Commit
•
28aecb0
1
Parent(s):
d800da3
huggingartists
Browse files- README.md +3 -3
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +2 -2
- pytorch_model.bin +2 -2
- rng_state.pth +2 -2
- scheduler.pt +1 -1
- trainer_state.json +475 -489
- training_args.bin +1 -1
README.md
CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
45 |
dataset = load_dataset("huggingartists/eminem")
|
46 |
```
|
47 |
|
48 |
-
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Eminem's lyrics.
|
53 |
|
54 |
-
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/
|
55 |
|
56 |
-
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/
|
57 |
|
58 |
## How to use
|
59 |
|
|
|
45 |
dataset = load_dataset("huggingartists/eminem")
|
46 |
```
|
47 |
|
48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2et9dwlh/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Eminem's lyrics.
|
53 |
|
54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/22e6ddab) for full transparency and reproducibility.
|
55 |
|
56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/22e6ddab/artifacts) is logged and versioned.
|
57 |
|
58 |
## How to use
|
59 |
|
evaluation.txt
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"eval_loss": 0.
|
|
|
1 |
+
{"eval_loss": 0.3614741861820221, "eval_runtime": 29.4359, "eval_samples_per_second": 21.878, "eval_steps_per_second": 2.752, "epoch": 3.0}
|
flax_model.msgpack
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 497764120
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d821cc38972b02052486350836768147963aaeb46cf69d3f41e17bfd52a8d71d
|
3 |
size 497764120
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35bd2c569b3675430dd87749d2e9e88b6c22f4d1277083f1eb323e5a78ccb1b7
|
3 |
+
size 995604017
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:58d89bc75fcb0d32ac9fa6dcb15dddc93183abf0fb0e94b1c0e81fa3acad906e
|
3 |
+
size 510403817
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:049b9bcbcaee102a87fbbdf3ad0f08b27c66519b58ffa7a05152155bdbc8fb60
|
3 |
+
size 14503
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02117a90c598eeedf1944bd81ff7344fb6927e0e3a13871b9e0ceb3095f3a8d3
|
3 |
size 623
|
trainer_state.json
CHANGED
@@ -1,1138 +1,1124 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "output/eminem/checkpoint-
|
4 |
"epoch": 2.0,
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 0.01,
|
12 |
-
"learning_rate":
|
13 |
-
"loss": 0.
|
14 |
"step": 5
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.02,
|
18 |
-
"learning_rate":
|
19 |
-
"loss": 0.
|
20 |
"step": 10
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.03,
|
24 |
-
"learning_rate":
|
25 |
-
"loss": 0.
|
26 |
"step": 15
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.04,
|
30 |
-
"learning_rate":
|
31 |
-
"loss": 0.
|
32 |
"step": 20
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.05,
|
36 |
-
"learning_rate":
|
37 |
-
"loss": 0.
|
38 |
"step": 25
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.07,
|
42 |
-
"learning_rate":
|
43 |
-
"loss": 0.
|
44 |
"step": 30
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.08,
|
48 |
-
"learning_rate":
|
49 |
-
"loss": 0.
|
50 |
"step": 35
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.09,
|
54 |
-
"learning_rate":
|
55 |
-
"loss": 0.
|
56 |
"step": 40
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.1,
|
60 |
-
"learning_rate":
|
61 |
-
"loss": 0.
|
62 |
"step": 45
|
63 |
},
|
64 |
{
|
65 |
"epoch": 0.11,
|
66 |
-
"learning_rate":
|
67 |
-
"loss": 0.
|
68 |
"step": 50
|
69 |
},
|
70 |
{
|
71 |
"epoch": 0.12,
|
72 |
-
"learning_rate": 1.
|
73 |
-
"loss": 0.
|
74 |
"step": 55
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.13,
|
78 |
-
"learning_rate":
|
79 |
-
"loss": 0.
|
80 |
"step": 60
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.14,
|
84 |
-
"learning_rate":
|
85 |
-
"loss": 0.
|
86 |
"step": 65
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.15,
|
90 |
-
"learning_rate": 2.
|
91 |
-
"loss": 0.
|
92 |
"step": 70
|
93 |
},
|
94 |
{
|
95 |
"epoch": 0.16,
|
96 |
-
"learning_rate":
|
97 |
-
"loss": 0.
|
98 |
"step": 75
|
99 |
},
|
100 |
{
|
101 |
"epoch": 0.18,
|
102 |
-
"learning_rate":
|
103 |
-
"loss": 0.
|
104 |
"step": 80
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.19,
|
108 |
-
"learning_rate":
|
109 |
-
"loss": 0.
|
110 |
"step": 85
|
111 |
},
|
112 |
{
|
113 |
"epoch": 0.2,
|
114 |
-
"learning_rate":
|
115 |
-
"loss": 0.
|
116 |
"step": 90
|
117 |
},
|
118 |
{
|
119 |
"epoch": 0.21,
|
120 |
-
"learning_rate":
|
121 |
-
"loss": 0.
|
122 |
"step": 95
|
123 |
},
|
124 |
{
|
125 |
"epoch": 0.22,
|
126 |
-
"learning_rate":
|
127 |
-
"loss": 0.
|
128 |
"step": 100
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.23,
|
132 |
-
"learning_rate":
|
133 |
-
"loss": 0.
|
134 |
"step": 105
|
135 |
},
|
136 |
{
|
137 |
"epoch": 0.24,
|
138 |
-
"learning_rate": 1.
|
139 |
-
"loss": 0.
|
140 |
"step": 110
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.25,
|
144 |
-
"learning_rate": 1.
|
145 |
-
"loss": 0.
|
146 |
"step": 115
|
147 |
},
|
148 |
{
|
149 |
"epoch": 0.26,
|
150 |
-
"learning_rate":
|
151 |
-
"loss": 0.
|
152 |
"step": 120
|
153 |
},
|
154 |
{
|
155 |
"epoch": 0.27,
|
156 |
-
"learning_rate":
|
157 |
-
"loss": 0.
|
158 |
"step": 125
|
159 |
},
|
160 |
{
|
161 |
"epoch": 0.29,
|
162 |
-
"learning_rate":
|
163 |
-
"loss": 0.
|
164 |
"step": 130
|
165 |
},
|
166 |
{
|
167 |
"epoch": 0.3,
|
168 |
-
"learning_rate":
|
169 |
-
"loss": 0.
|
170 |
"step": 135
|
171 |
},
|
172 |
{
|
173 |
"epoch": 0.31,
|
174 |
-
"learning_rate":
|
175 |
-
"loss": 0.
|
176 |
"step": 140
|
177 |
},
|
178 |
{
|
179 |
"epoch": 0.32,
|
180 |
-
"learning_rate":
|
181 |
-
"loss": 0.
|
182 |
"step": 145
|
183 |
},
|
184 |
{
|
185 |
"epoch": 0.33,
|
186 |
-
"learning_rate":
|
187 |
-
"loss": 0.
|
188 |
"step": 150
|
189 |
},
|
190 |
{
|
191 |
"epoch": 0.34,
|
192 |
-
"learning_rate":
|
193 |
-
"loss": 0.
|
194 |
"step": 155
|
195 |
},
|
196 |
{
|
197 |
"epoch": 0.35,
|
198 |
-
"learning_rate":
|
199 |
-
"loss": 0.
|
200 |
"step": 160
|
201 |
},
|
202 |
{
|
203 |
"epoch": 0.36,
|
204 |
-
"learning_rate":
|
205 |
-
"loss": 0.
|
206 |
"step": 165
|
207 |
},
|
208 |
{
|
209 |
"epoch": 0.37,
|
210 |
-
"learning_rate":
|
211 |
-
"loss": 0.
|
212 |
"step": 170
|
213 |
},
|
214 |
{
|
215 |
"epoch": 0.38,
|
216 |
-
"learning_rate":
|
217 |
-
"loss": 0.
|
218 |
"step": 175
|
219 |
},
|
220 |
{
|
221 |
-
"epoch": 0.
|
222 |
-
"learning_rate":
|
223 |
-
"loss": 0.
|
224 |
"step": 180
|
225 |
},
|
226 |
{
|
227 |
"epoch": 0.41,
|
228 |
-
"learning_rate":
|
229 |
-
"loss": 0.
|
230 |
"step": 185
|
231 |
},
|
232 |
{
|
233 |
"epoch": 0.42,
|
234 |
-
"learning_rate":
|
235 |
-
"loss": 0.
|
236 |
"step": 190
|
237 |
},
|
238 |
{
|
239 |
"epoch": 0.43,
|
240 |
-
"learning_rate":
|
241 |
-
"loss": 0.
|
242 |
"step": 195
|
243 |
},
|
244 |
{
|
245 |
"epoch": 0.44,
|
246 |
-
"learning_rate":
|
247 |
-
"loss": 0.
|
248 |
"step": 200
|
249 |
},
|
250 |
{
|
251 |
"epoch": 0.45,
|
252 |
-
"learning_rate":
|
253 |
-
"loss": 0.
|
254 |
"step": 205
|
255 |
},
|
256 |
{
|
257 |
"epoch": 0.46,
|
258 |
-
"learning_rate":
|
259 |
-
"loss": 0.
|
260 |
"step": 210
|
261 |
},
|
262 |
{
|
263 |
"epoch": 0.47,
|
264 |
-
"learning_rate":
|
265 |
-
"loss": 0.
|
266 |
"step": 215
|
267 |
},
|
268 |
{
|
269 |
"epoch": 0.48,
|
270 |
-
"learning_rate":
|
271 |
-
"loss": 0.
|
272 |
"step": 220
|
273 |
},
|
274 |
{
|
275 |
"epoch": 0.49,
|
276 |
-
"learning_rate":
|
277 |
-
"loss": 0.
|
278 |
"step": 225
|
279 |
},
|
280 |
{
|
281 |
-
"epoch": 0.
|
282 |
-
"learning_rate":
|
283 |
-
"loss": 0.
|
284 |
"step": 230
|
285 |
},
|
286 |
{
|
287 |
"epoch": 0.52,
|
288 |
-
"learning_rate":
|
289 |
-
"loss": 0.
|
290 |
"step": 235
|
291 |
},
|
292 |
{
|
293 |
"epoch": 0.53,
|
294 |
-
"learning_rate":
|
295 |
-
"loss": 0.
|
296 |
"step": 240
|
297 |
},
|
298 |
{
|
299 |
"epoch": 0.54,
|
300 |
-
"learning_rate":
|
301 |
-
"loss": 0.
|
302 |
"step": 245
|
303 |
},
|
304 |
{
|
305 |
"epoch": 0.55,
|
306 |
-
"learning_rate":
|
307 |
-
"loss": 0.
|
308 |
"step": 250
|
309 |
},
|
310 |
{
|
311 |
"epoch": 0.56,
|
312 |
-
"learning_rate":
|
313 |
-
"loss": 0.
|
314 |
"step": 255
|
315 |
},
|
316 |
{
|
317 |
"epoch": 0.57,
|
318 |
-
"learning_rate":
|
319 |
-
"loss": 0.
|
320 |
"step": 260
|
321 |
},
|
322 |
{
|
323 |
"epoch": 0.58,
|
324 |
-
"learning_rate":
|
325 |
-
"loss": 0.
|
326 |
"step": 265
|
327 |
},
|
328 |
{
|
329 |
"epoch": 0.59,
|
330 |
-
"learning_rate":
|
331 |
-
"loss": 0.
|
332 |
"step": 270
|
333 |
},
|
334 |
{
|
335 |
"epoch": 0.6,
|
336 |
-
"learning_rate":
|
337 |
-
"loss": 0.
|
338 |
"step": 275
|
339 |
},
|
340 |
{
|
341 |
-
"epoch": 0.
|
342 |
-
"learning_rate":
|
343 |
-
"loss": 0.
|
344 |
"step": 280
|
345 |
},
|
346 |
{
|
347 |
-
"epoch": 0.
|
348 |
-
"learning_rate":
|
349 |
-
"loss": 0.
|
350 |
"step": 285
|
351 |
},
|
352 |
{
|
353 |
"epoch": 0.64,
|
354 |
-
"learning_rate":
|
355 |
-
"loss": 0.
|
356 |
"step": 290
|
357 |
},
|
358 |
{
|
359 |
"epoch": 0.65,
|
360 |
-
"learning_rate":
|
361 |
-
"loss": 0.
|
362 |
"step": 295
|
363 |
},
|
364 |
{
|
365 |
"epoch": 0.66,
|
366 |
-
"learning_rate":
|
367 |
-
"loss": 0.
|
368 |
"step": 300
|
369 |
},
|
370 |
{
|
371 |
"epoch": 0.67,
|
372 |
-
"learning_rate":
|
373 |
-
"loss": 0.
|
374 |
"step": 305
|
375 |
},
|
376 |
{
|
377 |
"epoch": 0.68,
|
378 |
-
"learning_rate":
|
379 |
-
"loss": 0.
|
380 |
"step": 310
|
381 |
},
|
382 |
{
|
383 |
"epoch": 0.69,
|
384 |
-
"learning_rate":
|
385 |
-
"loss": 0.
|
386 |
"step": 315
|
387 |
},
|
388 |
{
|
389 |
"epoch": 0.7,
|
390 |
-
"learning_rate":
|
391 |
-
"loss": 0.
|
392 |
"step": 320
|
393 |
},
|
394 |
{
|
395 |
"epoch": 0.71,
|
396 |
-
"learning_rate":
|
397 |
-
"loss": 0.
|
398 |
"step": 325
|
399 |
},
|
400 |
{
|
401 |
-
"epoch": 0.
|
402 |
-
"learning_rate":
|
403 |
-
"loss": 0.
|
404 |
"step": 330
|
405 |
},
|
406 |
{
|
407 |
-
"epoch": 0.
|
408 |
-
"learning_rate":
|
409 |
-
"loss": 0.
|
410 |
"step": 335
|
411 |
},
|
412 |
{
|
413 |
"epoch": 0.75,
|
414 |
-
"learning_rate":
|
415 |
-
"loss": 0.
|
416 |
"step": 340
|
417 |
},
|
418 |
{
|
419 |
"epoch": 0.76,
|
420 |
-
"learning_rate":
|
421 |
-
"loss": 0.
|
422 |
"step": 345
|
423 |
},
|
424 |
{
|
425 |
"epoch": 0.77,
|
426 |
-
"learning_rate":
|
427 |
-
"loss": 0.
|
428 |
"step": 350
|
429 |
},
|
430 |
{
|
431 |
"epoch": 0.78,
|
432 |
-
"learning_rate":
|
433 |
-
"loss": 0.
|
434 |
"step": 355
|
435 |
},
|
436 |
{
|
437 |
"epoch": 0.79,
|
438 |
-
"learning_rate":
|
439 |
-
"loss": 0.
|
440 |
"step": 360
|
441 |
},
|
442 |
{
|
443 |
"epoch": 0.8,
|
444 |
-
"learning_rate":
|
445 |
-
"loss": 0.
|
446 |
"step": 365
|
447 |
},
|
448 |
{
|
449 |
"epoch": 0.81,
|
450 |
-
"learning_rate":
|
451 |
-
"loss": 0.
|
452 |
"step": 370
|
453 |
},
|
454 |
{
|
455 |
"epoch": 0.82,
|
456 |
-
"learning_rate":
|
457 |
-
"loss": 0.
|
458 |
"step": 375
|
459 |
},
|
460 |
{
|
461 |
-
"epoch": 0.
|
462 |
-
"learning_rate":
|
463 |
-
"loss": 0.
|
464 |
"step": 380
|
465 |
},
|
466 |
{
|
467 |
-
"epoch": 0.
|
468 |
-
"learning_rate": 0.
|
469 |
-
"loss": 0.
|
470 |
"step": 385
|
471 |
},
|
472 |
{
|
473 |
"epoch": 0.86,
|
474 |
-
"learning_rate": 0.
|
475 |
-
"loss": 0.
|
476 |
"step": 390
|
477 |
},
|
478 |
{
|
479 |
"epoch": 0.87,
|
480 |
-
"learning_rate": 0.
|
481 |
-
"loss": 0.
|
482 |
"step": 395
|
483 |
},
|
484 |
{
|
485 |
"epoch": 0.88,
|
486 |
-
"learning_rate": 0.
|
487 |
-
"loss": 0.
|
488 |
"step": 400
|
489 |
},
|
490 |
{
|
491 |
"epoch": 0.89,
|
492 |
-
"learning_rate": 0.
|
493 |
-
"loss": 0.
|
494 |
"step": 405
|
495 |
},
|
496 |
{
|
497 |
"epoch": 0.9,
|
498 |
-
"learning_rate": 0.
|
499 |
-
"loss": 0.
|
500 |
"step": 410
|
501 |
},
|
502 |
{
|
503 |
"epoch": 0.91,
|
504 |
-
"learning_rate": 0.
|
505 |
-
"loss": 0.
|
506 |
"step": 415
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.92,
|
510 |
-
"learning_rate": 0.
|
511 |
-
"loss": 0.
|
512 |
"step": 420
|
513 |
},
|
514 |
{
|
515 |
"epoch": 0.93,
|
516 |
-
"learning_rate": 0.
|
517 |
-
"loss": 0.
|
518 |
"step": 425
|
519 |
},
|
520 |
{
|
521 |
-
"epoch": 0.
|
522 |
-
"learning_rate": 0.
|
523 |
-
"loss": 0.
|
524 |
"step": 430
|
525 |
},
|
526 |
{
|
527 |
-
"epoch": 0.
|
528 |
-
"learning_rate": 0.
|
529 |
-
"loss": 0.
|
530 |
"step": 435
|
531 |
},
|
532 |
{
|
533 |
-
"epoch": 0.
|
534 |
-
"learning_rate": 0.
|
535 |
-
"loss": 0.
|
536 |
"step": 440
|
537 |
},
|
538 |
{
|
539 |
"epoch": 0.98,
|
540 |
-
"learning_rate": 0.
|
541 |
-
"loss": 0.
|
542 |
"step": 445
|
543 |
},
|
544 |
{
|
545 |
"epoch": 0.99,
|
546 |
-
"learning_rate": 0.
|
547 |
-
"loss": 0.
|
548 |
"step": 450
|
549 |
},
|
550 |
{
|
551 |
"epoch": 1.0,
|
552 |
-
"learning_rate": 0.
|
553 |
-
"loss": 0.
|
554 |
"step": 455
|
555 |
},
|
556 |
{
|
557 |
"epoch": 1.0,
|
558 |
-
"eval_loss": 0.
|
559 |
-
"eval_runtime":
|
560 |
-
"eval_samples_per_second":
|
561 |
-
"eval_steps_per_second":
|
562 |
-
"step":
|
563 |
},
|
564 |
{
|
565 |
-
"epoch": 1.
|
566 |
-
"learning_rate": 0.
|
567 |
-
"loss": 0.
|
568 |
"step": 460
|
569 |
},
|
570 |
{
|
571 |
-
"epoch": 1.
|
572 |
-
"learning_rate": 0.
|
573 |
-
"loss": 0.
|
574 |
"step": 465
|
575 |
},
|
576 |
{
|
577 |
-
"epoch": 1.
|
578 |
-
"learning_rate": 0.
|
579 |
-
"loss": 0.
|
580 |
"step": 470
|
581 |
},
|
582 |
{
|
583 |
-
"epoch": 1.
|
584 |
-
"learning_rate": 0.
|
585 |
-
"loss": 0.
|
586 |
"step": 475
|
587 |
},
|
588 |
{
|
589 |
-
"epoch": 1.
|
590 |
-
"learning_rate": 0.
|
591 |
-
"loss": 0.
|
592 |
"step": 480
|
593 |
},
|
594 |
{
|
595 |
"epoch": 1.07,
|
596 |
-
"learning_rate": 0.
|
597 |
-
"loss": 0.
|
598 |
"step": 485
|
599 |
},
|
600 |
{
|
601 |
"epoch": 1.08,
|
602 |
-
"learning_rate": 0.
|
603 |
-
"loss": 0.
|
604 |
"step": 490
|
605 |
},
|
606 |
{
|
607 |
-
"epoch": 1.
|
608 |
-
"learning_rate": 0.
|
609 |
-
"loss": 0.
|
610 |
"step": 495
|
611 |
},
|
612 |
{
|
613 |
-
"epoch": 1.
|
614 |
-
"learning_rate": 0.
|
615 |
-
"loss": 0.
|
616 |
"step": 500
|
617 |
},
|
618 |
{
|
619 |
-
"epoch": 1.
|
620 |
-
"learning_rate":
|
621 |
-
"loss": 0.
|
622 |
"step": 505
|
623 |
},
|
624 |
{
|
625 |
-
"epoch": 1.
|
626 |
-
"learning_rate":
|
627 |
-
"loss": 0.
|
628 |
"step": 510
|
629 |
},
|
630 |
{
|
631 |
-
"epoch": 1.
|
632 |
-
"learning_rate":
|
633 |
-
"loss": 0.
|
634 |
"step": 515
|
635 |
},
|
636 |
{
|
637 |
-
"epoch": 1.
|
638 |
-
"learning_rate":
|
639 |
-
"loss": 0.
|
640 |
"step": 520
|
641 |
},
|
642 |
{
|
643 |
-
"epoch": 1.
|
644 |
-
"learning_rate":
|
645 |
-
"loss": 0.
|
646 |
"step": 525
|
647 |
},
|
648 |
{
|
649 |
-
"epoch": 1.
|
650 |
-
"learning_rate":
|
651 |
-
"loss": 0.
|
652 |
"step": 530
|
653 |
},
|
654 |
{
|
655 |
"epoch": 1.18,
|
656 |
-
"learning_rate":
|
657 |
-
"loss": 0.
|
658 |
"step": 535
|
659 |
},
|
660 |
{
|
661 |
"epoch": 1.19,
|
662 |
-
"learning_rate":
|
663 |
-
"loss": 0.
|
664 |
"step": 540
|
665 |
},
|
666 |
{
|
667 |
-
"epoch": 1.
|
668 |
-
"learning_rate":
|
669 |
-
"loss": 0.
|
670 |
"step": 545
|
671 |
},
|
672 |
{
|
673 |
-
"epoch": 1.
|
674 |
-
"learning_rate":
|
675 |
-
"loss": 0.
|
676 |
"step": 550
|
677 |
},
|
678 |
{
|
679 |
-
"epoch": 1.
|
680 |
-
"learning_rate":
|
681 |
-
"loss": 0.
|
682 |
"step": 555
|
683 |
},
|
684 |
{
|
685 |
-
"epoch": 1.
|
686 |
-
"learning_rate":
|
687 |
-
"loss": 0.
|
688 |
"step": 560
|
689 |
},
|
690 |
{
|
691 |
-
"epoch": 1.
|
692 |
-
"learning_rate":
|
693 |
-
"loss": 0.
|
694 |
"step": 565
|
695 |
},
|
696 |
{
|
697 |
-
"epoch": 1.
|
698 |
-
"learning_rate":
|
699 |
-
"loss": 0.
|
700 |
"step": 570
|
701 |
},
|
702 |
{
|
703 |
-
"epoch": 1.
|
704 |
-
"learning_rate":
|
705 |
-
"loss": 0.
|
706 |
"step": 575
|
707 |
},
|
708 |
{
|
709 |
-
"epoch": 1.
|
710 |
-
"learning_rate":
|
711 |
-
"loss": 0.
|
712 |
"step": 580
|
713 |
},
|
714 |
{
|
715 |
"epoch": 1.29,
|
716 |
-
"learning_rate":
|
717 |
-
"loss": 0.
|
718 |
"step": 585
|
719 |
},
|
720 |
{
|
721 |
-
"epoch": 1.
|
722 |
-
"learning_rate":
|
723 |
-
"loss": 0.
|
724 |
"step": 590
|
725 |
},
|
726 |
{
|
727 |
-
"epoch": 1.
|
728 |
-
"learning_rate":
|
729 |
-
"loss": 0.
|
730 |
"step": 595
|
731 |
},
|
732 |
{
|
733 |
-
"epoch": 1.
|
734 |
-
"learning_rate":
|
735 |
-
"loss": 0.
|
736 |
"step": 600
|
737 |
},
|
738 |
{
|
739 |
-
"epoch": 1.
|
740 |
-
"learning_rate":
|
741 |
-
"loss": 0.
|
742 |
"step": 605
|
743 |
},
|
744 |
{
|
745 |
-
"epoch": 1.
|
746 |
-
"learning_rate":
|
747 |
-
"loss": 0.
|
748 |
"step": 610
|
749 |
},
|
750 |
{
|
751 |
-
"epoch": 1.
|
752 |
-
"learning_rate":
|
753 |
-
"loss": 0.
|
754 |
"step": 615
|
755 |
},
|
756 |
{
|
757 |
-
"epoch": 1.
|
758 |
-
"learning_rate":
|
759 |
-
"loss": 0.
|
760 |
"step": 620
|
761 |
},
|
762 |
{
|
763 |
-
"epoch": 1.
|
764 |
-
"learning_rate":
|
765 |
-
"loss": 0.
|
766 |
"step": 625
|
767 |
},
|
768 |
{
|
769 |
-
"epoch": 1.
|
770 |
-
"learning_rate":
|
771 |
-
"loss": 0.
|
772 |
"step": 630
|
773 |
},
|
774 |
{
|
775 |
"epoch": 1.4,
|
776 |
-
"learning_rate":
|
777 |
-
"loss": 0.
|
778 |
"step": 635
|
779 |
},
|
780 |
{
|
781 |
-
"epoch": 1.
|
782 |
-
"learning_rate":
|
783 |
-
"loss": 0.
|
784 |
"step": 640
|
785 |
},
|
786 |
{
|
787 |
-
"epoch": 1.
|
788 |
-
"learning_rate":
|
789 |
-
"loss": 0.
|
790 |
"step": 645
|
791 |
},
|
792 |
{
|
793 |
-
"epoch": 1.
|
794 |
-
"learning_rate":
|
795 |
-
"loss": 0.
|
796 |
"step": 650
|
797 |
},
|
798 |
{
|
799 |
-
"epoch": 1.
|
800 |
-
"learning_rate":
|
801 |
-
"loss": 0.
|
802 |
"step": 655
|
803 |
},
|
804 |
{
|
805 |
-
"epoch": 1.
|
806 |
-
"learning_rate":
|
807 |
-
"loss": 0.
|
808 |
"step": 660
|
809 |
},
|
810 |
{
|
811 |
-
"epoch": 1.
|
812 |
-
"learning_rate":
|
813 |
-
"loss": 0.
|
814 |
"step": 665
|
815 |
},
|
816 |
{
|
817 |
-
"epoch": 1.
|
818 |
-
"learning_rate":
|
819 |
-
"loss": 0.
|
820 |
"step": 670
|
821 |
},
|
822 |
{
|
823 |
-
"epoch": 1.
|
824 |
-
"learning_rate":
|
825 |
-
"loss": 0.
|
826 |
"step": 675
|
827 |
},
|
828 |
{
|
829 |
-
"epoch": 1.
|
830 |
-
"learning_rate":
|
831 |
-
"loss": 0.
|
832 |
"step": 680
|
833 |
},
|
834 |
{
|
835 |
-
"epoch": 1.
|
836 |
-
"learning_rate":
|
837 |
-
"loss": 0.
|
838 |
"step": 685
|
839 |
},
|
840 |
{
|
841 |
-
"epoch": 1.
|
842 |
-
"learning_rate":
|
843 |
-
"loss": 0.
|
844 |
"step": 690
|
845 |
},
|
846 |
{
|
847 |
-
"epoch": 1.
|
848 |
-
"learning_rate":
|
849 |
-
"loss": 0.
|
850 |
"step": 695
|
851 |
},
|
852 |
{
|
853 |
-
"epoch": 1.
|
854 |
-
"learning_rate":
|
855 |
-
"loss": 0.
|
856 |
"step": 700
|
857 |
},
|
858 |
{
|
859 |
-
"epoch": 1.
|
860 |
-
"learning_rate":
|
861 |
-
"loss": 0.
|
862 |
"step": 705
|
863 |
},
|
864 |
{
|
865 |
-
"epoch": 1.
|
866 |
-
"learning_rate":
|
867 |
-
"loss": 0.
|
868 |
"step": 710
|
869 |
},
|
870 |
{
|
871 |
-
"epoch": 1.
|
872 |
-
"learning_rate":
|
873 |
-
"loss": 0.
|
874 |
"step": 715
|
875 |
},
|
876 |
{
|
877 |
-
"epoch": 1.
|
878 |
-
"learning_rate":
|
879 |
-
"loss": 0.
|
880 |
"step": 720
|
881 |
},
|
882 |
{
|
883 |
-
"epoch": 1.
|
884 |
-
"learning_rate":
|
885 |
-
"loss": 0.
|
886 |
"step": 725
|
887 |
},
|
888 |
{
|
889 |
-
"epoch": 1.
|
890 |
-
"learning_rate":
|
891 |
-
"loss": 0.
|
892 |
"step": 730
|
893 |
},
|
894 |
{
|
895 |
-
"epoch": 1.
|
896 |
-
"learning_rate":
|
897 |
-
"loss": 0.
|
898 |
"step": 735
|
899 |
},
|
900 |
{
|
901 |
-
"epoch": 1.
|
902 |
-
"learning_rate": 5.
|
903 |
-
"loss": 0.
|
904 |
"step": 740
|
905 |
},
|
906 |
{
|
907 |
-
"epoch": 1.
|
908 |
-
"learning_rate":
|
909 |
-
"loss": 0.
|
910 |
"step": 745
|
911 |
},
|
912 |
{
|
913 |
-
"epoch": 1.
|
914 |
-
"learning_rate":
|
915 |
-
"loss": 0.
|
916 |
"step": 750
|
917 |
},
|
918 |
{
|
919 |
-
"epoch": 1.
|
920 |
-
"learning_rate":
|
921 |
-
"loss": 0.
|
922 |
"step": 755
|
923 |
},
|
924 |
{
|
925 |
-
"epoch": 1.
|
926 |
-
"learning_rate":
|
927 |
-
"loss": 0.
|
928 |
"step": 760
|
929 |
},
|
930 |
{
|
931 |
-
"epoch": 1.
|
932 |
-
"learning_rate":
|
933 |
-
"loss": 0.
|
934 |
"step": 765
|
935 |
},
|
936 |
{
|
937 |
-
"epoch": 1.
|
938 |
-
"learning_rate":
|
939 |
-
"loss": 0.
|
940 |
"step": 770
|
941 |
},
|
942 |
{
|
943 |
-
"epoch": 1.
|
944 |
-
"learning_rate":
|
945 |
-
"loss": 0.
|
946 |
"step": 775
|
947 |
},
|
948 |
{
|
949 |
-
"epoch": 1.
|
950 |
-
"learning_rate":
|
951 |
-
"loss": 0.
|
952 |
"step": 780
|
953 |
},
|
954 |
{
|
955 |
-
"epoch": 1.
|
956 |
-
"learning_rate": 3.
|
957 |
-
"loss": 0.
|
958 |
"step": 785
|
959 |
},
|
960 |
{
|
961 |
-
"epoch": 1.
|
962 |
-
"learning_rate":
|
963 |
-
"loss": 0.
|
964 |
"step": 790
|
965 |
},
|
966 |
{
|
967 |
-
"epoch": 1.
|
968 |
-
"learning_rate":
|
969 |
-
"loss": 0.
|
970 |
"step": 795
|
971 |
},
|
972 |
{
|
973 |
-
"epoch": 1.
|
974 |
-
"learning_rate":
|
975 |
-
"loss": 0.
|
976 |
"step": 800
|
977 |
},
|
978 |
{
|
979 |
-
"epoch": 1.
|
980 |
-
"learning_rate":
|
981 |
-
"loss": 0.
|
982 |
"step": 805
|
983 |
},
|
984 |
{
|
985 |
-
"epoch": 1.
|
986 |
-
"learning_rate": 2.
|
987 |
-
"loss": 0.
|
988 |
"step": 810
|
989 |
},
|
990 |
{
|
991 |
-
"epoch": 1.
|
992 |
-
"learning_rate":
|
993 |
-
"loss": 0.
|
994 |
"step": 815
|
995 |
},
|
996 |
{
|
997 |
-
"epoch": 1.
|
998 |
-
"learning_rate":
|
999 |
-
"loss": 0.
|
1000 |
"step": 820
|
1001 |
},
|
1002 |
{
|
1003 |
-
"epoch": 1.
|
1004 |
-
"learning_rate":
|
1005 |
-
"loss": 0.
|
1006 |
"step": 825
|
1007 |
},
|
1008 |
{
|
1009 |
-
"epoch": 1.
|
1010 |
-
"learning_rate": 1.
|
1011 |
-
"loss": 0.
|
1012 |
"step": 830
|
1013 |
},
|
1014 |
{
|
1015 |
-
"epoch": 1.
|
1016 |
-
"learning_rate":
|
1017 |
-
"loss": 0.
|
1018 |
"step": 835
|
1019 |
},
|
1020 |
{
|
1021 |
-
"epoch": 1.
|
1022 |
-
"learning_rate":
|
1023 |
-
"loss": 0.
|
1024 |
"step": 840
|
1025 |
},
|
1026 |
{
|
1027 |
-
"epoch": 1.
|
1028 |
-
"learning_rate":
|
1029 |
-
"loss": 0.
|
1030 |
"step": 845
|
1031 |
},
|
1032 |
{
|
1033 |
-
"epoch": 1.
|
1034 |
-
"learning_rate":
|
1035 |
-
"loss": 0.
|
1036 |
"step": 850
|
1037 |
},
|
1038 |
{
|
1039 |
-
"epoch": 1.
|
1040 |
-
"learning_rate":
|
1041 |
-
"loss": 0.
|
1042 |
"step": 855
|
1043 |
},
|
1044 |
{
|
1045 |
-
"epoch": 1.
|
1046 |
-
"learning_rate":
|
1047 |
-
"loss": 0.
|
1048 |
"step": 860
|
1049 |
},
|
1050 |
{
|
1051 |
-
"epoch": 1.
|
1052 |
-
"learning_rate":
|
1053 |
-
"loss": 0.
|
1054 |
"step": 865
|
1055 |
},
|
1056 |
{
|
1057 |
-
"epoch": 1.
|
1058 |
-
"learning_rate":
|
1059 |
-
"loss": 0.
|
1060 |
"step": 870
|
1061 |
},
|
1062 |
{
|
1063 |
-
"epoch": 1.
|
1064 |
-
"learning_rate":
|
1065 |
-
"loss": 0.
|
1066 |
"step": 875
|
1067 |
},
|
1068 |
{
|
1069 |
-
"epoch": 1.
|
1070 |
-
"learning_rate":
|
1071 |
-
"loss": 0.
|
1072 |
"step": 880
|
1073 |
},
|
1074 |
{
|
1075 |
-
"epoch": 1.
|
1076 |
-
"learning_rate":
|
1077 |
-
"loss": 0.
|
1078 |
"step": 885
|
1079 |
},
|
1080 |
{
|
1081 |
-
"epoch": 1.
|
1082 |
-
"learning_rate":
|
1083 |
-
"loss": 0.
|
1084 |
"step": 890
|
1085 |
},
|
1086 |
{
|
1087 |
-
"epoch": 1.
|
1088 |
-
"learning_rate":
|
1089 |
-
"loss": 0.
|
1090 |
"step": 895
|
1091 |
},
|
1092 |
-
{
|
1093 |
-
"epoch": 1.99,
|
1094 |
-
"learning_rate": 1.6220009567696506e-05,
|
1095 |
-
"loss": 0.7118,
|
1096 |
-
"step": 900
|
1097 |
-
},
|
1098 |
-
{
|
1099 |
-
"epoch": 2.0,
|
1100 |
-
"eval_loss": 0.3754318356513977,
|
1101 |
-
"eval_runtime": 8.7593,
|
1102 |
-
"eval_samples_per_second": 76.147,
|
1103 |
-
"eval_steps_per_second": 9.59,
|
1104 |
-
"step": 904
|
1105 |
-
},
|
1106 |
{
|
1107 |
"epoch": 1.98,
|
1108 |
-
"learning_rate":
|
1109 |
-
"loss": 0.
|
1110 |
-
"step":
|
1111 |
},
|
1112 |
{
|
1113 |
"epoch": 1.99,
|
1114 |
-
"learning_rate":
|
1115 |
-
"loss": 0.
|
1116 |
-
"step":
|
1117 |
},
|
1118 |
{
|
1119 |
"epoch": 2.0,
|
1120 |
-
"learning_rate":
|
1121 |
-
"loss": 0.
|
1122 |
-
"step":
|
1123 |
},
|
1124 |
{
|
1125 |
"epoch": 2.0,
|
1126 |
-
"eval_loss": 0.
|
1127 |
-
"eval_runtime":
|
1128 |
-
"eval_samples_per_second":
|
1129 |
-
"eval_steps_per_second":
|
1130 |
-
"step":
|
1131 |
}
|
1132 |
],
|
1133 |
-
"max_steps":
|
1134 |
"num_train_epochs": 3,
|
1135 |
-
"total_flos":
|
1136 |
"trial_name": null,
|
1137 |
"trial_params": null
|
1138 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.3614741861820221,
|
3 |
+
"best_model_checkpoint": "output/eminem/checkpoint-910",
|
4 |
"epoch": 2.0,
|
5 |
+
"global_step": 910,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
"epoch": 0.01,
|
12 |
+
"learning_rate": 1.0194653534426477e-06,
|
13 |
+
"loss": 0.6762,
|
14 |
"step": 5
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.02,
|
18 |
+
"learning_rate": 6.530415424531046e-07,
|
19 |
+
"loss": 0.6579,
|
20 |
"step": 10
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.03,
|
24 |
+
"learning_rate": 3.675914059099763e-07,
|
25 |
+
"loss": 0.7016,
|
26 |
"step": 15
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.04,
|
30 |
+
"learning_rate": 1.6345512013444254e-07,
|
31 |
+
"loss": 0.6176,
|
32 |
"step": 20
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.05,
|
36 |
+
"learning_rate": 4.087595819659287e-08,
|
37 |
+
"loss": 0.563,
|
38 |
"step": 25
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.07,
|
42 |
+
"learning_rate": 0.0,
|
43 |
+
"loss": 0.689,
|
44 |
"step": 30
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.08,
|
48 |
+
"learning_rate": 4.0875958196577634e-08,
|
49 |
+
"loss": 0.6532,
|
50 |
"step": 35
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.09,
|
54 |
+
"learning_rate": 1.634551201344197e-07,
|
55 |
+
"loss": 0.5922,
|
56 |
"step": 40
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.1,
|
60 |
+
"learning_rate": 3.6759140590974026e-07,
|
61 |
+
"loss": 0.6565,
|
62 |
"step": 45
|
63 |
},
|
64 |
{
|
65 |
"epoch": 0.11,
|
66 |
+
"learning_rate": 6.530415424530588e-07,
|
67 |
+
"loss": 0.7043,
|
68 |
"step": 50
|
69 |
},
|
70 |
{
|
71 |
"epoch": 0.12,
|
72 |
+
"learning_rate": 1.0194653534425943e-06,
|
73 |
+
"loss": 0.597,
|
74 |
"step": 55
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.13,
|
78 |
+
"learning_rate": 1.4664261646975495e-06,
|
79 |
+
"loss": 0.7246,
|
80 |
"step": 60
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.14,
|
84 |
+
"learning_rate": 1.993391324572832e-06,
|
85 |
+
"loss": 0.639,
|
86 |
"step": 65
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.15,
|
90 |
+
"learning_rate": 2.5997328387288936e-06,
|
91 |
+
"loss": 0.6834,
|
92 |
"step": 70
|
93 |
},
|
94 |
{
|
95 |
"epoch": 0.16,
|
96 |
+
"learning_rate": 3.2847281185253694e-06,
|
97 |
+
"loss": 0.6745,
|
98 |
"step": 75
|
99 |
},
|
100 |
{
|
101 |
"epoch": 0.18,
|
102 |
+
"learning_rate": 4.0475608421405796e-06,
|
103 |
+
"loss": 0.5851,
|
104 |
"step": 80
|
105 |
},
|
106 |
{
|
107 |
"epoch": 0.19,
|
108 |
+
"learning_rate": 4.887321927404397e-06,
|
109 |
+
"loss": 0.6805,
|
110 |
"step": 85
|
111 |
},
|
112 |
{
|
113 |
"epoch": 0.2,
|
114 |
+
"learning_rate": 5.803010615159864e-06,
|
115 |
+
"loss": 0.6536,
|
116 |
"step": 90
|
117 |
},
|
118 |
{
|
119 |
"epoch": 0.21,
|
120 |
+
"learning_rate": 6.793535661893871e-06,
|
121 |
+
"loss": 0.6119,
|
122 |
"step": 95
|
123 |
},
|
124 |
{
|
125 |
"epoch": 0.22,
|
126 |
+
"learning_rate": 7.857716640189427e-06,
|
127 |
+
"loss": 0.6954,
|
128 |
"step": 100
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.23,
|
132 |
+
"learning_rate": 8.994285345464919e-06,
|
133 |
+
"loss": 0.7077,
|
134 |
"step": 105
|
135 |
},
|
136 |
{
|
137 |
"epoch": 0.24,
|
138 |
+
"learning_rate": 1.0201887307313696e-05,
|
139 |
+
"loss": 0.6852,
|
140 |
"step": 110
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.25,
|
144 |
+
"learning_rate": 1.147908340365762e-05,
|
145 |
+
"loss": 0.6743,
|
146 |
"step": 115
|
147 |
},
|
148 |
{
|
149 |
"epoch": 0.26,
|
150 |
+
"learning_rate": 1.2824351575772418e-05,
|
151 |
+
"loss": 0.7093,
|
152 |
"step": 120
|
153 |
},
|
154 |
{
|
155 |
"epoch": 0.27,
|
156 |
+
"learning_rate": 1.4236088642155879e-05,
|
157 |
+
"loss": 0.696,
|
158 |
"step": 125
|
159 |
},
|
160 |
{
|
161 |
"epoch": 0.29,
|
162 |
+
"learning_rate": 1.5712612209063624e-05,
|
163 |
+
"loss": 0.6607,
|
164 |
"step": 130
|
165 |
},
|
166 |
{
|
167 |
"epoch": 0.3,
|
168 |
+
"learning_rate": 1.7252162675462687e-05,
|
169 |
+
"loss": 0.5847,
|
170 |
"step": 135
|
171 |
},
|
172 |
{
|
173 |
"epoch": 0.31,
|
174 |
+
"learning_rate": 1.8852905329964338e-05,
|
175 |
+
"loss": 0.6638,
|
176 |
"step": 140
|
177 |
},
|
178 |
{
|
179 |
"epoch": 0.32,
|
180 |
+
"learning_rate": 2.051293253729783e-05,
|
181 |
+
"loss": 0.6221,
|
182 |
"step": 145
|
183 |
},
|
184 |
{
|
185 |
"epoch": 0.33,
|
186 |
+
"learning_rate": 2.2230266011669234e-05,
|
187 |
+
"loss": 0.5733,
|
188 |
"step": 150
|
189 |
},
|
190 |
{
|
191 |
"epoch": 0.34,
|
192 |
+
"learning_rate": 2.4002859174324688e-05,
|
193 |
+
"loss": 0.6478,
|
194 |
"step": 155
|
195 |
},
|
196 |
{
|
197 |
"epoch": 0.35,
|
198 |
+
"learning_rate": 2.5828599592491126e-05,
|
199 |
+
"loss": 0.5237,
|
200 |
"step": 160
|
201 |
},
|
202 |
{
|
203 |
"epoch": 0.36,
|
204 |
+
"learning_rate": 2.770531149681108e-05,
|
205 |
+
"loss": 0.6559,
|
206 |
"step": 165
|
207 |
},
|
208 |
{
|
209 |
"epoch": 0.37,
|
210 |
+
"learning_rate": 2.9630758374242324e-05,
|
211 |
+
"loss": 0.6755,
|
212 |
"step": 170
|
213 |
},
|
214 |
{
|
215 |
"epoch": 0.38,
|
216 |
+
"learning_rate": 3.1602645633354207e-05,
|
217 |
+
"loss": 0.6611,
|
218 |
"step": 175
|
219 |
},
|
220 |
{
|
221 |
+
"epoch": 0.4,
|
222 |
+
"learning_rate": 3.3618623338835595e-05,
|
223 |
+
"loss": 0.651,
|
224 |
"step": 180
|
225 |
},
|
226 |
{
|
227 |
"epoch": 0.41,
|
228 |
+
"learning_rate": 3.5676289011958925e-05,
|
229 |
+
"loss": 0.7284,
|
230 |
"step": 185
|
231 |
},
|
232 |
{
|
233 |
"epoch": 0.42,
|
234 |
+
"learning_rate": 3.7773190493652644e-05,
|
235 |
+
"loss": 0.6301,
|
236 |
"step": 190
|
237 |
},
|
238 |
{
|
239 |
"epoch": 0.43,
|
240 |
+
"learning_rate": 3.990682886679578e-05,
|
241 |
+
"loss": 0.6188,
|
242 |
"step": 195
|
243 |
},
|
244 |
{
|
245 |
"epoch": 0.44,
|
246 |
+
"learning_rate": 4.2074661434217846e-05,
|
247 |
+
"loss": 0.586,
|
248 |
"step": 200
|
249 |
},
|
250 |
{
|
251 |
"epoch": 0.45,
|
252 |
+
"learning_rate": 4.427410474888269e-05,
|
253 |
+
"loss": 0.6809,
|
254 |
"step": 205
|
255 |
},
|
256 |
{
|
257 |
"epoch": 0.46,
|
258 |
+
"learning_rate": 4.650253769262196e-05,
|
259 |
+
"loss": 0.6343,
|
260 |
"step": 210
|
261 |
},
|
262 |
{
|
263 |
"epoch": 0.47,
|
264 |
+
"learning_rate": 4.875730459979135e-05,
|
265 |
+
"loss": 0.7333,
|
266 |
"step": 215
|
267 |
},
|
268 |
{
|
269 |
"epoch": 0.48,
|
270 |
+
"learning_rate": 5.103571842205178e-05,
|
271 |
+
"loss": 0.6158,
|
272 |
"step": 220
|
273 |
},
|
274 |
{
|
275 |
"epoch": 0.49,
|
276 |
+
"learning_rate": 5.3335063930595955e-05,
|
277 |
+
"loss": 0.6216,
|
278 |
"step": 225
|
279 |
},
|
280 |
{
|
281 |
+
"epoch": 0.51,
|
282 |
+
"learning_rate": 5.565260095192864e-05,
|
283 |
+
"loss": 0.7031,
|
284 |
"step": 230
|
285 |
},
|
286 |
{
|
287 |
"epoch": 0.52,
|
288 |
+
"learning_rate": 5.7985567633386964e-05,
|
289 |
+
"loss": 0.7186,
|
290 |
"step": 235
|
291 |
},
|
292 |
{
|
293 |
"epoch": 0.53,
|
294 |
+
"learning_rate": 6.033118373448471e-05,
|
295 |
+
"loss": 0.6841,
|
296 |
"step": 240
|
297 |
},
|
298 |
{
|
299 |
"epoch": 0.54,
|
300 |
+
"learning_rate": 6.268665394018899e-05,
|
301 |
+
"loss": 0.6929,
|
302 |
"step": 245
|
303 |
},
|
304 |
{
|
305 |
"epoch": 0.55,
|
306 |
+
"learning_rate": 6.504917119214232e-05,
|
307 |
+
"loss": 0.8001,
|
308 |
"step": 250
|
309 |
},
|
310 |
{
|
311 |
"epoch": 0.56,
|
312 |
+
"learning_rate": 6.741592003389098e-05,
|
313 |
+
"loss": 0.6497,
|
314 |
"step": 255
|
315 |
},
|
316 |
{
|
317 |
"epoch": 0.57,
|
318 |
+
"learning_rate": 6.978407996610794e-05,
|
319 |
+
"loss": 0.6001,
|
320 |
"step": 260
|
321 |
},
|
322 |
{
|
323 |
"epoch": 0.58,
|
324 |
+
"learning_rate": 7.21508288078566e-05,
|
325 |
+
"loss": 0.7792,
|
326 |
"step": 265
|
327 |
},
|
328 |
{
|
329 |
"epoch": 0.59,
|
330 |
+
"learning_rate": 7.451334605980994e-05,
|
331 |
+
"loss": 0.6229,
|
332 |
"step": 270
|
333 |
},
|
334 |
{
|
335 |
"epoch": 0.6,
|
336 |
+
"learning_rate": 7.686881626551423e-05,
|
337 |
+
"loss": 0.6103,
|
338 |
"step": 275
|
339 |
},
|
340 |
{
|
341 |
+
"epoch": 0.62,
|
342 |
+
"learning_rate": 7.921443236661197e-05,
|
343 |
+
"loss": 0.6509,
|
344 |
"step": 280
|
345 |
},
|
346 |
{
|
347 |
+
"epoch": 0.63,
|
348 |
+
"learning_rate": 8.15473990480703e-05,
|
349 |
+
"loss": 0.7008,
|
350 |
"step": 285
|
351 |
},
|
352 |
{
|
353 |
"epoch": 0.64,
|
354 |
+
"learning_rate": 8.3864936069403e-05,
|
355 |
+
"loss": 0.6919,
|
356 |
"step": 290
|
357 |
},
|
358 |
{
|
359 |
"epoch": 0.65,
|
360 |
+
"learning_rate": 8.616428157794718e-05,
|
361 |
+
"loss": 0.7347,
|
362 |
"step": 295
|
363 |
},
|
364 |
{
|
365 |
"epoch": 0.66,
|
366 |
+
"learning_rate": 8.844269540020762e-05,
|
367 |
+
"loss": 0.7356,
|
368 |
"step": 300
|
369 |
},
|
370 |
{
|
371 |
"epoch": 0.67,
|
372 |
+
"learning_rate": 9.069746230737702e-05,
|
373 |
+
"loss": 0.7,
|
374 |
"step": 305
|
375 |
},
|
376 |
{
|
377 |
"epoch": 0.68,
|
378 |
+
"learning_rate": 9.29258952511163e-05,
|
379 |
+
"loss": 0.6359,
|
380 |
"step": 310
|
381 |
},
|
382 |
{
|
383 |
"epoch": 0.69,
|
384 |
+
"learning_rate": 9.512533856578116e-05,
|
385 |
+
"loss": 0.7165,
|
386 |
"step": 315
|
387 |
},
|
388 |
{
|
389 |
"epoch": 0.7,
|
390 |
+
"learning_rate": 9.729317113320324e-05,
|
391 |
+
"loss": 0.7637,
|
392 |
"step": 320
|
393 |
},
|
394 |
{
|
395 |
"epoch": 0.71,
|
396 |
+
"learning_rate": 9.942680950634639e-05,
|
397 |
+
"loss": 0.6223,
|
398 |
"step": 325
|
399 |
},
|
400 |
{
|
401 |
+
"epoch": 0.73,
|
402 |
+
"learning_rate": 0.00010152371098804014,
|
403 |
+
"loss": 0.5555,
|
404 |
"step": 330
|
405 |
},
|
406 |
{
|
407 |
+
"epoch": 0.74,
|
408 |
+
"learning_rate": 0.00010358137666116348,
|
409 |
+
"loss": 0.7219,
|
410 |
"step": 335
|
411 |
},
|
412 |
{
|
413 |
"epoch": 0.75,
|
414 |
+
"learning_rate": 0.00010559735436664489,
|
415 |
+
"loss": 0.7501,
|
416 |
"step": 340
|
417 |
},
|
418 |
{
|
419 |
"epoch": 0.76,
|
420 |
+
"learning_rate": 0.0001075692416257568,
|
421 |
+
"loss": 0.7454,
|
422 |
"step": 345
|
423 |
},
|
424 |
{
|
425 |
"epoch": 0.77,
|
426 |
+
"learning_rate": 0.00010949468850318805,
|
427 |
+
"loss": 0.6956,
|
428 |
"step": 350
|
429 |
},
|
430 |
{
|
431 |
"epoch": 0.78,
|
432 |
+
"learning_rate": 0.00011137140040750957,
|
433 |
+
"loss": 0.7294,
|
434 |
"step": 355
|
435 |
},
|
436 |
{
|
437 |
"epoch": 0.79,
|
438 |
+
"learning_rate": 0.00011319714082567451,
|
439 |
+
"loss": 0.7416,
|
440 |
"step": 360
|
441 |
},
|
442 |
{
|
443 |
"epoch": 0.8,
|
444 |
+
"learning_rate": 0.00011496973398832998,
|
445 |
+
"loss": 0.6926,
|
446 |
"step": 365
|
447 |
},
|
448 |
{
|
449 |
"epoch": 0.81,
|
450 |
+
"learning_rate": 0.00011668706746270142,
|
451 |
+
"loss": 0.7841,
|
452 |
"step": 370
|
453 |
},
|
454 |
{
|
455 |
"epoch": 0.82,
|
456 |
+
"learning_rate": 0.00011834709467003491,
|
457 |
+
"loss": 0.7603,
|
458 |
"step": 375
|
459 |
},
|
460 |
{
|
461 |
+
"epoch": 0.84,
|
462 |
+
"learning_rate": 0.00011994783732453659,
|
463 |
+
"loss": 0.7249,
|
464 |
"step": 380
|
465 |
},
|
466 |
{
|
467 |
+
"epoch": 0.85,
|
468 |
+
"learning_rate": 0.0001214873877909357,
|
469 |
+
"loss": 0.7231,
|
470 |
"step": 385
|
471 |
},
|
472 |
{
|
473 |
"epoch": 0.86,
|
474 |
+
"learning_rate": 0.00012296391135784465,
|
475 |
+
"loss": 0.7101,
|
476 |
"step": 390
|
477 |
},
|
478 |
{
|
479 |
"epoch": 0.87,
|
480 |
+
"learning_rate": 0.00012437564842422694,
|
481 |
+
"loss": 0.7521,
|
482 |
"step": 395
|
483 |
},
|
484 |
{
|
485 |
"epoch": 0.88,
|
486 |
+
"learning_rate": 0.00012572091659634178,
|
487 |
+
"loss": 0.6694,
|
488 |
"step": 400
|
489 |
},
|
490 |
{
|
491 |
"epoch": 0.89,
|
492 |
+
"learning_rate": 0.00012699811269268675,
|
493 |
+
"loss": 0.6763,
|
494 |
"step": 405
|
495 |
},
|
496 |
{
|
497 |
"epoch": 0.9,
|
498 |
+
"learning_rate": 0.00012820571465453455,
|
499 |
+
"loss": 0.7881,
|
500 |
"step": 410
|
501 |
},
|
502 |
{
|
503 |
"epoch": 0.91,
|
504 |
+
"learning_rate": 0.00012934228335981007,
|
505 |
+
"loss": 0.7906,
|
506 |
"step": 415
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.92,
|
510 |
+
"learning_rate": 0.00013040646433810568,
|
511 |
+
"loss": 0.7662,
|
512 |
"step": 420
|
513 |
},
|
514 |
{
|
515 |
"epoch": 0.93,
|
516 |
+
"learning_rate": 0.00013139698938483972,
|
517 |
+
"loss": 0.7165,
|
518 |
"step": 425
|
519 |
},
|
520 |
{
|
521 |
+
"epoch": 0.95,
|
522 |
+
"learning_rate": 0.00013231267807259521,
|
523 |
+
"loss": 0.7159,
|
524 |
"step": 430
|
525 |
},
|
526 |
{
|
527 |
+
"epoch": 0.96,
|
528 |
+
"learning_rate": 0.00013315243915785907,
|
529 |
+
"loss": 0.7796,
|
530 |
"step": 435
|
531 |
},
|
532 |
{
|
533 |
+
"epoch": 0.97,
|
534 |
+
"learning_rate": 0.0001339152718814749,
|
535 |
+
"loss": 0.8222,
|
536 |
"step": 440
|
537 |
},
|
538 |
{
|
539 |
"epoch": 0.98,
|
540 |
+
"learning_rate": 0.0001346002671612708,
|
541 |
+
"loss": 0.7428,
|
542 |
"step": 445
|
543 |
},
|
544 |
{
|
545 |
"epoch": 0.99,
|
546 |
+
"learning_rate": 0.00013520660867542692,
|
547 |
+
"loss": 0.7597,
|
548 |
"step": 450
|
549 |
},
|
550 |
{
|
551 |
"epoch": 1.0,
|
552 |
+
"learning_rate": 0.00013573357383530262,
|
553 |
+
"loss": 0.8769,
|
554 |
"step": 455
|
555 |
},
|
556 |
{
|
557 |
"epoch": 1.0,
|
558 |
+
"eval_loss": 0.36263224482536316,
|
559 |
+
"eval_runtime": 29.2619,
|
560 |
+
"eval_samples_per_second": 22.008,
|
561 |
+
"eval_steps_per_second": 2.768,
|
562 |
+
"step": 455
|
563 |
},
|
564 |
{
|
565 |
+
"epoch": 1.01,
|
566 |
+
"learning_rate": 0.0001361805346465572,
|
567 |
+
"loss": 0.6009,
|
568 |
"step": 460
|
569 |
},
|
570 |
{
|
571 |
+
"epoch": 1.02,
|
572 |
+
"learning_rate": 0.00013654695845754679,
|
573 |
+
"loss": 0.6886,
|
574 |
"step": 465
|
575 |
},
|
576 |
{
|
577 |
+
"epoch": 1.03,
|
578 |
+
"learning_rate": 0.00013683240859409016,
|
579 |
+
"loss": 0.648,
|
580 |
"step": 470
|
581 |
},
|
582 |
{
|
583 |
+
"epoch": 1.04,
|
584 |
+
"learning_rate": 0.0001370365448798655,
|
585 |
+
"loss": 0.6639,
|
586 |
"step": 475
|
587 |
},
|
588 |
{
|
589 |
+
"epoch": 1.05,
|
590 |
+
"learning_rate": 0.00013715912404180336,
|
591 |
+
"loss": 0.6411,
|
592 |
"step": 480
|
593 |
},
|
594 |
{
|
595 |
"epoch": 1.07,
|
596 |
+
"learning_rate": 0.0001372,
|
597 |
+
"loss": 0.5669,
|
598 |
"step": 485
|
599 |
},
|
600 |
{
|
601 |
"epoch": 1.08,
|
602 |
+
"learning_rate": 0.00013715912404180345,
|
603 |
+
"loss": 0.7041,
|
604 |
"step": 490
|
605 |
},
|
606 |
{
|
607 |
+
"epoch": 1.09,
|
608 |
+
"learning_rate": 0.00013703654487986564,
|
609 |
+
"loss": 0.7196,
|
610 |
"step": 495
|
611 |
},
|
612 |
{
|
613 |
+
"epoch": 1.1,
|
614 |
+
"learning_rate": 0.00013683240859409013,
|
615 |
+
"loss": 0.6656,
|
616 |
"step": 500
|
617 |
},
|
618 |
{
|
619 |
+
"epoch": 1.11,
|
620 |
+
"learning_rate": 0.00013654695845754679,
|
621 |
+
"loss": 0.7371,
|
622 |
"step": 505
|
623 |
},
|
624 |
{
|
625 |
+
"epoch": 1.12,
|
626 |
+
"learning_rate": 0.00013618053464655754,
|
627 |
+
"loss": 0.6728,
|
628 |
"step": 510
|
629 |
},
|
630 |
{
|
631 |
+
"epoch": 1.13,
|
632 |
+
"learning_rate": 0.0001357335738353026,
|
633 |
+
"loss": 0.663,
|
634 |
"step": 515
|
635 |
},
|
636 |
{
|
637 |
+
"epoch": 1.14,
|
638 |
+
"learning_rate": 0.00013520660867542687,
|
639 |
+
"loss": 0.5962,
|
640 |
"step": 520
|
641 |
},
|
642 |
{
|
643 |
+
"epoch": 1.15,
|
644 |
+
"learning_rate": 0.0001346002671612713,
|
645 |
+
"loss": 0.6391,
|
646 |
"step": 525
|
647 |
},
|
648 |
{
|
649 |
+
"epoch": 1.16,
|
650 |
+
"learning_rate": 0.00013391527188147485,
|
651 |
+
"loss": 0.7195,
|
652 |
"step": 530
|
653 |
},
|
654 |
{
|
655 |
"epoch": 1.18,
|
656 |
+
"learning_rate": 0.00013315243915785902,
|
657 |
+
"loss": 0.7751,
|
658 |
"step": 535
|
659 |
},
|
660 |
{
|
661 |
"epoch": 1.19,
|
662 |
+
"learning_rate": 0.0001323126780725959,
|
663 |
+
"loss": 0.6413,
|
664 |
"step": 540
|
665 |
},
|
666 |
{
|
667 |
+
"epoch": 1.2,
|
668 |
+
"learning_rate": 0.00013139698938484045,
|
669 |
+
"loss": 0.7913,
|
670 |
"step": 545
|
671 |
},
|
672 |
{
|
673 |
+
"epoch": 1.21,
|
674 |
+
"learning_rate": 0.0001304064643381056,
|
675 |
+
"loss": 0.7761,
|
676 |
"step": 550
|
677 |
},
|
678 |
{
|
679 |
+
"epoch": 1.22,
|
680 |
+
"learning_rate": 0.00012934228335981002,
|
681 |
+
"loss": 0.6472,
|
682 |
"step": 555
|
683 |
},
|
684 |
{
|
685 |
+
"epoch": 1.23,
|
686 |
+
"learning_rate": 0.00012820571465453544,
|
687 |
+
"loss": 0.6222,
|
688 |
"step": 560
|
689 |
},
|
690 |
{
|
691 |
+
"epoch": 1.24,
|
692 |
+
"learning_rate": 0.0001269981126926867,
|
693 |
+
"loss": 0.6736,
|
694 |
"step": 565
|
695 |
},
|
696 |
{
|
697 |
+
"epoch": 1.25,
|
698 |
+
"learning_rate": 0.00012572091659634172,
|
699 |
+
"loss": 0.73,
|
700 |
"step": 570
|
701 |
},
|
702 |
{
|
703 |
+
"epoch": 1.26,
|
704 |
+
"learning_rate": 0.000124375648424228,
|
705 |
+
"loss": 0.6426,
|
706 |
"step": 575
|
707 |
},
|
708 |
{
|
709 |
+
"epoch": 1.27,
|
710 |
+
"learning_rate": 0.00012296391135784457,
|
711 |
+
"loss": 0.7043,
|
712 |
"step": 580
|
713 |
},
|
714 |
{
|
715 |
"epoch": 1.29,
|
716 |
+
"learning_rate": 0.00012148738779093562,
|
717 |
+
"loss": 0.7227,
|
718 |
"step": 585
|
719 |
},
|
720 |
{
|
721 |
+
"epoch": 1.3,
|
722 |
+
"learning_rate": 0.00011994783732453781,
|
723 |
+
"loss": 0.7366,
|
724 |
"step": 590
|
725 |
},
|
726 |
{
|
727 |
+
"epoch": 1.31,
|
728 |
+
"learning_rate": 0.00011834709467003617,
|
729 |
+
"loss": 0.7144,
|
730 |
"step": 595
|
731 |
},
|
732 |
{
|
733 |
+
"epoch": 1.32,
|
734 |
+
"learning_rate": 0.00011668706746270132,
|
735 |
+
"loss": 0.6817,
|
736 |
"step": 600
|
737 |
},
|
738 |
{
|
739 |
+
"epoch": 1.33,
|
740 |
+
"learning_rate": 0.00011496973398833133,
|
741 |
+
"loss": 0.6274,
|
742 |
"step": 605
|
743 |
},
|
744 |
{
|
745 |
+
"epoch": 1.34,
|
746 |
+
"learning_rate": 0.00011319714082567588,
|
747 |
+
"loss": 0.6754,
|
748 |
"step": 610
|
749 |
},
|
750 |
{
|
751 |
+
"epoch": 1.35,
|
752 |
+
"learning_rate": 0.00011137140040750945,
|
753 |
+
"loss": 0.6134,
|
754 |
"step": 615
|
755 |
},
|
756 |
{
|
757 |
+
"epoch": 1.36,
|
758 |
+
"learning_rate": 0.00010949468850318951,
|
759 |
+
"loss": 0.698,
|
760 |
"step": 620
|
761 |
},
|
762 |
{
|
763 |
+
"epoch": 1.37,
|
764 |
+
"learning_rate": 0.00010756924162575829,
|
765 |
+
"loss": 0.8638,
|
766 |
"step": 625
|
767 |
},
|
768 |
{
|
769 |
+
"epoch": 1.38,
|
770 |
+
"learning_rate": 0.00010559735436664478,
|
771 |
+
"loss": 0.7123,
|
772 |
"step": 630
|
773 |
},
|
774 |
{
|
775 |
"epoch": 1.4,
|
776 |
+
"learning_rate": 0.00010358137666116336,
|
777 |
+
"loss": 0.666,
|
778 |
"step": 635
|
779 |
},
|
780 |
{
|
781 |
+
"epoch": 1.41,
|
782 |
+
"learning_rate": 0.00010152371098804174,
|
783 |
+
"loss": 0.6977,
|
784 |
"step": 640
|
785 |
},
|
786 |
{
|
787 |
+
"epoch": 1.42,
|
788 |
+
"learning_rate": 9.942680950634801e-05,
|
789 |
+
"loss": 0.5854,
|
790 |
"step": 645
|
791 |
},
|
792 |
{
|
793 |
+
"epoch": 1.43,
|
794 |
+
"learning_rate": 9.729317113320311e-05,
|
795 |
+
"loss": 0.6888,
|
796 |
"step": 650
|
797 |
},
|
798 |
{
|
799 |
+
"epoch": 1.44,
|
800 |
+
"learning_rate": 9.512533856578284e-05,
|
801 |
+
"loss": 0.6905,
|
802 |
"step": 655
|
803 |
},
|
804 |
{
|
805 |
+
"epoch": 1.45,
|
806 |
+
"learning_rate": 9.2925895251118e-05,
|
807 |
+
"loss": 0.664,
|
808 |
"step": 660
|
809 |
},
|
810 |
{
|
811 |
+
"epoch": 1.46,
|
812 |
+
"learning_rate": 9.069746230737689e-05,
|
813 |
+
"loss": 0.7013,
|
814 |
"step": 665
|
815 |
},
|
816 |
{
|
817 |
+
"epoch": 1.47,
|
818 |
+
"learning_rate": 8.844269540020936e-05,
|
819 |
+
"loss": 0.6946,
|
820 |
"step": 670
|
821 |
},
|
822 |
{
|
823 |
+
"epoch": 1.48,
|
824 |
+
"learning_rate": 8.616428157794893e-05,
|
825 |
+
"loss": 0.7819,
|
826 |
"step": 675
|
827 |
},
|
828 |
{
|
829 |
+
"epoch": 1.49,
|
830 |
+
"learning_rate": 8.386493606940288e-05,
|
831 |
+
"loss": 0.6662,
|
832 |
"step": 680
|
833 |
},
|
834 |
{
|
835 |
+
"epoch": 1.51,
|
836 |
+
"learning_rate": 8.154739904807017e-05,
|
837 |
+
"loss": 0.5889,
|
838 |
"step": 685
|
839 |
},
|
840 |
{
|
841 |
+
"epoch": 1.52,
|
842 |
+
"learning_rate": 7.921443236661376e-05,
|
843 |
+
"loss": 0.7833,
|
844 |
"step": 690
|
845 |
},
|
846 |
{
|
847 |
+
"epoch": 1.53,
|
848 |
+
"learning_rate": 7.686881626551602e-05,
|
849 |
+
"loss": 0.7548,
|
850 |
"step": 695
|
851 |
},
|
852 |
{
|
853 |
+
"epoch": 1.54,
|
854 |
+
"learning_rate": 7.45133460598098e-05,
|
855 |
+
"loss": 0.6886,
|
856 |
"step": 700
|
857 |
},
|
858 |
{
|
859 |
+
"epoch": 1.55,
|
860 |
+
"learning_rate": 7.215082880785842e-05,
|
861 |
+
"loss": 0.7179,
|
862 |
"step": 705
|
863 |
},
|
864 |
{
|
865 |
+
"epoch": 1.56,
|
866 |
+
"learning_rate": 6.978407996610975e-05,
|
867 |
+
"loss": 0.745,
|
868 |
"step": 710
|
869 |
},
|
870 |
{
|
871 |
+
"epoch": 1.57,
|
872 |
+
"learning_rate": 6.741592003389085e-05,
|
873 |
+
"loss": 0.5811,
|
874 |
"step": 715
|
875 |
},
|
876 |
{
|
877 |
+
"epoch": 1.58,
|
878 |
+
"learning_rate": 6.504917119214413e-05,
|
879 |
+
"loss": 0.6962,
|
880 |
"step": 720
|
881 |
},
|
882 |
{
|
883 |
+
"epoch": 1.59,
|
884 |
+
"learning_rate": 6.268665394019079e-05,
|
885 |
+
"loss": 0.6763,
|
886 |
"step": 725
|
887 |
},
|
888 |
{
|
889 |
+
"epoch": 1.6,
|
890 |
+
"learning_rate": 6.033118373448457e-05,
|
891 |
+
"loss": 0.6247,
|
892 |
"step": 730
|
893 |
},
|
894 |
{
|
895 |
+
"epoch": 1.62,
|
896 |
+
"learning_rate": 5.7985567633386836e-05,
|
897 |
+
"loss": 0.7763,
|
898 |
"step": 735
|
899 |
},
|
900 |
{
|
901 |
+
"epoch": 1.63,
|
902 |
+
"learning_rate": 5.5652600951930425e-05,
|
903 |
+
"loss": 0.6239,
|
904 |
"step": 740
|
905 |
},
|
906 |
{
|
907 |
+
"epoch": 1.64,
|
908 |
+
"learning_rate": 5.333506393059772e-05,
|
909 |
+
"loss": 0.7218,
|
910 |
"step": 745
|
911 |
},
|
912 |
{
|
913 |
+
"epoch": 1.65,
|
914 |
+
"learning_rate": 5.103571842205165e-05,
|
915 |
+
"loss": 0.695,
|
916 |
"step": 750
|
917 |
},
|
918 |
{
|
919 |
+
"epoch": 1.66,
|
920 |
+
"learning_rate": 4.8757304599793096e-05,
|
921 |
+
"loss": 0.7938,
|
922 |
"step": 755
|
923 |
},
|
924 |
{
|
925 |
+
"epoch": 1.67,
|
926 |
+
"learning_rate": 4.650253769262368e-05,
|
927 |
+
"loss": 0.7333,
|
928 |
"step": 760
|
929 |
},
|
930 |
{
|
931 |
+
"epoch": 1.68,
|
932 |
+
"learning_rate": 4.4274104748882565e-05,
|
933 |
+
"loss": 0.7092,
|
934 |
"step": 765
|
935 |
},
|
936 |
{
|
937 |
+
"epoch": 1.69,
|
938 |
+
"learning_rate": 4.2074661434219527e-05,
|
939 |
+
"loss": 0.764,
|
940 |
"step": 770
|
941 |
},
|
942 |
{
|
943 |
+
"epoch": 1.7,
|
944 |
+
"learning_rate": 3.9906828866797437e-05,
|
945 |
+
"loss": 0.7929,
|
946 |
"step": 775
|
947 |
},
|
948 |
{
|
949 |
+
"epoch": 1.71,
|
950 |
+
"learning_rate": 3.777319049365253e-05,
|
951 |
+
"loss": 0.6747,
|
952 |
"step": 780
|
953 |
},
|
954 |
{
|
955 |
+
"epoch": 1.73,
|
956 |
+
"learning_rate": 3.56762890119588e-05,
|
957 |
+
"loss": 0.6921,
|
958 |
"step": 785
|
959 |
},
|
960 |
{
|
961 |
+
"epoch": 1.74,
|
962 |
+
"learning_rate": 3.361862333883716e-05,
|
963 |
+
"loss": 0.7049,
|
964 |
"step": 790
|
965 |
},
|
966 |
{
|
967 |
+
"epoch": 1.75,
|
968 |
+
"learning_rate": 3.160264563335574e-05,
|
969 |
+
"loss": 0.778,
|
970 |
"step": 795
|
971 |
},
|
972 |
{
|
973 |
+
"epoch": 1.76,
|
974 |
+
"learning_rate": 2.9630758374242215e-05,
|
975 |
+
"loss": 0.6684,
|
976 |
"step": 800
|
977 |
},
|
978 |
{
|
979 |
+
"epoch": 1.77,
|
980 |
+
"learning_rate": 2.7705311496812532e-05,
|
981 |
+
"loss": 0.6652,
|
982 |
"step": 805
|
983 |
},
|
984 |
{
|
985 |
+
"epoch": 1.78,
|
986 |
+
"learning_rate": 2.582859959249102e-05,
|
987 |
+
"loss": 0.7384,
|
988 |
"step": 810
|
989 |
},
|
990 |
{
|
991 |
+
"epoch": 1.79,
|
992 |
+
"learning_rate": 2.400285917432458e-05,
|
993 |
+
"loss": 0.6104,
|
994 |
"step": 815
|
995 |
},
|
996 |
{
|
997 |
+
"epoch": 1.8,
|
998 |
+
"learning_rate": 2.2230266011670566e-05,
|
999 |
+
"loss": 0.7144,
|
1000 |
"step": 820
|
1001 |
},
|
1002 |
{
|
1003 |
+
"epoch": 1.81,
|
1004 |
+
"learning_rate": 2.0512932537299123e-05,
|
1005 |
+
"loss": 0.5937,
|
1006 |
"step": 825
|
1007 |
},
|
1008 |
{
|
1009 |
+
"epoch": 1.82,
|
1010 |
+
"learning_rate": 1.885290532996424e-05,
|
1011 |
+
"loss": 0.7779,
|
1012 |
"step": 830
|
1013 |
},
|
1014 |
{
|
1015 |
+
"epoch": 1.84,
|
1016 |
+
"learning_rate": 1.7252162675462595e-05,
|
1017 |
+
"loss": 0.7421,
|
1018 |
"step": 835
|
1019 |
},
|
1020 |
{
|
1021 |
+
"epoch": 1.85,
|
1022 |
+
"learning_rate": 1.571261220906478e-05,
|
1023 |
+
"loss": 0.7221,
|
1024 |
"step": 840
|
1025 |
},
|
1026 |
{
|
1027 |
+
"epoch": 1.86,
|
1028 |
+
"learning_rate": 1.4236088642155802e-05,
|
1029 |
+
"loss": 0.757,
|
1030 |
"step": 845
|
1031 |
},
|
1032 |
{
|
1033 |
+
"epoch": 1.87,
|
1034 |
+
"learning_rate": 1.2824351575772341e-05,
|
1035 |
+
"loss": 0.752,
|
1036 |
"step": 850
|
1037 |
},
|
1038 |
{
|
1039 |
+
"epoch": 1.88,
|
1040 |
+
"learning_rate": 1.1479083403658627e-05,
|
1041 |
+
"loss": 0.743,
|
1042 |
"step": 855
|
1043 |
},
|
1044 |
{
|
1045 |
+
"epoch": 1.89,
|
1046 |
+
"learning_rate": 1.0201887307313627e-05,
|
1047 |
+
"loss": 0.5773,
|
1048 |
"step": 860
|
1049 |
},
|
1050 |
{
|
1051 |
+
"epoch": 1.9,
|
1052 |
+
"learning_rate": 8.994285345464858e-06,
|
1053 |
+
"loss": 0.7125,
|
1054 |
"step": 865
|
1055 |
},
|
1056 |
{
|
1057 |
+
"epoch": 1.91,
|
1058 |
+
"learning_rate": 7.857716640190273e-06,
|
1059 |
+
"loss": 0.6377,
|
1060 |
"step": 870
|
1061 |
},
|
1062 |
{
|
1063 |
+
"epoch": 1.92,
|
1064 |
+
"learning_rate": 6.793535661894656e-06,
|
1065 |
+
"loss": 0.7089,
|
1066 |
"step": 875
|
1067 |
},
|
1068 |
{
|
1069 |
+
"epoch": 1.93,
|
1070 |
+
"learning_rate": 5.803010615159811e-06,
|
1071 |
+
"loss": 0.61,
|
1072 |
"step": 880
|
1073 |
},
|
1074 |
{
|
1075 |
+
"epoch": 1.95,
|
1076 |
+
"learning_rate": 4.887321927404351e-06,
|
1077 |
+
"loss": 0.696,
|
1078 |
"step": 885
|
1079 |
},
|
1080 |
{
|
1081 |
+
"epoch": 1.96,
|
1082 |
+
"learning_rate": 4.047560842141189e-06,
|
1083 |
+
"loss": 0.6485,
|
1084 |
"step": 890
|
1085 |
},
|
1086 |
{
|
1087 |
+
"epoch": 1.97,
|
1088 |
+
"learning_rate": 3.2847281185253237e-06,
|
1089 |
+
"loss": 0.6823,
|
1090 |
"step": 895
|
1091 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1092 |
{
|
1093 |
"epoch": 1.98,
|
1094 |
+
"learning_rate": 2.599732838728863e-06,
|
1095 |
+
"loss": 0.7038,
|
1096 |
+
"step": 900
|
1097 |
},
|
1098 |
{
|
1099 |
"epoch": 1.99,
|
1100 |
+
"learning_rate": 1.993391324573266e-06,
|
1101 |
+
"loss": 0.6816,
|
1102 |
+
"step": 905
|
1103 |
},
|
1104 |
{
|
1105 |
"epoch": 2.0,
|
1106 |
+
"learning_rate": 1.4664261646975266e-06,
|
1107 |
+
"loss": 0.6658,
|
1108 |
+
"step": 910
|
1109 |
},
|
1110 |
{
|
1111 |
"epoch": 2.0,
|
1112 |
+
"eval_loss": 0.3614741861820221,
|
1113 |
+
"eval_runtime": 29.3257,
|
1114 |
+
"eval_samples_per_second": 21.96,
|
1115 |
+
"eval_steps_per_second": 2.762,
|
1116 |
+
"step": 910
|
1117 |
}
|
1118 |
],
|
1119 |
+
"max_steps": 1365,
|
1120 |
"num_train_epochs": 3,
|
1121 |
+
"total_flos": 949535244288000.0,
|
1122 |
"trial_name": null,
|
1123 |
"trial_params": null
|
1124 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3055
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:20a154128815e7bbba517b0e4c2b881a6382b44d7ef3cd0aa609175592021c15
|
3 |
size 3055
|