857321652065,
|
1140 |
-
"grad_norm": 2.
|
1141 |
-
"learning_rate":
|
1142 |
-
"loss": 0.
|
1143 |
"step": 1530
|
1144 |
},
|
1145 |
{
|
1146 |
"epoch": 7.735919899874844,
|
1147 |
-
"grad_norm":
|
1148 |
-
"learning_rate":
|
1149 |
-
"loss": 0.
|
1150 |
"step": 1540
|
1151 |
},
|
1152 |
{
|
1153 |
"epoch": 7.785982478097622,
|
1154 |
-
"grad_norm":
|
1155 |
-
"learning_rate":
|
1156 |
-
"loss": 0.
|
1157 |
"step": 1550
|
1158 |
},
|
1159 |
{
|
1160 |
"epoch": 7.8360450563204,
|
1161 |
-
"grad_norm": 1.
|
1162 |
-
"learning_rate":
|
1163 |
-
"loss": 0.
|
1164 |
"step": 1560
|
1165 |
},
|
1166 |
{
|
1167 |
"epoch": 7.886107634543179,
|
1168 |
-
"grad_norm":
|
1169 |
-
"learning_rate":
|
1170 |
-
"loss": 0.
|
1171 |
"step": 1570
|
1172 |
},
|
1173 |
{
|
1174 |
"epoch": 7.9361702127659575,
|
1175 |
-
"grad_norm": 1.
|
1176 |
-
"learning_rate":
|
1177 |
-
"loss": 0.
|
1178 |
"step": 1580
|
1179 |
},
|
1180 |
{
|
1181 |
"epoch": 7.986232790988736,
|
1182 |
-
"grad_norm":
|
1183 |
-
"learning_rate":
|
1184 |
-
"loss": 0.
|
1185 |
"step": 1590
|
1186 |
},
|
1187 |
{
|
1188 |
"epoch": 7.996245306633291,
|
1189 |
-
"eval_accuracy": 0.
|
1190 |
-
"eval_loss": 0.
|
1191 |
-
"eval_runtime": 5.
|
1192 |
-
"eval_samples_per_second":
|
1193 |
-
"eval_steps_per_second": 38.
|
1194 |
"step": 1592
|
1195 |
},
|
1196 |
{
|
1197 |
"epoch": 7.996245306633291,
|
1198 |
"step": 1592,
|
1199 |
"total_flos": 3.777723239743488e+18,
|
1200 |
-
"train_loss": 0.
|
1201 |
-
"train_runtime":
|
1202 |
-
"train_samples_per_second":
|
1203 |
-
"train_steps_per_second": 2.
|
1204 |
}
|
1205 |
],
|
1206 |
"logging_steps": 10,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.9814651368049426,
|
3 |
+
"best_model_checkpoint": "wav2vec2-base-ft-keyword-spotting/checkpoint-1393",
|
4 |
"epoch": 7.996245306633291,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 1592,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.05006257822277847,
|
13 |
+
"grad_norm": 2.2161571979522705,
|
14 |
+
"learning_rate": 1.25e-06,
|
15 |
+
"loss": 4.0993,
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.10012515644555695,
|
20 |
+
"grad_norm": 3.068307399749756,
|
21 |
+
"learning_rate": 2.5e-06,
|
22 |
+
"loss": 4.1801,
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.15018773466833543,
|
27 |
+
"grad_norm": 3.121358871459961,
|
28 |
+
"learning_rate": 3.7500000000000005e-06,
|
29 |
+
"loss": 4.11,
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.2002503128911139,
|
34 |
+
"grad_norm": 3.456892728805542,
|
35 |
+
"learning_rate": 5e-06,
|
36 |
+
"loss": 3.9697,
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.2503128911138924,
|
41 |
+
"grad_norm": 4.699773788452148,
|
42 |
+
"learning_rate": 6.25e-06,
|
43 |
+
"loss": 3.7271,
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.30037546933667086,
|
48 |
+
"grad_norm": 6.0911078453063965,
|
49 |
+
"learning_rate": 7.500000000000001e-06,
|
50 |
+
"loss": 3.3436,
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.3504380475594493,
|
55 |
+
"grad_norm": 5.996876239776611,
|
56 |
+
"learning_rate": 8.750000000000001e-06,
|
57 |
+
"loss": 2.8803,
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
"epoch": 0.4005006257822278,
|
62 |
+
"grad_norm": 5.593225479125977,
|
63 |
+
"learning_rate": 1e-05,
|
64 |
+
"loss": 2.54,
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
"epoch": 0.45056320400500627,
|
69 |
+
"grad_norm": 5.177072525024414,
|
70 |
+
"learning_rate": 1.125e-05,
|
71 |
+
"loss": 2.326,
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
"epoch": 0.5006257822277848,
|
76 |
+
"grad_norm": 4.610163688659668,
|
77 |
+
"learning_rate": 1.25e-05,
|
78 |
+
"loss": 2.1907,
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
"epoch": 0.5506883604505632,
|
83 |
+
"grad_norm": 4.401670932769775,
|
84 |
+
"learning_rate": 1.375e-05,
|
85 |
+
"loss": 2.054,
|
86 |
"step": 110
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.6007509386733417,
|
90 |
+
"grad_norm": 3.190901041030884,
|
91 |
+
"learning_rate": 1.5000000000000002e-05,
|
92 |
+
"loss": 1.994,
|
93 |
"step": 120
|
94 |
},
|
95 |
{
|
96 |
"epoch": 0.6508135168961201,
|
97 |
+
"grad_norm": 2.314962148666382,
|
98 |
+
"learning_rate": 1.6250000000000002e-05,
|
99 |
+
"loss": 1.9256,
|
100 |
"step": 130
|
101 |
},
|
102 |
{
|
103 |
"epoch": 0.7008760951188986,
|
104 |
+
"grad_norm": 2.310375690460205,
|
105 |
+
"learning_rate": 1.7500000000000002e-05,
|
106 |
+
"loss": 1.7774,
|
107 |
"step": 140
|
108 |
},
|
109 |
{
|
110 |
"epoch": 0.7509386733416771,
|
111 |
+
"grad_norm": 1.4242396354675293,
|
112 |
+
"learning_rate": 1.8750000000000002e-05,
|
113 |
+
"loss": 1.7961,
|
114 |
"step": 150
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.8010012515644556,
|
118 |
+
"grad_norm": 0.6981024742126465,
|
119 |
+
"learning_rate": 2e-05,
|
120 |
+
"loss": 1.8108,
|
121 |
"step": 160
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.851063829787234,
|
125 |
+
"grad_norm": 0.7541513442993164,
|
126 |
+
"learning_rate": 1.9860335195530728e-05,
|
127 |
+
"loss": 1.733,
|
128 |
"step": 170
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.9011264080100125,
|
132 |
+
"grad_norm": 1.1699227094650269,
|
133 |
+
"learning_rate": 1.9720670391061455e-05,
|
134 |
+
"loss": 1.759,
|
135 |
"step": 180
|
136 |
},
|
137 |
{
|
138 |
"epoch": 0.951188986232791,
|
139 |
+
"grad_norm": 2.1834521293640137,
|
140 |
+
"learning_rate": 1.958100558659218e-05,
|
141 |
+
"loss": 1.7191,
|
142 |
"step": 190
|
143 |
},
|
144 |
{
|
145 |
"epoch": 0.9962453066332916,
|
146 |
+
"eval_accuracy": 0.6209179170344219,
|
147 |
+
"eval_loss": 1.5815445184707642,
|
148 |
+
"eval_runtime": 4.9783,
|
149 |
+
"eval_samples_per_second": 1365.528,
|
150 |
+
"eval_steps_per_second": 42.786,
|
151 |
"step": 199
|
152 |
},
|
153 |
{
|
154 |
"epoch": 1.0050062578222778,
|
155 |
+
"grad_norm": 3.1251320838928223,
|
156 |
+
"learning_rate": 1.9441340782122907e-05,
|
157 |
+
"loss": 1.7482,
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
"epoch": 1.0550688360450564,
|
162 |
+
"grad_norm": 4.351515293121338,
|
163 |
+
"learning_rate": 1.9301675977653634e-05,
|
164 |
+
"loss": 1.5606,
|
165 |
"step": 210
|
166 |
},
|
167 |
{
|
168 |
"epoch": 1.1051314142678348,
|
169 |
+
"grad_norm": 2.624279022216797,
|
170 |
+
"learning_rate": 1.9162011173184357e-05,
|
171 |
+
"loss": 1.5657,
|
172 |
"step": 220
|
173 |
},
|
174 |
{
|
175 |
"epoch": 1.1551939924906134,
|
176 |
+
"grad_norm": 2.795576333999634,
|
177 |
+
"learning_rate": 1.9022346368715087e-05,
|
178 |
+
"loss": 1.4578,
|
179 |
"step": 230
|
180 |
},
|
181 |
{
|
182 |
"epoch": 1.2052565707133918,
|
183 |
+
"grad_norm": 2.7234373092651367,
|
184 |
+
"learning_rate": 1.888268156424581e-05,
|
185 |
+
"loss": 1.449,
|
186 |
"step": 240
|
187 |
},
|
188 |
{
|
189 |
"epoch": 1.2553191489361701,
|
190 |
+
"grad_norm": 5.743226528167725,
|
191 |
+
"learning_rate": 1.8743016759776536e-05,
|
192 |
+
"loss": 1.4041,
|
193 |
"step": 250
|
194 |
},
|
195 |
{
|
196 |
"epoch": 1.3053817271589487,
|
197 |
+
"grad_norm": 6.164392471313477,
|
198 |
+
"learning_rate": 1.8603351955307266e-05,
|
199 |
+
"loss": 1.3667,
|
200 |
"step": 260
|
201 |
},
|
202 |
{
|
203 |
"epoch": 1.355444305381727,
|
204 |
+
"grad_norm": 3.474940299987793,
|
205 |
+
"learning_rate": 1.846368715083799e-05,
|
206 |
+
"loss": 1.2949,
|
207 |
"step": 270
|
208 |
},
|
209 |
{
|
210 |
"epoch": 1.4055068836045057,
|
211 |
+
"grad_norm": 4.262696266174316,
|
212 |
+
"learning_rate": 1.8324022346368716e-05,
|
213 |
+
"loss": 1.2814,
|
214 |
"step": 280
|
215 |
},
|
216 |
{
|
217 |
"epoch": 1.455569461827284,
|
218 |
+
"grad_norm": 7.003787994384766,
|
219 |
+
"learning_rate": 1.8184357541899442e-05,
|
220 |
+
"loss": 1.2096,
|
221 |
"step": 290
|
222 |
},
|
223 |
{
|
224 |
"epoch": 1.5056320400500627,
|
225 |
+
"grad_norm": 3.957937240600586,
|
226 |
+
"learning_rate": 1.804469273743017e-05,
|
227 |
+
"loss": 1.183,
|
228 |
"step": 300
|
229 |
},
|
230 |
{
|
231 |
"epoch": 1.555694618272841,
|
232 |
+
"grad_norm": 4.9982991218566895,
|
233 |
+
"learning_rate": 1.7905027932960895e-05,
|
234 |
+
"loss": 1.2105,
|
235 |
"step": 310
|
236 |
},
|
237 |
{
|
238 |
"epoch": 1.6057571964956194,
|
239 |
+
"grad_norm": 3.900266647338867,
|
240 |
+
"learning_rate": 1.776536312849162e-05,
|
241 |
+
"loss": 1.1167,
|
242 |
"step": 320
|
243 |
},
|
244 |
{
|
245 |
"epoch": 1.655819774718398,
|
246 |
+
"grad_norm": 4.397927284240723,
|
247 |
+
"learning_rate": 1.7625698324022348e-05,
|
248 |
+
"loss": 1.144,
|
249 |
"step": 330
|
250 |
},
|
251 |
{
|
252 |
"epoch": 1.7058823529411766,
|
253 |
+
"grad_norm": 3.0843544006347656,
|
254 |
+
"learning_rate": 1.7486033519553075e-05,
|
255 |
+
"loss": 1.101,
|
256 |
"step": 340
|
257 |
},
|
258 |
{
|
259 |
"epoch": 1.7559449311639548,
|
260 |
+
"grad_norm": 3.6377310752868652,
|
261 |
+
"learning_rate": 1.73463687150838e-05,
|
262 |
+
"loss": 1.0482,
|
263 |
"step": 350
|
264 |
},
|
265 |
{
|
266 |
"epoch": 1.8060075093867334,
|
267 |
+
"grad_norm": 2.659217357635498,
|
268 |
+
"learning_rate": 1.7206703910614527e-05,
|
269 |
+
"loss": 1.0428,
|
270 |
"step": 360
|
271 |
},
|
272 |
{
|
273 |
"epoch": 1.856070087609512,
|
274 |
+
"grad_norm": 4.358920097351074,
|
275 |
+
"learning_rate": 1.7067039106145254e-05,
|
276 |
+
"loss": 1.0483,
|
277 |
"step": 370
|
278 |
},
|
279 |
{
|
280 |
"epoch": 1.9061326658322904,
|
281 |
+
"grad_norm": 4.7584452629089355,
|
282 |
+
"learning_rate": 1.6927374301675977e-05,
|
283 |
+
"loss": 0.9886,
|
284 |
"step": 380
|
285 |
},
|
286 |
{
|
287 |
"epoch": 1.9561952440550687,
|
288 |
+
"grad_norm": 3.638371229171753,
|
289 |
+
"learning_rate": 1.6787709497206707e-05,
|
290 |
+
"loss": 1.0088,
|
291 |
"step": 390
|
292 |
},
|
293 |
{
|
294 |
"epoch": 1.9962453066332917,
|
295 |
+
"eval_accuracy": 0.83480435422183,
|
296 |
+
"eval_loss": 0.9595487713813782,
|
297 |
+
"eval_runtime": 5.0807,
|
298 |
+
"eval_samples_per_second": 1338.014,
|
299 |
+
"eval_steps_per_second": 41.924,
|
300 |
"step": 398
|
301 |
},
|
302 |
{
|
303 |
"epoch": 2.0100125156445556,
|
304 |
+
"grad_norm": 2.447006940841675,
|
305 |
+
"learning_rate": 1.664804469273743e-05,
|
306 |
+
"loss": 1.0295,
|
307 |
"step": 400
|
308 |
},
|
309 |
{
|
310 |
"epoch": 2.0600750938673342,
|
311 |
+
"grad_norm": 3.891425132751465,
|
312 |
+
"learning_rate": 1.6508379888268156e-05,
|
313 |
+
"loss": 0.9607,
|
314 |
"step": 410
|
315 |
},
|
316 |
{
|
317 |
"epoch": 2.110137672090113,
|
318 |
+
"grad_norm": 3.9283361434936523,
|
319 |
+
"learning_rate": 1.6368715083798886e-05,
|
320 |
+
"loss": 0.9797,
|
321 |
"step": 420
|
322 |
},
|
323 |
{
|
324 |
"epoch": 2.160200250312891,
|
325 |
+
"grad_norm": 3.9875011444091797,
|
326 |
+
"learning_rate": 1.622905027932961e-05,
|
327 |
+
"loss": 0.923,
|
328 |
"step": 430
|
329 |
},
|
330 |
{
|
331 |
"epoch": 2.2102628285356696,
|
332 |
+
"grad_norm": 2.9517266750335693,
|
333 |
+
"learning_rate": 1.6089385474860336e-05,
|
334 |
+
"loss": 0.872,
|
335 |
"step": 440
|
336 |
},
|
337 |
{
|
338 |
"epoch": 2.260325406758448,
|
339 |
+
"grad_norm": 6.958038330078125,
|
340 |
+
"learning_rate": 1.5949720670391062e-05,
|
341 |
+
"loss": 0.8477,
|
342 |
"step": 450
|
343 |
},
|
344 |
{
|
345 |
"epoch": 2.3103879849812268,
|
346 |
+
"grad_norm": 2.592712640762329,
|
347 |
+
"learning_rate": 1.581005586592179e-05,
|
348 |
+
"loss": 0.803,
|
349 |
"step": 460
|
350 |
},
|
351 |
{
|
352 |
"epoch": 2.360450563204005,
|
353 |
+
"grad_norm": 3.0653076171875,
|
354 |
+
"learning_rate": 1.5670391061452515e-05,
|
355 |
+
"loss": 0.7824,
|
356 |
"step": 470
|
357 |
},
|
358 |
{
|
359 |
"epoch": 2.4105131414267835,
|
360 |
+
"grad_norm": 4.134679794311523,
|
361 |
+
"learning_rate": 1.553072625698324e-05,
|
362 |
+
"loss": 0.7681,
|
363 |
"step": 480
|
364 |
},
|
365 |
{
|
366 |
"epoch": 2.460575719649562,
|
367 |
+
"grad_norm": 2.6600096225738525,
|
368 |
+
"learning_rate": 1.5391061452513968e-05,
|
369 |
+
"loss": 0.6916,
|
370 |
"step": 490
|
371 |
},
|
372 |
{
|
373 |
"epoch": 2.5106382978723403,
|
374 |
+
"grad_norm": 2.778372287750244,
|
375 |
+
"learning_rate": 1.5251396648044694e-05,
|
376 |
+
"loss": 0.7158,
|
377 |
"step": 500
|
378 |
},
|
379 |
{
|
380 |
"epoch": 2.560700876095119,
|
381 |
+
"grad_norm": 2.7881994247436523,
|
382 |
+
"learning_rate": 1.5111731843575421e-05,
|
383 |
+
"loss": 0.689,
|
384 |
"step": 510
|
385 |
},
|
386 |
{
|
387 |
"epoch": 2.6107634543178975,
|
388 |
+
"grad_norm": 3.771796226501465,
|
389 |
+
"learning_rate": 1.4972067039106146e-05,
|
390 |
+
"loss": 0.6383,
|
391 |
"step": 520
|
392 |
},
|
393 |
{
|
394 |
"epoch": 2.660826032540676,
|
395 |
+
"grad_norm": 3.580981731414795,
|
396 |
+
"learning_rate": 1.4832402234636874e-05,
|
397 |
+
"loss": 0.6086,
|
398 |
"step": 530
|
399 |
},
|
400 |
{
|
401 |
"epoch": 2.710888610763454,
|
402 |
+
"grad_norm": 2.298293352127075,
|
403 |
+
"learning_rate": 1.4692737430167599e-05,
|
404 |
+
"loss": 0.581,
|
405 |
"step": 540
|
406 |
},
|
407 |
{
|
408 |
"epoch": 2.760951188986233,
|
409 |
+
"grad_norm": 3.767306327819824,
|
410 |
+
"learning_rate": 1.4553072625698325e-05,
|
411 |
+
"loss": 0.5556,
|
412 |
"step": 550
|
413 |
},
|
414 |
{
|
415 |
"epoch": 2.8110137672090114,
|
416 |
+
"grad_norm": 3.712104558944702,
|
417 |
+
"learning_rate": 1.4413407821229052e-05,
|
418 |
+
"loss": 0.5637,
|
419 |
"step": 560
|
420 |
},
|
421 |
{
|
422 |
"epoch": 2.8610763454317896,
|
423 |
+
"grad_norm": 2.500260829925537,
|
424 |
+
"learning_rate": 1.4273743016759778e-05,
|
425 |
+
"loss": 0.525,
|
426 |
"step": 570
|
427 |
},
|
428 |
{
|
429 |
"epoch": 2.911138923654568,
|
430 |
+
"grad_norm": 3.4421885013580322,
|
431 |
+
"learning_rate": 1.4134078212290503e-05,
|
432 |
+
"loss": 0.5235,
|
433 |
"step": 580
|
434 |
},
|
435 |
{
|
436 |
"epoch": 2.9612015018773468,
|
437 |
+
"grad_norm": 3.3740108013153076,
|
438 |
+
"learning_rate": 1.3994413407821231e-05,
|
439 |
+
"loss": 0.4964,
|
440 |
"step": 590
|
441 |
},
|
442 |
{
|
443 |
"epoch": 2.9962453066332917,
|
444 |
+
"eval_accuracy": 0.9727861135628126,
|
445 |
+
"eval_loss": 0.37298157811164856,
|
446 |
+
"eval_runtime": 4.944,
|
447 |
+
"eval_samples_per_second": 1374.987,
|
448 |
+
"eval_steps_per_second": 43.082,
|
449 |
"step": 597
|
450 |
},
|
451 |
{
|
452 |
"epoch": 3.0150187734668337,
|
453 |
+
"grad_norm": 5.472856521606445,
|
454 |
+
"learning_rate": 1.3854748603351957e-05,
|
455 |
+
"loss": 0.5243,
|
456 |
"step": 600
|
457 |
},
|
458 |
{
|
459 |
"epoch": 3.065081351689612,
|
460 |
+
"grad_norm": 3.919224977493286,
|
461 |
+
"learning_rate": 1.3715083798882682e-05,
|
462 |
+
"loss": 0.4697,
|
463 |
"step": 610
|
464 |
},
|
465 |
{
|
466 |
"epoch": 3.1151439299123904,
|
467 |
+
"grad_norm": 2.8592793941497803,
|
468 |
+
"learning_rate": 1.357541899441341e-05,
|
469 |
+
"loss": 0.4581,
|
470 |
"step": 620
|
471 |
},
|
472 |
{
|
473 |
"epoch": 3.165206508135169,
|
474 |
+
"grad_norm": 3.517875909805298,
|
475 |
+
"learning_rate": 1.3435754189944135e-05,
|
476 |
+
"loss": 0.4609,
|
477 |
"step": 630
|
478 |
},
|
479 |
{
|
480 |
"epoch": 3.2152690863579476,
|
481 |
+
"grad_norm": 3.0219643115997314,
|
482 |
+
"learning_rate": 1.3296089385474861e-05,
|
483 |
+
"loss": 0.4251,
|
484 |
"step": 640
|
485 |
},
|
486 |
{
|
487 |
"epoch": 3.2653316645807258,
|
488 |
+
"grad_norm": 4.166126728057861,
|
489 |
+
"learning_rate": 1.3156424581005586e-05,
|
490 |
+
"loss": 0.4026,
|
491 |
"step": 650
|
492 |
},
|
493 |
{
|
494 |
"epoch": 3.3153942428035044,
|
495 |
+
"grad_norm": 2.9157118797302246,
|
496 |
+
"learning_rate": 1.3016759776536314e-05,
|
497 |
+
"loss": 0.4101,
|
498 |
"step": 660
|
499 |
},
|
500 |
{
|
501 |
"epoch": 3.365456821026283,
|
502 |
+
"grad_norm": 2.605928897857666,
|
503 |
+
"learning_rate": 1.287709497206704e-05,
|
504 |
+
"loss": 0.3951,
|
505 |
"step": 670
|
506 |
},
|
507 |
{
|
508 |
"epoch": 3.415519399249061,
|
509 |
+
"grad_norm": 2.282757043838501,
|
510 |
+
"learning_rate": 1.2737430167597766e-05,
|
511 |
+
"loss": 0.3876,
|
512 |
"step": 680
|
513 |
},
|
514 |
{
|
515 |
"epoch": 3.4655819774718397,
|
516 |
+
"grad_norm": 2.8429336547851562,
|
517 |
+
"learning_rate": 1.2597765363128494e-05,
|
518 |
+
"loss": 0.3735,
|
519 |
"step": 690
|
520 |
},
|
521 |
{
|
522 |
"epoch": 3.5156445556946183,
|
523 |
+
"grad_norm": 3.3356730937957764,
|
524 |
+
"learning_rate": 1.2458100558659219e-05,
|
525 |
+
"loss": 0.3873,
|
526 |
"step": 700
|
527 |
},
|
528 |
{
|
529 |
"epoch": 3.565707133917397,
|
530 |
+
"grad_norm": 3.1262238025665283,
|
531 |
+
"learning_rate": 1.2318435754189945e-05,
|
532 |
+
"loss": 0.3572,
|
533 |
"step": 710
|
534 |
},
|
535 |
{
|
536 |
"epoch": 3.615769712140175,
|
537 |
+
"grad_norm": 3.589921712875366,
|
538 |
+
"learning_rate": 1.2178770949720671e-05,
|
539 |
+
"loss": 0.3558,
|
540 |
"step": 720
|
541 |
},
|
542 |
{
|
543 |
"epoch": 3.6658322903629537,
|
544 |
+
"grad_norm": 3.085702896118164,
|
545 |
+
"learning_rate": 1.2039106145251398e-05,
|
546 |
+
"loss": 0.3499,
|
547 |
"step": 730
|
548 |
},
|
549 |
{
|
550 |
"epoch": 3.7158948685857323,
|
551 |
+
"grad_norm": 3.2283403873443604,
|
552 |
+
"learning_rate": 1.1899441340782123e-05,
|
553 |
+
"loss": 0.3428,
|
554 |
"step": 740
|
555 |
},
|
556 |
{
|
557 |
"epoch": 3.7659574468085104,
|
558 |
+
"grad_norm": 2.824887990951538,
|
559 |
+
"learning_rate": 1.175977653631285e-05,
|
560 |
+
"loss": 0.3399,
|
561 |
"step": 750
|
562 |
},
|
563 |
{
|
564 |
"epoch": 3.816020025031289,
|
565 |
+
"grad_norm": 3.595198392868042,
|
566 |
+
"learning_rate": 1.1620111731843577e-05,
|
567 |
+
"loss": 0.3264,
|
568 |
"step": 760
|
569 |
},
|
570 |
{
|
571 |
"epoch": 3.8660826032540676,
|
572 |
+
"grad_norm": 4.541857719421387,
|
573 |
+
"learning_rate": 1.1480446927374302e-05,
|
574 |
+
"loss": 0.3322,
|
575 |
"step": 770
|
576 |
},
|
577 |
{
|
578 |
"epoch": 3.916145181476846,
|
579 |
+
"grad_norm": 3.149718999862671,
|
580 |
+
"learning_rate": 1.134078212290503e-05,
|
581 |
+
"loss": 0.3032,
|
582 |
"step": 780
|
583 |
},
|
584 |
{
|
585 |
"epoch": 3.966207759699625,
|
586 |
+
"grad_norm": 3.3817384243011475,
|
587 |
+
"learning_rate": 1.1201117318435755e-05,
|
588 |
+
"loss": 0.3263,
|
589 |
"step": 790
|
590 |
},
|
591 |
{
|
592 |
"epoch": 3.9962453066332917,
|
593 |
+
"eval_accuracy": 0.9783759929390997,
|
594 |
+
"eval_loss": 0.21611753106117249,
|
595 |
+
"eval_runtime": 4.9235,
|
596 |
+
"eval_samples_per_second": 1380.729,
|
597 |
+
"eval_steps_per_second": 43.262,
|
598 |
"step": 796
|
599 |
},
|
600 |
{
|
601 |
"epoch": 4.020025031289111,
|
602 |
+
"grad_norm": 3.4890668392181396,
|
603 |
+
"learning_rate": 1.1061452513966481e-05,
|
604 |
+
"loss": 0.3262,
|
605 |
"step": 800
|
606 |
},
|
607 |
{
|
608 |
"epoch": 4.07008760951189,
|
609 |
+
"grad_norm": 2.202880382537842,
|
610 |
+
"learning_rate": 1.0921787709497206e-05,
|
611 |
+
"loss": 0.3101,
|
612 |
"step": 810
|
613 |
},
|
614 |
{
|
615 |
"epoch": 4.1201501877346685,
|
616 |
+
"grad_norm": 2.4988725185394287,
|
617 |
+
"learning_rate": 1.0782122905027934e-05,
|
618 |
+
"loss": 0.3022,
|
619 |
"step": 820
|
620 |
},
|
621 |
{
|
622 |
"epoch": 4.170212765957447,
|
623 |
+
"grad_norm": 3.495452404022217,
|
624 |
+
"learning_rate": 1.0642458100558659e-05,
|
625 |
+
"loss": 0.2997,
|
626 |
"step": 830
|
627 |
},
|
628 |
{
|
629 |
"epoch": 4.220275344180226,
|
630 |
+
"grad_norm": 2.8825905323028564,
|
631 |
+
"learning_rate": 1.0502793296089386e-05,
|
632 |
+
"loss": 0.3182,
|
633 |
"step": 840
|
634 |
},
|
635 |
{
|
636 |
"epoch": 4.270337922403003,
|
637 |
+
"grad_norm": 2.3349449634552,
|
638 |
+
"learning_rate": 1.0363128491620114e-05,
|
639 |
+
"loss": 0.2884,
|
640 |
"step": 850
|
641 |
},
|
642 |
{
|
643 |
"epoch": 4.320400500625782,
|
644 |
+
"grad_norm": 3.996652603149414,
|
645 |
+
"learning_rate": 1.0223463687150838e-05,
|
646 |
+
"loss": 0.29,
|
647 |
"step": 860
|
648 |
},
|
649 |
{
|
650 |
"epoch": 4.370463078848561,
|
651 |
+
"grad_norm": 1.9067742824554443,
|
652 |
+
"learning_rate": 1.0083798882681567e-05,
|
653 |
+
"loss": 0.2703,
|
654 |
"step": 870
|
655 |
},
|
656 |
{
|
657 |
"epoch": 4.420525657071339,
|
658 |
+
"grad_norm": 3.410153388977051,
|
659 |
+
"learning_rate": 9.944134078212291e-06,
|
660 |
+
"loss": 0.2864,
|
661 |
"step": 880
|
662 |
},
|
663 |
{
|
664 |
"epoch": 4.470588235294118,
|
665 |
+
"grad_norm": 2.604459285736084,
|
666 |
+
"learning_rate": 9.804469273743018e-06,
|
667 |
+
"loss": 0.2814,
|
668 |
"step": 890
|
669 |
},
|
670 |
{
|
671 |
"epoch": 4.520650813516896,
|
672 |
+
"grad_norm": 1.9956718683242798,
|
673 |
+
"learning_rate": 9.664804469273744e-06,
|
674 |
+
"loss": 0.2501,
|
675 |
"step": 900
|
676 |
},
|
677 |
{
|
678 |
"epoch": 4.570713391739675,
|
679 |
+
"grad_norm": 3.1284875869750977,
|
680 |
+
"learning_rate": 9.52513966480447e-06,
|
681 |
+
"loss": 0.2584,
|
682 |
"step": 910
|
683 |
},
|
684 |
{
|
685 |
"epoch": 4.6207759699624535,
|
686 |
+
"grad_norm": 3.3868837356567383,
|
687 |
+
"learning_rate": 9.385474860335197e-06,
|
688 |
+
"loss": 0.2486,
|
689 |
"step": 920
|
690 |
},
|
691 |
{
|
692 |
"epoch": 4.670838548185231,
|
693 |
+
"grad_norm": 2.5720443725585938,
|
694 |
+
"learning_rate": 9.245810055865922e-06,
|
695 |
+
"loss": 0.2624,
|
696 |
"step": 930
|
697 |
},
|
698 |
{
|
699 |
"epoch": 4.72090112640801,
|
700 |
+
"grad_norm": 3.343977689743042,
|
701 |
+
"learning_rate": 9.106145251396648e-06,
|
702 |
+
"loss": 0.2506,
|
703 |
"step": 940
|
704 |
},
|
705 |
{
|
706 |
"epoch": 4.7709637046307884,
|
707 |
+
"grad_norm": 3.791302442550659,
|
708 |
+
"learning_rate": 8.966480446927375e-06,
|
709 |
+
"loss": 0.2712,
|
710 |
"step": 950
|
711 |
},
|
712 |
{
|
713 |
"epoch": 4.821026282853567,
|
714 |
+
"grad_norm": 3.8650593757629395,
|
715 |
+
"learning_rate": 8.826815642458101e-06,
|
716 |
+
"loss": 0.2508,
|
717 |
"step": 960
|
718 |
},
|
719 |
{
|
720 |
"epoch": 4.871088861076346,
|
721 |
+
"grad_norm": 3.790891408920288,
|
722 |
+
"learning_rate": 8.687150837988828e-06,
|
723 |
+
"loss": 0.2653,
|
724 |
"step": 970
|
725 |
},
|
726 |
{
|
727 |
"epoch": 4.921151439299124,
|
728 |
+
"grad_norm": 3.150134325027466,
|
729 |
+
"learning_rate": 8.547486033519554e-06,
|
730 |
+
"loss": 0.2431,
|
731 |
"step": 980
|
732 |
},
|
733 |
{
|
734 |
"epoch": 4.971214017521902,
|
735 |
+
"grad_norm": 3.999809741973877,
|
736 |
+
"learning_rate": 8.40782122905028e-06,
|
737 |
+
"loss": 0.2512,
|
738 |
"step": 990
|
739 |
},
|
740 |
{
|
741 |
"epoch": 4.996245306633291,
|
742 |
+
"eval_accuracy": 0.979552809649897,
|
743 |
+
"eval_loss": 0.16172775626182556,
|
744 |
+
"eval_runtime": 5.0947,
|
745 |
+
"eval_samples_per_second": 1334.339,
|
746 |
+
"eval_steps_per_second": 41.809,
|
747 |
"step": 995
|
748 |
},
|
749 |
{
|
750 |
"epoch": 5.025031289111389,
|
751 |
+
"grad_norm": 3.607999563217163,
|
752 |
+
"learning_rate": 8.268156424581007e-06,
|
753 |
+
"loss": 0.261,
|
754 |
"step": 1000
|
755 |
},
|
756 |
{
|
757 |
"epoch": 5.075093867334168,
|
758 |
+
"grad_norm": 2.384503126144409,
|
759 |
+
"learning_rate": 8.128491620111732e-06,
|
760 |
+
"loss": 0.2301,
|
761 |
"step": 1010
|
762 |
},
|
763 |
{
|
764 |
"epoch": 5.1251564455569465,
|
765 |
+
"grad_norm": 3.3627846240997314,
|
766 |
+
"learning_rate": 7.988826815642458e-06,
|
767 |
+
"loss": 0.2363,
|
768 |
"step": 1020
|
769 |
},
|
770 |
{
|
771 |
"epoch": 5.175219023779725,
|
772 |
+
"grad_norm": 3.3233959674835205,
|
773 |
+
"learning_rate": 7.849162011173185e-06,
|
774 |
+
"loss": 0.2275,
|
775 |
"step": 1030
|
776 |
},
|
777 |
{
|
778 |
"epoch": 5.225281602002503,
|
779 |
+
"grad_norm": 3.21708083152771,
|
780 |
+
"learning_rate": 7.709497206703911e-06,
|
781 |
+
"loss": 0.2247,
|
782 |
"step": 1040
|
783 |
},
|
784 |
{
|
785 |
"epoch": 5.275344180225281,
|
786 |
+
"grad_norm": 2.4232187271118164,
|
787 |
+
"learning_rate": 7.569832402234637e-06,
|
788 |
+
"loss": 0.2044,
|
789 |
"step": 1050
|
790 |
},
|
791 |
{
|
792 |
"epoch": 5.32540675844806,
|
793 |
+
"grad_norm": 2.9417362213134766,
|
794 |
+
"learning_rate": 7.430167597765364e-06,
|
795 |
+
"loss": 0.2156,
|
796 |
"step": 1060
|
797 |
},
|
798 |
{
|
799 |
"epoch": 5.375469336670839,
|
800 |
+
"grad_norm": 5.781579971313477,
|
801 |
+
"learning_rate": 7.290502793296091e-06,
|
802 |
+
"loss": 0.2479,
|
803 |
"step": 1070
|
804 |
},
|
805 |
{
|
806 |
"epoch": 5.425531914893617,
|
807 |
+
"grad_norm": 2.417659044265747,
|
808 |
+
"learning_rate": 7.150837988826816e-06,
|
809 |
+
"loss": 0.2349,
|
810 |
"step": 1080
|
811 |
},
|
812 |
{
|
813 |
"epoch": 5.475594493116396,
|
814 |
+
"grad_norm": 3.229393482208252,
|
815 |
+
"learning_rate": 7.011173184357543e-06,
|
816 |
+
"loss": 0.218,
|
817 |
"step": 1090
|
818 |
},
|
819 |
{
|
820 |
"epoch": 5.5256570713391735,
|
821 |
+
"grad_norm": 3.5707740783691406,
|
822 |
+
"learning_rate": 6.871508379888268e-06,
|
823 |
+
"loss": 0.2287,
|
824 |
"step": 1100
|
825 |
},
|
826 |
{
|
827 |
"epoch": 5.575719649561952,
|
828 |
+
"grad_norm": 2.99178147315979,
|
829 |
+
"learning_rate": 6.731843575418995e-06,
|
830 |
+
"loss": 0.2474,
|
831 |
"step": 1110
|
832 |
},
|
833 |
{
|
834 |
"epoch": 5.625782227784731,
|
835 |
+
"grad_norm": 4.041738986968994,
|
836 |
+
"learning_rate": 6.592178770949721e-06,
|
837 |
+
"loss": 0.254,
|
838 |
"step": 1120
|
839 |
},
|
840 |
{
|
841 |
"epoch": 5.675844806007509,
|
842 |
+
"grad_norm": 3.7862696647644043,
|
843 |
+
"learning_rate": 6.452513966480447e-06,
|
844 |
+
"loss": 0.2336,
|
845 |
"step": 1130
|
846 |
},
|
847 |
{
|
848 |
"epoch": 5.725907384230288,
|
849 |
+
"grad_norm": 3.772305488586426,
|
850 |
+
"learning_rate": 6.312849162011173e-06,
|
851 |
+
"loss": 0.2366,
|
852 |
"step": 1140
|
853 |
},
|
854 |
{
|
855 |
"epoch": 5.7759699624530665,
|
856 |
+
"grad_norm": 3.721914291381836,
|
857 |
+
"learning_rate": 6.173184357541901e-06,
|
858 |
+
"loss": 0.2056,
|
859 |
"step": 1150
|
860 |
},
|
861 |
{
|
862 |
"epoch": 5.826032540675845,
|
863 |
+
"grad_norm": 3.321104049682617,
|
864 |
+
"learning_rate": 6.033519553072626e-06,
|
865 |
+
"loss": 0.2083,
|
866 |
"step": 1160
|
867 |
},
|
868 |
{
|
869 |
"epoch": 5.876095118898624,
|
870 |
+
"grad_norm": 2.396998405456543,
|
871 |
+
"learning_rate": 5.893854748603353e-06,
|
872 |
+
"loss": 0.2413,
|
873 |
"step": 1170
|
874 |
},
|
875 |
{
|
876 |
"epoch": 5.926157697121401,
|
877 |
+
"grad_norm": 2.6875176429748535,
|
878 |
+
"learning_rate": 5.754189944134078e-06,
|
879 |
+
"loss": 0.2049,
|
880 |
"step": 1180
|
881 |
},
|
882 |
{
|
883 |
"epoch": 5.97622027534418,
|
884 |
+
"grad_norm": 3.377181053161621,
|
885 |
+
"learning_rate": 5.614525139664805e-06,
|
886 |
+
"loss": 0.214,
|
887 |
"step": 1190
|
888 |
},
|
889 |
{
|
890 |
"epoch": 5.996245306633291,
|
891 |
+
"eval_accuracy": 0.9807296263606943,
|
892 |
+
"eval_loss": 0.13626398146152496,
|
893 |
+
"eval_runtime": 4.9535,
|
894 |
+
"eval_samples_per_second": 1372.375,
|
895 |
+
"eval_steps_per_second": 43.0,
|
896 |
"step": 1194
|
897 |
},
|
898 |
{
|
899 |
"epoch": 6.030037546933667,
|
900 |
+
"grad_norm": 2.5127053260803223,
|
901 |
+
"learning_rate": 5.474860335195531e-06,
|
902 |
+
"loss": 0.2374,
|
903 |
"step": 1200
|
904 |
},
|
905 |
{
|
906 |
"epoch": 6.080100125156446,
|
907 |
+
"grad_norm": 2.9415111541748047,
|
908 |
+
"learning_rate": 5.335195530726257e-06,
|
909 |
+
"loss": 0.203,
|
910 |
"step": 1210
|
911 |
},
|
912 |
{
|
913 |
"epoch": 6.130162703379224,
|
914 |
+
"grad_norm": 1.5115219354629517,
|
915 |
+
"learning_rate": 5.195530726256983e-06,
|
916 |
+
"loss": 0.2151,
|
917 |
"step": 1220
|
918 |
},
|
919 |
{
|
920 |
"epoch": 6.180225281602002,
|
921 |
+
"grad_norm": 1.8936338424682617,
|
922 |
+
"learning_rate": 5.055865921787711e-06,
|
923 |
+
"loss": 0.2159,
|
924 |
"step": 1230
|
925 |
},
|
926 |
{
|
927 |
"epoch": 6.230287859824781,
|
928 |
+
"grad_norm": 2.6654956340789795,
|
929 |
+
"learning_rate": 4.916201117318436e-06,
|
930 |
+
"loss": 0.2013,
|
931 |
"step": 1240
|
932 |
},
|
933 |
{
|
934 |
"epoch": 6.280350438047559,
|
935 |
+
"grad_norm": 3.0140798091888428,
|
936 |
+
"learning_rate": 4.776536312849163e-06,
|
937 |
+
"loss": 0.1985,
|
938 |
"step": 1250
|
939 |
},
|
940 |
{
|
941 |
"epoch": 6.330413016270338,
|
942 |
+
"grad_norm": 3.579763412475586,
|
943 |
+
"learning_rate": 4.636871508379888e-06,
|
944 |
+
"loss": 0.207,
|
945 |
"step": 1260
|
946 |
},
|
947 |
{
|
948 |
"epoch": 6.380475594493117,
|
949 |
+
"grad_norm": 2.5355074405670166,
|
950 |
+
"learning_rate": 4.497206703910615e-06,
|
951 |
+
"loss": 0.193,
|
952 |
"step": 1270
|
953 |
},
|
954 |
{
|
955 |
"epoch": 6.430538172715895,
|
956 |
+
"grad_norm": 2.3967621326446533,
|
957 |
+
"learning_rate": 4.357541899441341e-06,
|
958 |
+
"loss": 0.2139,
|
959 |
"step": 1280
|
960 |
},
|
961 |
{
|
962 |
"epoch": 6.480600750938673,
|
963 |
+
"grad_norm": 2.9278903007507324,
|
964 |
+
"learning_rate": 4.217877094972068e-06,
|
965 |
+
"loss": 0.2048,
|
966 |
"step": 1290
|
967 |
},
|
968 |
{
|
969 |
"epoch": 6.5306633291614515,
|
970 |
+
"grad_norm": 5.488080024719238,
|
971 |
+
"learning_rate": 4.078212290502794e-06,
|
972 |
+
"loss": 0.2098,
|
973 |
"step": 1300
|
974 |
},
|
975 |
{
|
976 |
"epoch": 6.58072590738423,
|
977 |
+
"grad_norm": 4.781853199005127,
|
978 |
+
"learning_rate": 3.93854748603352e-06,
|
979 |
+
"loss": 0.208,
|
980 |
"step": 1310
|
981 |
},
|
982 |
{
|
983 |
"epoch": 6.630788485607009,
|
984 |
+
"grad_norm": 2.9733641147613525,
|
985 |
+
"learning_rate": 3.798882681564246e-06,
|
986 |
+
"loss": 0.1923,
|
987 |
"step": 1320
|
988 |
},
|
989 |
{
|
990 |
"epoch": 6.680851063829787,
|
991 |
+
"grad_norm": 2.823633909225464,
|
992 |
+
"learning_rate": 3.6592178770949723e-06,
|
993 |
+
"loss": 0.185,
|
994 |
"step": 1330
|
995 |
},
|
996 |
{
|
997 |
"epoch": 6.730913642052566,
|
998 |
+
"grad_norm": 2.7353272438049316,
|
999 |
+
"learning_rate": 3.5195530726256988e-06,
|
1000 |
+
"loss": 0.1977,
|
1001 |
"step": 1340
|
1002 |
},
|
1003 |
{
|
1004 |
"epoch": 6.7809762202753445,
|
1005 |
+
"grad_norm": 2.489300489425659,
|
1006 |
+
"learning_rate": 3.3798882681564248e-06,
|
1007 |
+
"loss": 0.1999,
|
1008 |
"step": 1350
|
1009 |
},
|
1010 |
{
|
1011 |
"epoch": 6.831038798498122,
|
1012 |
+
"grad_norm": 2.8995063304901123,
|
1013 |
+
"learning_rate": 3.240223463687151e-06,
|
1014 |
+
"loss": 0.2111,
|
1015 |
"step": 1360
|
1016 |
},
|
1017 |
{
|
1018 |
"epoch": 6.881101376720901,
|
1019 |
+
"grad_norm": 2.4604063034057617,
|
1020 |
+
"learning_rate": 3.1005586592178773e-06,
|
1021 |
+
"loss": 0.2043,
|
1022 |
"step": 1370
|
1023 |
},
|
1024 |
{
|
1025 |
"epoch": 6.931163954943679,
|
1026 |
+
"grad_norm": 2.1975669860839844,
|
1027 |
+
"learning_rate": 2.9608938547486037e-06,
|
1028 |
+
"loss": 0.203,
|
1029 |
"step": 1380
|
1030 |
},
|
1031 |
{
|
1032 |
"epoch": 6.981226533166458,
|
1033 |
+
"grad_norm": 2.8632118701934814,
|
1034 |
+
"learning_rate": 2.8212290502793298e-06,
|
1035 |
+
"loss": 0.1826,
|
1036 |
"step": 1390
|
1037 |
},
|
1038 |
{
|
1039 |
"epoch": 6.996245306633291,
|
1040 |
+
"eval_accuracy": 0.9814651368049426,
|
1041 |
+
"eval_loss": 0.12438357621431351,
|
1042 |
+
"eval_runtime": 4.9437,
|
1043 |
+
"eval_samples_per_second": 1375.072,
|
1044 |
+
"eval_steps_per_second": 43.085,
|
1045 |
"step": 1393
|
1046 |
},
|
1047 |
{
|
1048 |
"epoch": 7.035043804755945,
|
1049 |
+
"grad_norm": 2.0548951625823975,
|
1050 |
+
"learning_rate": 2.6815642458100562e-06,
|
1051 |
+
"loss": 0.2089,
|
1052 |
"step": 1400
|
1053 |
},
|
1054 |
{
|
1055 |
"epoch": 7.085106382978723,
|
1056 |
+
"grad_norm": 2.192319631576538,
|
1057 |
+
"learning_rate": 2.5418994413407823e-06,
|
1058 |
+
"loss": 0.1803,
|
1059 |
"step": 1410
|
1060 |
},
|
1061 |
{
|
1062 |
"epoch": 7.135168961201502,
|
1063 |
+
"grad_norm": 3.190675973892212,
|
1064 |
+
"learning_rate": 2.4022346368715087e-06,
|
1065 |
+
"loss": 0.1985,
|
1066 |
"step": 1420
|
1067 |
},
|
1068 |
{
|
1069 |
"epoch": 7.18523153942428,
|
1070 |
+
"grad_norm": 3.319995880126953,
|
1071 |
+
"learning_rate": 2.2625698324022348e-06,
|
1072 |
+
"loss": 0.1815,
|
1073 |
"step": 1430
|
1074 |
},
|
1075 |
{
|
1076 |
"epoch": 7.235294117647059,
|
1077 |
+
"grad_norm": 3.1613974571228027,
|
1078 |
+
"learning_rate": 2.1229050279329612e-06,
|
1079 |
+
"loss": 0.1874,
|
1080 |
"step": 1440
|
1081 |
},
|
1082 |
{
|
1083 |
"epoch": 7.2853566958698375,
|
1084 |
+
"grad_norm": 5.030276298522949,
|
1085 |
+
"learning_rate": 1.9832402234636873e-06,
|
1086 |
+
"loss": 0.2316,
|
1087 |
"step": 1450
|
1088 |
},
|
1089 |
{
|
1090 |
"epoch": 7.335419274092616,
|
1091 |
+
"grad_norm": 4.310720443725586,
|
1092 |
+
"learning_rate": 1.8435754189944135e-06,
|
1093 |
+
"loss": 0.187,
|
1094 |
"step": 1460
|
1095 |
},
|
1096 |
{
|
1097 |
"epoch": 7.385481852315394,
|
1098 |
+
"grad_norm": 3.277670383453369,
|
1099 |
+
"learning_rate": 1.7039106145251397e-06,
|
1100 |
+
"loss": 0.189,
|
1101 |
"step": 1470
|
1102 |
},
|
1103 |
{
|
1104 |
"epoch": 7.435544430538172,
|
1105 |
+
"grad_norm": 2.657017707824707,
|
1106 |
+
"learning_rate": 1.564245810055866e-06,
|
1107 |
+
"loss": 0.1963,
|
1108 |
"step": 1480
|
1109 |
},
|
1110 |
{
|
1111 |
"epoch": 7.485607008760951,
|
1112 |
+
"grad_norm": 1.9806122779846191,
|
1113 |
+
"learning_rate": 1.4245810055865922e-06,
|
1114 |
+
"loss": 0.1736,
|
1115 |
"step": 1490
|
1116 |
},
|
1117 |
{
|
1118 |
"epoch": 7.53566958698373,
|
1119 |
+
"grad_norm": 2.318047285079956,
|
1120 |
+
"learning_rate": 1.2849162011173185e-06,
|
1121 |
+
"loss": 0.192,
|
1122 |
"step": 1500
|
1123 |
},
|
1124 |
{
|
1125 |
"epoch": 7.585732165206508,
|
1126 |
+
"grad_norm": 1.8546510934829712,
|
1127 |
+
"learning_rate": 1.1452513966480447e-06,
|
1128 |
+
"loss": 0.1915,
|
1129 |
"step": 1510
|
1130 |
},
|
1131 |
{
|
1132 |
"epoch": 7.635794743429287,
|
1133 |
+
"grad_norm": 2.3725483417510986,
|
1134 |
+
"learning_rate": 1.005586592178771e-06,
|
1135 |
+
"loss": 0.1924,
|
1136 |
"step": 1520
|
1137 |
},
|
1138 |
{
|
1139 |
"epoch": 7.685857321652065,
|
1140 |
+
"grad_norm": 2.3467774391174316,
|
1141 |
+
"learning_rate": 8.659217877094973e-07,
|
1142 |
+
"loss": 0.1997,
|
1143 |
"step": 1530
|
1144 |
},
|
1145 |
{
|
1146 |
"epoch": 7.735919899874844,
|
1147 |
+
"grad_norm": 2.653728485107422,
|
1148 |
+
"learning_rate": 7.262569832402236e-07,
|
1149 |
+
"loss": 0.2081,
|
1150 |
"step": 1540
|
1151 |
},
|
1152 |
{
|
1153 |
"epoch": 7.785982478097622,
|
1154 |
+
"grad_norm": 3.015143394470215,
|
1155 |
+
"learning_rate": 5.865921787709498e-07,
|
1156 |
+
"loss": 0.2194,
|
1157 |
"step": 1550
|
1158 |
},
|
1159 |
{
|
1160 |
"epoch": 7.8360450563204,
|
1161 |
+
"grad_norm": 1.836916446685791,
|
1162 |
+
"learning_rate": 4.46927374301676e-07,
|
1163 |
+
"loss": 0.2127,
|
1164 |
"step": 1560
|
1165 |
},
|
1166 |
{
|
1167 |
"epoch": 7.886107634543179,
|
1168 |
+
"grad_norm": 1.9544241428375244,
|
1169 |
+
"learning_rate": 3.0726256983240227e-07,
|
1170 |
+
"loss": 0.1995,
|
1171 |
"step": 1570
|
1172 |
},
|
1173 |
{
|
1174 |
"epoch": 7.9361702127659575,
|
1175 |
+
"grad_norm": 1.683449625968933,
|
1176 |
+
"learning_rate": 1.6759776536312851e-07,
|
1177 |
+
"loss": 0.1974,
|
1178 |
"step": 1580
|
1179 |
},
|
1180 |
{
|
1181 |
"epoch": 7.986232790988736,
|
1182 |
+
"grad_norm": 2.348433494567871,
|
1183 |
+
"learning_rate": 2.793296089385475e-08,
|
1184 |
+
"loss": 0.1982,
|
1185 |
"step": 1590
|
1186 |
},
|
1187 |
{
|
1188 |
"epoch": 7.996245306633291,
|
1189 |
+
"eval_accuracy": 0.9814651368049426,
|
1190 |
+
"eval_loss": 0.12064122408628464,
|
1191 |
+
"eval_runtime": 5.5563,
|
1192 |
+
"eval_samples_per_second": 1223.467,
|
1193 |
+
"eval_steps_per_second": 38.335,
|
1194 |
"step": 1592
|
1195 |
},
|
1196 |
{
|
1197 |
"epoch": 7.996245306633291,
|
1198 |
"step": 1592,
|
1199 |
"total_flos": 3.777723239743488e+18,
|
1200 |
+
"train_loss": 0.726146658611058,
|
1201 |
+
"train_runtime": 635.8846,
|
1202 |
+
"train_samples_per_second": 642.808,
|
1203 |
+
"train_steps_per_second": 2.504
|
1204 |
}
|
1205 |
],
|
1206 |
"logging_steps": 10,
|