AlekseyKorshuk
commited on
Commit
•
fc38921
1
Parent(s):
6d2c1bd
huggingartists
Browse files- README.md +3 -3
- config.json +1 -1
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +243 -797
- training_args.bin +2 -2
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/3rukw5ac/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/2mua6tz8) for full transparency and reproducibility.
|
55 |
|
56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/2mua6tz8/artifacts) is logged and versioned.
|
57 |
|
58 |
## How to use
|
59 |
|
config.json
CHANGED
@@ -37,7 +37,7 @@
|
|
37 |
}
|
38 |
},
|
39 |
"torch_dtype": "float32",
|
40 |
-
"transformers_version": "4.
|
41 |
"use_cache": true,
|
42 |
"vocab_size": 50257
|
43 |
}
|
|
|
37 |
}
|
38 |
},
|
39 |
"torch_dtype": "float32",
|
40 |
+
"transformers_version": "4.19.2",
|
41 |
"use_cache": true,
|
42 |
"vocab_size": 50257
|
43 |
}
|
evaluation.txt
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"eval_loss": 0.
|
|
|
1 |
+
{"eval_loss": 0.22213919460773468, "eval_runtime": 29.4043, "eval_samples_per_second": 21.936, "eval_steps_per_second": 2.755, "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:f2c3b167b74a63ec8fb077ce24d21025947fef9ecb32878a46ee4d6977cc15f0
|
3 |
size 497764120
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 995604017
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:26834f0e4208938d811a9dfa581c60096828db9664c29122cbe7c108c2266f68
|
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 510396521
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c21f7e095359913b7c6cf81d01da61fcc6b0f7f55c7374a09691747e7e324411
|
3 |
size 510396521
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14567
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:091b8c645ee921fc1fb21bf66e87b6404370564c2f1e26f8e92fd9a5afcce570
|
3 |
size 14567
|
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:c3e6d52e9eda83b53901112c80d072ee8b7df374009cb1437cc93a39ef8c5f83
|
3 |
size 623
|
trainer_state.json
CHANGED
@@ -1,1124 +1,570 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "output/eminem/checkpoint-
|
4 |
-
"epoch":
|
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": 0.
|
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":
|
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":
|
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.
|
102 |
-
"learning_rate":
|
103 |
-
"loss": 0.
|
104 |
"step": 80
|
105 |
},
|
106 |
{
|
107 |
-
"epoch": 0.
|
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":
|
139 |
-
"loss": 0.
|
140 |
"step": 110
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.25,
|
144 |
-
"learning_rate":
|
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.
|
162 |
-
"learning_rate":
|
163 |
-
"loss": 0.
|
164 |
"step": 130
|
165 |
},
|
166 |
{
|
167 |
-
"epoch": 0.
|
168 |
-
"learning_rate":
|
169 |
-
"loss": 0.
|
170 |
"step": 135
|
171 |
},
|
172 |
{
|
173 |
-
"epoch": 0.
|
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.
|
228 |
-
"learning_rate":
|
229 |
-
"loss": 0.
|
230 |
"step": 185
|
231 |
},
|
232 |
{
|
233 |
-
"epoch": 0.
|
234 |
-
"learning_rate":
|
235 |
-
"loss": 0.
|
236 |
"step": 190
|
237 |
},
|
238 |
{
|
239 |
-
"epoch": 0.
|
240 |
-
"learning_rate":
|
241 |
-
"loss": 0.
|
242 |
"step": 195
|
243 |
},
|
244 |
{
|
245 |
-
"epoch": 0.
|
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.
|
288 |
-
"learning_rate":
|
289 |
-
"loss": 0.
|
290 |
"step": 235
|
291 |
},
|
292 |
{
|
293 |
-
"epoch": 0.
|
294 |
-
"learning_rate":
|
295 |
-
"loss": 0.
|
296 |
"step": 240
|
297 |
},
|
298 |
{
|
299 |
-
"epoch": 0.
|
300 |
-
"learning_rate": 6.
|
301 |
-
"loss": 0.
|
302 |
"step": 245
|
303 |
},
|
304 |
{
|
305 |
-
"epoch": 0.
|
306 |
-
"learning_rate":
|
307 |
-
"loss": 0.
|
308 |
"step": 250
|
309 |
},
|
310 |
{
|
311 |
-
"epoch": 0.
|
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.
|
354 |
-
"learning_rate": 4.
|
355 |
-
"loss": 0.
|
356 |
"step": 290
|
357 |
},
|
358 |
{
|
359 |
-
"epoch": 0.
|
360 |
-
"learning_rate":
|
361 |
-
"loss": 0.
|
362 |
"step": 295
|
363 |
},
|
364 |
{
|
365 |
-
"epoch": 0.
|
366 |
-
"learning_rate":
|
367 |
-
"loss": 0.
|
368 |
"step": 300
|
369 |
},
|
370 |
{
|
371 |
-
"epoch": 0.
|
372 |
-
"learning_rate":
|
373 |
-
"loss": 0.
|
374 |
"step": 305
|
375 |
},
|
376 |
{
|
377 |
-
"epoch": 0.
|
378 |
-
"learning_rate":
|
379 |
-
"loss": 0.
|
380 |
"step": 310
|
381 |
},
|
382 |
{
|
383 |
-
"epoch": 0.
|
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.
|
414 |
-
"learning_rate":
|
415 |
-
"loss": 0.
|
416 |
"step": 340
|
417 |
},
|
418 |
{
|
419 |
-
"epoch": 0.
|
420 |
-
"learning_rate": 2.
|
421 |
-
"loss": 0.
|
422 |
"step": 345
|
423 |
},
|
424 |
{
|
425 |
-
"epoch": 0.
|
426 |
-
"learning_rate": 2.
|
427 |
-
"loss": 0.
|
428 |
"step": 350
|
429 |
},
|
430 |
{
|
431 |
-
"epoch": 0.
|
432 |
-
"learning_rate":
|
433 |
-
"loss": 0.
|
434 |
"step": 355
|
435 |
},
|
436 |
{
|
437 |
-
"epoch": 0.
|
438 |
-
"learning_rate":
|
439 |
-
"loss": 0.
|
440 |
"step": 360
|
441 |
},
|
442 |
{
|
443 |
-
"epoch": 0.
|
444 |
-
"learning_rate":
|
445 |
-
"loss": 0.
|
446 |
"step": 365
|
447 |
},
|
448 |
{
|
449 |
-
"epoch": 0.
|
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": 0.83,
|
462 |
-
"learning_rate": 8.350443827565338e-06,
|
463 |
-
"loss": 0.4506,
|
464 |
-
"step": 380
|
465 |
-
},
|
466 |
{
|
467 |
"epoch": 0.84,
|
468 |
-
"learning_rate":
|
469 |
-
"loss": 0.
|
470 |
-
"step":
|
471 |
},
|
472 |
{
|
473 |
"epoch": 0.85,
|
474 |
-
"learning_rate": 1.
|
475 |
-
"loss": 0.
|
476 |
-
"step":
|
477 |
},
|
478 |
{
|
479 |
"epoch": 0.86,
|
480 |
-
"learning_rate":
|
481 |
-
"loss": 0.
|
482 |
-
"step":
|
483 |
},
|
484 |
{
|
485 |
"epoch": 0.87,
|
486 |
-
"learning_rate":
|
487 |
-
"loss": 0.
|
488 |
-
"step":
|
489 |
},
|
490 |
{
|
491 |
"epoch": 0.88,
|
492 |
-
"learning_rate":
|
493 |
-
"loss": 0.
|
494 |
-
"step":
|
495 |
},
|
496 |
{
|
497 |
"epoch": 0.89,
|
498 |
-
"learning_rate":
|
499 |
-
"loss": 0.
|
500 |
-
"step":
|
501 |
},
|
502 |
{
|
503 |
"epoch": 0.9,
|
504 |
-
"learning_rate":
|
505 |
-
"loss": 0.
|
506 |
-
"step":
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.91,
|
510 |
-
"learning_rate":
|
511 |
-
"loss": 0.
|
512 |
-
"step":
|
513 |
},
|
514 |
{
|
515 |
"epoch": 0.92,
|
516 |
-
"learning_rate":
|
517 |
-
"loss": 0.
|
518 |
-
"step":
|
519 |
},
|
520 |
{
|
521 |
"epoch": 0.93,
|
522 |
-
"learning_rate": 2.
|
523 |
-
"loss": 0.
|
524 |
-
"step":
|
525 |
},
|
526 |
{
|
527 |
"epoch": 0.95,
|
528 |
-
"learning_rate": 2.
|
529 |
-
"loss": 0.
|
530 |
-
"step":
|
531 |
},
|
532 |
{
|
533 |
"epoch": 0.96,
|
534 |
-
"learning_rate":
|
535 |
-
"loss": 0.
|
536 |
-
"step":
|
537 |
},
|
538 |
{
|
539 |
"epoch": 0.97,
|
540 |
-
"learning_rate":
|
541 |
-
"loss": 0.
|
542 |
-
"step":
|
543 |
},
|
544 |
{
|
545 |
"epoch": 0.98,
|
546 |
-
"learning_rate":
|
547 |
-
"loss": 0.
|
548 |
-
"step":
|
549 |
},
|
550 |
{
|
551 |
"epoch": 0.99,
|
552 |
-
"learning_rate":
|
553 |
-
"loss": 0.
|
554 |
-
"step":
|
555 |
},
|
556 |
{
|
557 |
"epoch": 1.0,
|
558 |
-
"learning_rate":
|
559 |
-
"loss": 0.
|
560 |
-
"step":
|
561 |
},
|
562 |
{
|
563 |
"epoch": 1.0,
|
564 |
-
"eval_loss": 0.
|
565 |
-
"eval_runtime":
|
566 |
-
"eval_samples_per_second":
|
567 |
-
"eval_steps_per_second":
|
568 |
-
"step":
|
569 |
-
},
|
570 |
-
{
|
571 |
-
"epoch": 1.02,
|
572 |
-
"learning_rate": 0.0001310364026633813,
|
573 |
-
"loss": 0.4239,
|
574 |
-
"step": 465
|
575 |
-
},
|
576 |
-
{
|
577 |
-
"epoch": 1.03,
|
578 |
-
"learning_rate": 0.0001319781359302741,
|
579 |
-
"loss": 0.5115,
|
580 |
-
"step": 470
|
581 |
-
},
|
582 |
-
{
|
583 |
-
"epoch": 1.04,
|
584 |
-
"learning_rate": 0.00013284467114611735,
|
585 |
-
"loss": 0.4756,
|
586 |
-
"step": 475
|
587 |
-
},
|
588 |
-
{
|
589 |
-
"epoch": 1.05,
|
590 |
-
"learning_rate": 0.00013363498016834413,
|
591 |
-
"loss": 0.4892,
|
592 |
-
"step": 480
|
593 |
-
},
|
594 |
-
{
|
595 |
-
"epoch": 1.06,
|
596 |
-
"learning_rate": 0.00013434812529663595,
|
597 |
-
"loss": 0.5002,
|
598 |
-
"step": 485
|
599 |
-
},
|
600 |
-
{
|
601 |
-
"epoch": 1.07,
|
602 |
-
"learning_rate": 0.00013498326038550478,
|
603 |
-
"loss": 0.5714,
|
604 |
-
"step": 490
|
605 |
-
},
|
606 |
-
{
|
607 |
-
"epoch": 1.09,
|
608 |
-
"learning_rate": 0.00013553963184824178,
|
609 |
-
"loss": 0.5845,
|
610 |
-
"step": 495
|
611 |
-
},
|
612 |
-
{
|
613 |
-
"epoch": 1.1,
|
614 |
-
"learning_rate": 0.0001360165795510487,
|
615 |
-
"loss": 0.4911,
|
616 |
-
"step": 500
|
617 |
-
},
|
618 |
-
{
|
619 |
-
"epoch": 1.11,
|
620 |
-
"learning_rate": 0.00013641353759628265,
|
621 |
-
"loss": 0.5275,
|
622 |
-
"step": 505
|
623 |
-
},
|
624 |
-
{
|
625 |
-
"epoch": 1.12,
|
626 |
-
"learning_rate": 0.00013673003499389456,
|
627 |
-
"loss": 0.5292,
|
628 |
-
"step": 510
|
629 |
-
},
|
630 |
-
{
|
631 |
-
"epoch": 1.13,
|
632 |
-
"learning_rate": 0.00013696569622025749,
|
633 |
-
"loss": 0.5506,
|
634 |
-
"step": 515
|
635 |
-
},
|
636 |
-
{
|
637 |
-
"epoch": 1.14,
|
638 |
-
"learning_rate": 0.00013712024166372548,
|
639 |
-
"loss": 0.5552,
|
640 |
-
"step": 520
|
641 |
-
},
|
642 |
-
{
|
643 |
-
"epoch": 1.15,
|
644 |
-
"learning_rate": 0.0001371934879563911,
|
645 |
-
"loss": 0.5404,
|
646 |
-
"step": 525
|
647 |
-
},
|
648 |
-
{
|
649 |
-
"epoch": 1.16,
|
650 |
-
"learning_rate": 0.0001371853481916521,
|
651 |
-
"loss": 0.5364,
|
652 |
-
"step": 530
|
653 |
-
},
|
654 |
-
{
|
655 |
-
"epoch": 1.17,
|
656 |
-
"learning_rate": 0.0001370958320273255,
|
657 |
-
"loss": 0.4954,
|
658 |
-
"step": 535
|
659 |
-
},
|
660 |
-
{
|
661 |
-
"epoch": 1.18,
|
662 |
-
"learning_rate": 0.00013692504567418905,
|
663 |
-
"loss": 0.6414,
|
664 |
-
"step": 540
|
665 |
-
},
|
666 |
-
{
|
667 |
-
"epoch": 1.2,
|
668 |
-
"learning_rate": 0.0001366731917699612,
|
669 |
-
"loss": 0.5806,
|
670 |
-
"step": 545
|
671 |
-
},
|
672 |
-
{
|
673 |
-
"epoch": 1.21,
|
674 |
-
"learning_rate": 0.00013634056913887406,
|
675 |
-
"loss": 0.5283,
|
676 |
-
"step": 550
|
677 |
-
},
|
678 |
-
{
|
679 |
-
"epoch": 1.22,
|
680 |
-
"learning_rate": 0.0001359275724371159,
|
681 |
-
"loss": 0.5036,
|
682 |
-
"step": 555
|
683 |
-
},
|
684 |
-
{
|
685 |
-
"epoch": 1.23,
|
686 |
-
"learning_rate": 0.00013543469168457526,
|
687 |
-
"loss": 0.583,
|
688 |
-
"step": 560
|
689 |
-
},
|
690 |
-
{
|
691 |
-
"epoch": 1.24,
|
692 |
-
"learning_rate": 0.00013486251168343032,
|
693 |
-
"loss": 0.5384,
|
694 |
-
"step": 565
|
695 |
-
},
|
696 |
-
{
|
697 |
-
"epoch": 1.25,
|
698 |
-
"learning_rate": 0.0001342117113242847,
|
699 |
-
"loss": 0.5136,
|
700 |
-
"step": 570
|
701 |
-
},
|
702 |
-
{
|
703 |
-
"epoch": 1.26,
|
704 |
-
"learning_rate": 0.00013348306278066345,
|
705 |
-
"loss": 0.5162,
|
706 |
-
"step": 575
|
707 |
-
},
|
708 |
-
{
|
709 |
-
"epoch": 1.27,
|
710 |
-
"learning_rate": 0.0001326774305928314,
|
711 |
-
"loss": 0.4983,
|
712 |
-
"step": 580
|
713 |
-
},
|
714 |
-
{
|
715 |
-
"epoch": 1.28,
|
716 |
-
"learning_rate": 0.00013179577064201505,
|
717 |
-
"loss": 0.5491,
|
718 |
-
"step": 585
|
719 |
-
},
|
720 |
-
{
|
721 |
-
"epoch": 1.29,
|
722 |
-
"learning_rate": 0.00013083912901625498,
|
723 |
-
"loss": 0.4611,
|
724 |
-
"step": 590
|
725 |
-
},
|
726 |
-
{
|
727 |
-
"epoch": 1.3,
|
728 |
-
"learning_rate": 0.0001298086407692224,
|
729 |
-
"loss": 0.5359,
|
730 |
-
"step": 595
|
731 |
-
},
|
732 |
-
{
|
733 |
-
"epoch": 1.32,
|
734 |
-
"learning_rate": 0.00012870552857347723,
|
735 |
-
"loss": 0.5226,
|
736 |
-
"step": 600
|
737 |
-
},
|
738 |
-
{
|
739 |
-
"epoch": 1.33,
|
740 |
-
"learning_rate": 0.00012753110126977415,
|
741 |
-
"loss": 0.5316,
|
742 |
-
"step": 605
|
743 |
-
},
|
744 |
-
{
|
745 |
-
"epoch": 1.34,
|
746 |
-
"learning_rate": 0.00012628675231412197,
|
747 |
-
"loss": 0.5227,
|
748 |
-
"step": 610
|
749 |
-
},
|
750 |
-
{
|
751 |
-
"epoch": 1.35,
|
752 |
-
"learning_rate": 0.0001249739581244523,
|
753 |
-
"loss": 0.4961,
|
754 |
-
"step": 615
|
755 |
-
},
|
756 |
-
{
|
757 |
-
"epoch": 1.36,
|
758 |
-
"learning_rate": 0.00012359427632884786,
|
759 |
-
"loss": 0.5487,
|
760 |
-
"step": 620
|
761 |
-
},
|
762 |
-
{
|
763 |
-
"epoch": 1.37,
|
764 |
-
"learning_rate": 0.0001221493439174271,
|
765 |
-
"loss": 0.5994,
|
766 |
-
"step": 625
|
767 |
-
},
|
768 |
-
{
|
769 |
-
"epoch": 1.38,
|
770 |
-
"learning_rate": 0.00012064087530004776,
|
771 |
-
"loss": 0.5401,
|
772 |
-
"step": 630
|
773 |
-
},
|
774 |
-
{
|
775 |
-
"epoch": 1.39,
|
776 |
-
"learning_rate": 0.00011907066027217711,
|
777 |
-
"loss": 0.5615,
|
778 |
-
"step": 635
|
779 |
-
},
|
780 |
-
{
|
781 |
-
"epoch": 1.4,
|
782 |
-
"learning_rate": 0.00011744056189128525,
|
783 |
-
"loss": 0.5705,
|
784 |
-
"step": 640
|
785 |
-
},
|
786 |
-
{
|
787 |
-
"epoch": 1.41,
|
788 |
-
"learning_rate": 0.00011575251426634572,
|
789 |
-
"loss": 0.5663,
|
790 |
-
"step": 645
|
791 |
-
},
|
792 |
-
{
|
793 |
-
"epoch": 1.43,
|
794 |
-
"learning_rate": 0.00011400852026300672,
|
795 |
-
"loss": 0.5623,
|
796 |
-
"step": 650
|
797 |
-
},
|
798 |
-
{
|
799 |
-
"epoch": 1.44,
|
800 |
-
"learning_rate": 0.00011221064912720382,
|
801 |
-
"loss": 0.5898,
|
802 |
-
"step": 655
|
803 |
-
},
|
804 |
-
{
|
805 |
-
"epoch": 1.45,
|
806 |
-
"learning_rate": 0.00011036103402999931,
|
807 |
-
"loss": 0.5708,
|
808 |
-
"step": 660
|
809 |
-
},
|
810 |
-
{
|
811 |
-
"epoch": 1.46,
|
812 |
-
"learning_rate": 0.00010846186953657361,
|
813 |
-
"loss": 0.5299,
|
814 |
-
"step": 665
|
815 |
-
},
|
816 |
-
{
|
817 |
-
"epoch": 1.47,
|
818 |
-
"learning_rate": 0.00010651540900238634,
|
819 |
-
"loss": 0.6369,
|
820 |
-
"step": 670
|
821 |
-
},
|
822 |
-
{
|
823 |
-
"epoch": 1.48,
|
824 |
-
"learning_rate": 0.00010452396189956808,
|
825 |
-
"loss": 0.532,
|
826 |
-
"step": 675
|
827 |
-
},
|
828 |
-
{
|
829 |
-
"epoch": 1.49,
|
830 |
-
"learning_rate": 0.0001024898910767386,
|
831 |
-
"loss": 0.5957,
|
832 |
-
"step": 680
|
833 |
-
},
|
834 |
-
{
|
835 |
-
"epoch": 1.5,
|
836 |
-
"learning_rate": 0.00010041560995548433,
|
837 |
-
"loss": 0.4907,
|
838 |
-
"step": 685
|
839 |
-
},
|
840 |
-
{
|
841 |
-
"epoch": 1.51,
|
842 |
-
"learning_rate": 9.830357966684779e-05,
|
843 |
-
"loss": 0.5726,
|
844 |
-
"step": 690
|
845 |
-
},
|
846 |
-
{
|
847 |
-
"epoch": 1.52,
|
848 |
-
"learning_rate": 9.615630613119297e-05,
|
849 |
-
"loss": 0.5582,
|
850 |
-
"step": 695
|
851 |
-
},
|
852 |
-
{
|
853 |
-
"epoch": 1.54,
|
854 |
-
"learning_rate": 9.397633708493872e-05,
|
855 |
-
"loss": 0.6085,
|
856 |
-
"step": 700
|
857 |
-
},
|
858 |
-
{
|
859 |
-
"epoch": 1.55,
|
860 |
-
"learning_rate": 9.17662590576683e-05,
|
861 |
-
"loss": 0.6075,
|
862 |
-
"step": 705
|
863 |
-
},
|
864 |
-
{
|
865 |
-
"epoch": 1.56,
|
866 |
-
"learning_rate": 8.95286943032199e-05,
|
867 |
-
"loss": 0.5199,
|
868 |
-
"step": 710
|
869 |
-
},
|
870 |
-
{
|
871 |
-
"epoch": 1.57,
|
872 |
-
"learning_rate": 8.726629768839068e-05,
|
873 |
-
"loss": 0.565,
|
874 |
-
"step": 715
|
875 |
-
},
|
876 |
-
{
|
877 |
-
"epoch": 1.58,
|
878 |
-
"learning_rate": 8.49817535429363e-05,
|
879 |
-
"loss": 0.5779,
|
880 |
-
"step": 720
|
881 |
-
},
|
882 |
-
{
|
883 |
-
"epoch": 1.59,
|
884 |
-
"learning_rate": 8.267777247462197e-05,
|
885 |
-
"loss": 0.514,
|
886 |
-
"step": 725
|
887 |
-
},
|
888 |
-
{
|
889 |
-
"epoch": 1.6,
|
890 |
-
"learning_rate": 8.035708815308325e-05,
|
891 |
-
"loss": 0.6287,
|
892 |
-
"step": 730
|
893 |
-
},
|
894 |
-
{
|
895 |
-
"epoch": 1.61,
|
896 |
-
"learning_rate": 7.802245406634311e-05,
|
897 |
-
"loss": 0.5617,
|
898 |
-
"step": 735
|
899 |
-
},
|
900 |
-
{
|
901 |
-
"epoch": 1.62,
|
902 |
-
"learning_rate": 7.56766402537954e-05,
|
903 |
-
"loss": 0.6135,
|
904 |
-
"step": 740
|
905 |
-
},
|
906 |
-
{
|
907 |
-
"epoch": 1.63,
|
908 |
-
"learning_rate": 7.332243001956172e-05,
|
909 |
-
"loss": 0.5151,
|
910 |
-
"step": 745
|
911 |
-
},
|
912 |
-
{
|
913 |
-
"epoch": 1.64,
|
914 |
-
"learning_rate": 7.096261663009792e-05,
|
915 |
-
"loss": 0.5282,
|
916 |
-
"step": 750
|
917 |
-
},
|
918 |
-
{
|
919 |
-
"epoch": 1.66,
|
920 |
-
"learning_rate": 6.85999999999999e-05,
|
921 |
-
"loss": 0.5488,
|
922 |
-
"step": 755
|
923 |
-
},
|
924 |
-
{
|
925 |
-
"epoch": 1.67,
|
926 |
-
"learning_rate": 6.623738336990383e-05,
|
927 |
-
"loss": 0.5678,
|
928 |
-
"step": 760
|
929 |
-
},
|
930 |
-
{
|
931 |
-
"epoch": 1.68,
|
932 |
-
"learning_rate": 6.387756998044003e-05,
|
933 |
-
"loss": 0.5534,
|
934 |
-
"step": 765
|
935 |
-
},
|
936 |
-
{
|
937 |
-
"epoch": 1.69,
|
938 |
-
"learning_rate": 6.152335974620634e-05,
|
939 |
-
"loss": 0.6249,
|
940 |
-
"step": 770
|
941 |
-
},
|
942 |
-
{
|
943 |
-
"epoch": 1.7,
|
944 |
-
"learning_rate": 5.917754593365669e-05,
|
945 |
-
"loss": 0.5588,
|
946 |
-
"step": 775
|
947 |
-
},
|
948 |
-
{
|
949 |
-
"epoch": 1.71,
|
950 |
-
"learning_rate": 5.684291184691846e-05,
|
951 |
-
"loss": 0.4985,
|
952 |
-
"step": 780
|
953 |
-
},
|
954 |
-
{
|
955 |
-
"epoch": 1.72,
|
956 |
-
"learning_rate": 5.452222752537974e-05,
|
957 |
-
"loss": 0.545,
|
958 |
-
"step": 785
|
959 |
-
},
|
960 |
-
{
|
961 |
-
"epoch": 1.73,
|
962 |
-
"learning_rate": 5.2218246457065396e-05,
|
963 |
-
"loss": 0.5905,
|
964 |
-
"step": 790
|
965 |
-
},
|
966 |
-
{
|
967 |
-
"epoch": 1.74,
|
968 |
-
"learning_rate": 4.993370231160913e-05,
|
969 |
-
"loss": 0.4741,
|
970 |
-
"step": 795
|
971 |
-
},
|
972 |
-
{
|
973 |
-
"epoch": 1.75,
|
974 |
-
"learning_rate": 4.767130569678176e-05,
|
975 |
-
"loss": 0.5473,
|
976 |
-
"step": 800
|
977 |
-
},
|
978 |
-
{
|
979 |
-
"epoch": 1.77,
|
980 |
-
"learning_rate": 4.543374094233335e-05,
|
981 |
-
"loss": 0.525,
|
982 |
-
"step": 805
|
983 |
-
},
|
984 |
-
{
|
985 |
-
"epoch": 1.78,
|
986 |
-
"learning_rate": 4.3223662915062914e-05,
|
987 |
-
"loss": 0.5134,
|
988 |
-
"step": 810
|
989 |
-
},
|
990 |
-
{
|
991 |
-
"epoch": 1.79,
|
992 |
-
"learning_rate": 4.104369386880686e-05,
|
993 |
-
"loss": 0.5726,
|
994 |
-
"step": 815
|
995 |
-
},
|
996 |
-
{
|
997 |
-
"epoch": 1.8,
|
998 |
-
"learning_rate": 3.8896420333152035e-05,
|
999 |
-
"loss": 0.4985,
|
1000 |
-
"step": 820
|
1001 |
-
},
|
1002 |
-
{
|
1003 |
-
"epoch": 1.81,
|
1004 |
-
"learning_rate": 3.6784390044517205e-05,
|
1005 |
-
"loss": 0.5251,
|
1006 |
-
"step": 825
|
1007 |
-
},
|
1008 |
-
{
|
1009 |
-
"epoch": 1.82,
|
1010 |
-
"learning_rate": 3.471010892326291e-05,
|
1011 |
-
"loss": 0.6068,
|
1012 |
-
"step": 830
|
1013 |
-
},
|
1014 |
-
{
|
1015 |
-
"epoch": 1.83,
|
1016 |
-
"learning_rate": 3.267603810043175e-05,
|
1017 |
-
"loss": 0.5227,
|
1018 |
-
"step": 835
|
1019 |
-
},
|
1020 |
-
{
|
1021 |
-
"epoch": 1.84,
|
1022 |
-
"learning_rate": 3.068459099761349e-05,
|
1023 |
-
"loss": 0.5061,
|
1024 |
-
"step": 840
|
1025 |
-
},
|
1026 |
-
{
|
1027 |
-
"epoch": 1.85,
|
1028 |
-
"learning_rate": 2.8738130463427823e-05,
|
1029 |
-
"loss": 0.5539,
|
1030 |
-
"step": 845
|
1031 |
-
},
|
1032 |
-
{
|
1033 |
-
"epoch": 1.86,
|
1034 |
-
"learning_rate": 2.683896597000207e-05,
|
1035 |
-
"loss": 0.5975,
|
1036 |
-
"step": 850
|
1037 |
-
},
|
1038 |
-
{
|
1039 |
-
"epoch": 1.88,
|
1040 |
-
"learning_rate": 2.4989350872796035e-05,
|
1041 |
-
"loss": 0.5843,
|
1042 |
-
"step": 855
|
1043 |
-
},
|
1044 |
-
{
|
1045 |
-
"epoch": 1.89,
|
1046 |
-
"learning_rate": 2.3191479736993138e-05,
|
1047 |
-
"loss": 0.571,
|
1048 |
-
"step": 860
|
1049 |
-
},
|
1050 |
-
{
|
1051 |
-
"epoch": 1.9,
|
1052 |
-
"learning_rate": 2.1447485733655544e-05,
|
1053 |
-
"loss": 0.5282,
|
1054 |
-
"step": 865
|
1055 |
-
},
|
1056 |
-
{
|
1057 |
-
"epoch": 1.91,
|
1058 |
-
"learning_rate": 1.975943810871599e-05,
|
1059 |
-
"loss": 0.5626,
|
1060 |
-
"step": 870
|
1061 |
-
},
|
1062 |
-
{
|
1063 |
-
"epoch": 1.92,
|
1064 |
-
"learning_rate": 1.8129339727824064e-05,
|
1065 |
-
"loss": 0.5394,
|
1066 |
-
"step": 875
|
1067 |
-
},
|
1068 |
-
{
|
1069 |
-
"epoch": 1.93,
|
1070 |
-
"learning_rate": 1.6559124699952118e-05,
|
1071 |
-
"loss": 0.4747,
|
1072 |
-
"step": 880
|
1073 |
-
},
|
1074 |
-
{
|
1075 |
-
"epoch": 1.94,
|
1076 |
-
"learning_rate": 1.5050656082574002e-05,
|
1077 |
-
"loss": 0.5601,
|
1078 |
-
"step": 885
|
1079 |
-
},
|
1080 |
-
{
|
1081 |
-
"epoch": 1.95,
|
1082 |
-
"learning_rate": 1.3605723671153187e-05,
|
1083 |
-
"loss": 0.4725,
|
1084 |
-
"step": 890
|
1085 |
-
},
|
1086 |
-
{
|
1087 |
-
"epoch": 1.96,
|
1088 |
-
"learning_rate": 1.22260418755487e-05,
|
1089 |
-
"loss": 0.5171,
|
1090 |
-
"step": 895
|
1091 |
-
},
|
1092 |
-
{
|
1093 |
-
"epoch": 1.97,
|
1094 |
-
"learning_rate": 1.0913247685877927e-05,
|
1095 |
-
"loss": 0.5425,
|
1096 |
-
"step": 900
|
1097 |
-
},
|
1098 |
-
{
|
1099 |
-
"epoch": 1.98,
|
1100 |
-
"learning_rate": 9.66889873022574e-06,
|
1101 |
-
"loss": 0.4969,
|
1102 |
-
"step": 905
|
1103 |
-
},
|
1104 |
-
{
|
1105 |
-
"epoch": 2.0,
|
1106 |
-
"learning_rate": 8.494471426523612e-06,
|
1107 |
-
"loss": 0.5235,
|
1108 |
-
"step": 910
|
1109 |
-
},
|
1110 |
-
{
|
1111 |
-
"epoch": 2.0,
|
1112 |
-
"eval_loss": 0.21034620702266693,
|
1113 |
-
"eval_runtime": 15.3022,
|
1114 |
-
"eval_samples_per_second": 41.432,
|
1115 |
-
"eval_steps_per_second": 5.228,
|
1116 |
-
"step": 912
|
1117 |
}
|
1118 |
],
|
1119 |
-
"max_steps":
|
1120 |
-
"num_train_epochs":
|
1121 |
-
"total_flos":
|
1122 |
"trial_name": null,
|
1123 |
"trial_params": null
|
1124 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.22213919460773468,
|
3 |
+
"best_model_checkpoint": "output/eminem/checkpoint-455",
|
4 |
+
"epoch": 1.0,
|
5 |
+
"global_step": 455,
|
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": 0.00013711989076069754,
|
13 |
+
"loss": 0.4887,
|
14 |
"step": 5
|
15 |
},
|
16 |
{
|
17 |
"epoch": 0.02,
|
18 |
+
"learning_rate": 0.00013719345930102362,
|
19 |
+
"loss": 0.4564,
|
20 |
"step": 10
|
21 |
},
|
22 |
{
|
23 |
"epoch": 0.03,
|
24 |
+
"learning_rate": 0.00013718528371963108,
|
25 |
+
"loss": 0.4132,
|
26 |
"step": 15
|
27 |
},
|
28 |
{
|
29 |
"epoch": 0.04,
|
30 |
+
"learning_rate": 0.00013709537375951485,
|
31 |
+
"loss": 0.4758,
|
32 |
"step": 20
|
33 |
},
|
34 |
{
|
35 |
"epoch": 0.05,
|
36 |
+
"learning_rate": 0.00013692383656807334,
|
37 |
+
"loss": 0.4929,
|
38 |
"step": 25
|
39 |
},
|
40 |
{
|
41 |
"epoch": 0.07,
|
42 |
+
"learning_rate": 0.00013667087656941957,
|
43 |
+
"loss": 0.4404,
|
44 |
"step": 30
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.08,
|
48 |
+
"learning_rate": 0.0001363367952207642,
|
49 |
+
"loss": 0.4726,
|
50 |
"step": 35
|
51 |
},
|
52 |
{
|
53 |
"epoch": 0.09,
|
54 |
+
"learning_rate": 0.00013592199065316386,
|
55 |
+
"loss": 0.4448,
|
56 |
"step": 40
|
57 |
},
|
58 |
{
|
59 |
"epoch": 0.1,
|
60 |
+
"learning_rate": 0.00013542695719706024,
|
61 |
+
"loss": 0.4852,
|
62 |
"step": 45
|
63 |
},
|
64 |
{
|
65 |
"epoch": 0.11,
|
66 |
+
"learning_rate": 0.00013485228479317936,
|
67 |
+
"loss": 0.4865,
|
68 |
"step": 50
|
69 |
},
|
70 |
{
|
71 |
"epoch": 0.12,
|
72 |
+
"learning_rate": 0.00013419865828948315,
|
73 |
+
"loss": 0.5125,
|
74 |
"step": 55
|
75 |
},
|
76 |
{
|
77 |
"epoch": 0.13,
|
78 |
+
"learning_rate": 0.00013346685662502967,
|
79 |
+
"loss": 0.5314,
|
80 |
"step": 60
|
81 |
},
|
82 |
{
|
83 |
"epoch": 0.14,
|
84 |
+
"learning_rate": 0.00013265775190169146,
|
85 |
+
"loss": 0.4813,
|
86 |
"step": 65
|
87 |
},
|
88 |
{
|
89 |
"epoch": 0.15,
|
90 |
+
"learning_rate": 0.00013177230834485767,
|
91 |
+
"loss": 0.4348,
|
92 |
"step": 70
|
93 |
},
|
94 |
{
|
95 |
"epoch": 0.16,
|
96 |
+
"learning_rate": 0.00013081158115434713,
|
97 |
+
"loss": 0.4715,
|
98 |
"step": 75
|
99 |
},
|
100 |
{
|
101 |
+
"epoch": 0.18,
|
102 |
+
"learning_rate": 0.00012977671524691005,
|
103 |
+
"loss": 0.4864,
|
104 |
"step": 80
|
105 |
},
|
106 |
{
|
107 |
+
"epoch": 0.19,
|
108 |
+
"learning_rate": 0.0001286689438918043,
|
109 |
+
"loss": 0.4764,
|
110 |
"step": 85
|
111 |
},
|
112 |
{
|
113 |
"epoch": 0.2,
|
114 |
+
"learning_rate": 0.0001274895872410883,
|
115 |
+
"loss": 0.5284,
|
116 |
"step": 90
|
117 |
},
|
118 |
{
|
119 |
"epoch": 0.21,
|
120 |
+
"learning_rate": 0.0001262400507563689,
|
121 |
+
"loss": 0.5136,
|
122 |
"step": 95
|
123 |
},
|
124 |
{
|
125 |
"epoch": 0.22,
|
126 |
+
"learning_rate": 0.00012492182353389172,
|
127 |
+
"loss": 0.5015,
|
128 |
"step": 100
|
129 |
},
|
130 |
{
|
131 |
"epoch": 0.23,
|
132 |
+
"learning_rate": 0.00012353647652994513,
|
133 |
+
"loss": 0.5157,
|
134 |
"step": 105
|
135 |
},
|
136 |
{
|
137 |
"epoch": 0.24,
|
138 |
+
"learning_rate": 0.00012208566068873714,
|
139 |
+
"loss": 0.5087,
|
140 |
"step": 110
|
141 |
},
|
142 |
{
|
143 |
"epoch": 0.25,
|
144 |
+
"learning_rate": 0.00012057110497492192,
|
145 |
+
"loss": 0.5977,
|
146 |
"step": 115
|
147 |
},
|
148 |
{
|
149 |
"epoch": 0.26,
|
150 |
+
"learning_rate": 0.00011899461431316551,
|
151 |
+
"loss": 0.4527,
|
152 |
"step": 120
|
153 |
},
|
154 |
{
|
155 |
"epoch": 0.27,
|
156 |
+
"learning_rate": 0.00011735806743718042,
|
157 |
+
"loss": 0.4758,
|
158 |
"step": 125
|
159 |
},
|
160 |
{
|
161 |
+
"epoch": 0.29,
|
162 |
+
"learning_rate": 0.00011566341465080868,
|
163 |
+
"loss": 0.4566,
|
164 |
"step": 130
|
165 |
},
|
166 |
{
|
167 |
+
"epoch": 0.3,
|
168 |
+
"learning_rate": 0.00011391267550379983,
|
169 |
+
"loss": 0.4963,
|
170 |
"step": 135
|
171 |
},
|
172 |
{
|
173 |
+
"epoch": 0.31,
|
174 |
+
"learning_rate": 0.00011210793638508056,
|
175 |
+
"loss": 0.4994,
|
176 |
"step": 140
|
177 |
},
|
178 |
{
|
179 |
"epoch": 0.32,
|
180 |
+
"learning_rate": 0.00011025134803636436,
|
181 |
+
"loss": 0.5175,
|
182 |
"step": 145
|
183 |
},
|
184 |
{
|
185 |
"epoch": 0.33,
|
186 |
+
"learning_rate": 0.00010834512298907191,
|
187 |
+
"loss": 0.5139,
|
188 |
"step": 150
|
189 |
},
|
190 |
{
|
191 |
"epoch": 0.34,
|
192 |
+
"learning_rate": 0.00010639153292761406,
|
193 |
+
"loss": 0.515,
|
194 |
"step": 155
|
195 |
},
|
196 |
{
|
197 |
"epoch": 0.35,
|
198 |
+
"learning_rate": 0.0001043929059821889,
|
199 |
+
"loss": 0.4672,
|
200 |
"step": 160
|
201 |
},
|
202 |
{
|
203 |
"epoch": 0.36,
|
204 |
+
"learning_rate": 0.00010235162395429551,
|
205 |
+
"loss": 0.5449,
|
206 |
"step": 165
|
207 |
},
|
208 |
{
|
209 |
"epoch": 0.37,
|
210 |
+
"learning_rate": 0.00010027011947830235,
|
211 |
+
"loss": 0.5187,
|
212 |
"step": 170
|
213 |
},
|
214 |
{
|
215 |
"epoch": 0.38,
|
216 |
+
"learning_rate": 9.81508731224277e-05,
|
217 |
+
"loss": 0.5283,
|
218 |
"step": 175
|
219 |
},
|
220 |
{
|
221 |
+
"epoch": 0.4,
|
222 |
+
"learning_rate": 9.599641043260833e-05,
|
223 |
+
"loss": 0.4869,
|
224 |
"step": 180
|
225 |
},
|
226 |
{
|
227 |
+
"epoch": 0.41,
|
228 |
+
"learning_rate": 9.380929892273937e-05,
|
229 |
+
"loss": 0.4807,
|
230 |
"step": 185
|
231 |
},
|
232 |
{
|
233 |
+
"epoch": 0.42,
|
234 |
+
"learning_rate": 9.15921450149442e-05,
|
235 |
+
"loss": 0.5009,
|
236 |
"step": 190
|
237 |
},
|
238 |
{
|
239 |
+
"epoch": 0.43,
|
240 |
+
"learning_rate": 8.93475909334324e-05,
|
241 |
+
"loss": 0.5661,
|
242 |
"step": 195
|
243 |
},
|
244 |
{
|
245 |
+
"epoch": 0.44,
|
246 |
+
"learning_rate": 8.70783115557202e-05,
|
247 |
+
"loss": 0.5181,
|
248 |
"step": 200
|
249 |
},
|
250 |
{
|
251 |
"epoch": 0.45,
|
252 |
+
"learning_rate": 8.478701122492464e-05,
|
253 |
+
"loss": 0.5132,
|
254 |
"step": 205
|
255 |
},
|
256 |
{
|
257 |
"epoch": 0.46,
|
258 |
+
"learning_rate": 8.247642052695435e-05,
|
259 |
+
"loss": 0.4738,
|
260 |
"step": 210
|
261 |
},
|
262 |
{
|
263 |
"epoch": 0.47,
|
264 |
+
"learning_rate": 8.014929303640746e-05,
|
265 |
+
"loss": 0.4718,
|
266 |
"step": 215
|
267 |
},
|
268 |
{
|
269 |
"epoch": 0.48,
|
270 |
+
"learning_rate": 7.780840203509217e-05,
|
271 |
+
"loss": 0.5202,
|
272 |
"step": 220
|
273 |
},
|
274 |
{
|
275 |
"epoch": 0.49,
|
276 |
+
"learning_rate": 7.545653720705064e-05,
|
277 |
+
"loss": 0.4643,
|
278 |
"step": 225
|
279 |
},
|
280 |
{
|
281 |
+
"epoch": 0.51,
|
282 |
+
"learning_rate": 7.309650131404948e-05,
|
283 |
+
"loss": 0.4846,
|
284 |
"step": 230
|
285 |
},
|
286 |
{
|
287 |
+
"epoch": 0.52,
|
288 |
+
"learning_rate": 7.073110685545424e-05,
|
289 |
+
"loss": 0.4528,
|
290 |
"step": 235
|
291 |
},
|
292 |
{
|
293 |
+
"epoch": 0.53,
|
294 |
+
"learning_rate": 6.836317271654793e-05,
|
295 |
+
"loss": 0.4695,
|
296 |
"step": 240
|
297 |
},
|
298 |
{
|
299 |
+
"epoch": 0.54,
|
300 |
+
"learning_rate": 6.599552080919084e-05,
|
301 |
+
"loss": 0.4723,
|
302 |
"step": 245
|
303 |
},
|
304 |
{
|
305 |
+
"epoch": 0.55,
|
306 |
+
"learning_rate": 6.363097270890452e-05,
|
307 |
+
"loss": 0.4357,
|
308 |
"step": 250
|
309 |
},
|
310 |
{
|
311 |
+
"epoch": 0.56,
|
312 |
+
"learning_rate": 6.127234629234303e-05,
|
313 |
+
"loss": 0.5521,
|
314 |
"step": 255
|
315 |
},
|
316 |
{
|
317 |
"epoch": 0.57,
|
318 |
+
"learning_rate": 5.892245237918387e-05,
|
319 |
+
"loss": 0.484,
|
320 |
"step": 260
|
321 |
},
|
322 |
{
|
323 |
"epoch": 0.58,
|
324 |
+
"learning_rate": 5.6584091382409376e-05,
|
325 |
+
"loss": 0.5221,
|
326 |
"step": 265
|
327 |
},
|
328 |
{
|
329 |
"epoch": 0.59,
|
330 |
+
"learning_rate": 5.426004997100939e-05,
|
331 |
+
"loss": 0.4865,
|
332 |
"step": 270
|
333 |
},
|
334 |
{
|
335 |
"epoch": 0.6,
|
336 |
+
"learning_rate": 5.195309774905142e-05,
|
337 |
+
"loss": 0.4395,
|
338 |
"step": 275
|
339 |
},
|
340 |
{
|
341 |
+
"epoch": 0.62,
|
342 |
+
"learning_rate": 4.966598395510049e-05,
|
343 |
+
"loss": 0.4693,
|
344 |
"step": 280
|
345 |
},
|
346 |
{
|
347 |
+
"epoch": 0.63,
|
348 |
+
"learning_rate": 4.7401434185878876e-05,
|
349 |
+
"loss": 0.5105,
|
350 |
"step": 285
|
351 |
},
|
352 |
{
|
353 |
+
"epoch": 0.64,
|
354 |
+
"learning_rate": 4.516214714814733e-05,
|
355 |
+
"loss": 0.406,
|
356 |
"step": 290
|
357 |
},
|
358 |
{
|
359 |
+
"epoch": 0.65,
|
360 |
+
"learning_rate": 4.2950791442584565e-05,
|
361 |
+
"loss": 0.4573,
|
362 |
"step": 295
|
363 |
},
|
364 |
{
|
365 |
+
"epoch": 0.66,
|
366 |
+
"learning_rate": 4.0770002383574794e-05,
|
367 |
+
"loss": 0.4153,
|
368 |
"step": 300
|
369 |
},
|
370 |
{
|
371 |
+
"epoch": 0.67,
|
372 |
+
"learning_rate": 3.862237885865012e-05,
|
373 |
+
"loss": 0.4713,
|
374 |
"step": 305
|
375 |
},
|
376 |
{
|
377 |
+
"epoch": 0.68,
|
378 |
+
"learning_rate": 3.651048023135423e-05,
|
379 |
+
"loss": 0.4615,
|
380 |
"step": 310
|
381 |
},
|
382 |
{
|
383 |
+
"epoch": 0.69,
|
384 |
+
"learning_rate": 3.4436823291189834e-05,
|
385 |
+
"loss": 0.4731,
|
386 |
"step": 315
|
387 |
},
|
388 |
{
|
389 |
"epoch": 0.7,
|
390 |
+
"learning_rate": 3.2403879254320225e-05,
|
391 |
+
"loss": 0.4743,
|
392 |
"step": 320
|
393 |
},
|
394 |
{
|
395 |
"epoch": 0.71,
|
396 |
+
"learning_rate": 3.0414070818570786e-05,
|
397 |
+
"loss": 0.4617,
|
398 |
"step": 325
|
399 |
},
|
400 |
{
|
401 |
+
"epoch": 0.73,
|
402 |
+
"learning_rate": 2.84697692762624e-05,
|
403 |
+
"loss": 0.4779,
|
404 |
"step": 330
|
405 |
},
|
406 |
{
|
407 |
+
"epoch": 0.74,
|
408 |
+
"learning_rate": 2.657329168827975e-05,
|
409 |
+
"loss": 0.4796,
|
410 |
"step": 335
|
411 |
},
|
412 |
{
|
413 |
+
"epoch": 0.75,
|
414 |
+
"learning_rate": 2.4726898122808552e-05,
|
415 |
+
"loss": 0.451,
|
416 |
"step": 340
|
417 |
},
|
418 |
{
|
419 |
+
"epoch": 0.76,
|
420 |
+
"learning_rate": 2.2932788961952088e-05,
|
421 |
+
"loss": 0.5316,
|
422 |
"step": 345
|
423 |
},
|
424 |
{
|
425 |
+
"epoch": 0.77,
|
426 |
+
"learning_rate": 2.1193102279502168e-05,
|
427 |
+
"loss": 0.4846,
|
428 |
"step": 350
|
429 |
},
|
430 |
{
|
431 |
+
"epoch": 0.78,
|
432 |
+
"learning_rate": 1.9509911292953254e-05,
|
433 |
+
"loss": 0.4435,
|
434 |
"step": 355
|
435 |
},
|
436 |
{
|
437 |
+
"epoch": 0.79,
|
438 |
+
"learning_rate": 1.7885221892815502e-05,
|
439 |
+
"loss": 0.4859,
|
440 |
"step": 360
|
441 |
},
|
442 |
{
|
443 |
+
"epoch": 0.8,
|
444 |
+
"learning_rate": 1.6320970252148786e-05,
|
445 |
+
"loss": 0.5439,
|
446 |
"step": 365
|
447 |
},
|
448 |
{
|
449 |
+
"epoch": 0.81,
|
450 |
+
"learning_rate": 1.48190205191943e-05,
|
451 |
+
"loss": 0.4491,
|
452 |
"step": 370
|
453 |
},
|
454 |
{
|
455 |
"epoch": 0.82,
|
456 |
+
"learning_rate": 1.3381162595831284e-05,
|
457 |
+
"loss": 0.4599,
|
458 |
"step": 375
|
459 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
460 |
{
|
461 |
"epoch": 0.84,
|
462 |
+
"learning_rate": 1.2009110004522916e-05,
|
463 |
+
"loss": 0.4688,
|
464 |
+
"step": 380
|
465 |
},
|
466 |
{
|
467 |
"epoch": 0.85,
|
468 |
+
"learning_rate": 1.0704497846266602e-05,
|
469 |
+
"loss": 0.445,
|
470 |
+
"step": 385
|
471 |
},
|
472 |
{
|
473 |
"epoch": 0.86,
|
474 |
+
"learning_rate": 9.468880852029007e-06,
|
475 |
+
"loss": 0.4632,
|
476 |
+
"step": 390
|
477 |
},
|
478 |
{
|
479 |
"epoch": 0.87,
|
480 |
+
"learning_rate": 8.303731529931584e-06,
|
481 |
+
"loss": 0.4699,
|
482 |
+
"step": 395
|
483 |
},
|
484 |
{
|
485 |
"epoch": 0.88,
|
486 |
+
"learning_rate": 7.210438410440295e-06,
|
487 |
+
"loss": 0.4994,
|
488 |
+
"step": 400
|
489 |
},
|
490 |
{
|
491 |
"epoch": 0.89,
|
492 |
+
"learning_rate": 6.190304391625742e-06,
|
493 |
+
"loss": 0.4727,
|
494 |
+
"step": 405
|
495 |
},
|
496 |
{
|
497 |
"epoch": 0.9,
|
498 |
+
"learning_rate": 5.24454518647817e-06,
|
499 |
+
"loss": 0.4766,
|
500 |
+
"step": 410
|
501 |
},
|
502 |
{
|
503 |
"epoch": 0.91,
|
504 |
+
"learning_rate": 4.374287874114302e-06,
|
505 |
+
"loss": 0.5309,
|
506 |
+
"step": 415
|
507 |
},
|
508 |
{
|
509 |
"epoch": 0.92,
|
510 |
+
"learning_rate": 3.5805695566192774e-06,
|
511 |
+
"loss": 0.4346,
|
512 |
+
"step": 420
|
513 |
},
|
514 |
{
|
515 |
"epoch": 0.93,
|
516 |
+
"learning_rate": 2.8643361231110826e-06,
|
517 |
+
"loss": 0.4458,
|
518 |
+
"step": 425
|
519 |
},
|
520 |
{
|
521 |
"epoch": 0.95,
|
522 |
+
"learning_rate": 2.2264411225093805e-06,
|
523 |
+
"loss": 0.416,
|
524 |
+
"step": 430
|
525 |
},
|
526 |
{
|
527 |
"epoch": 0.96,
|
528 |
+
"learning_rate": 1.6676447463391393e-06,
|
529 |
+
"loss": 0.524,
|
530 |
+
"step": 435
|
531 |
},
|
532 |
{
|
533 |
"epoch": 0.97,
|
534 |
+
"learning_rate": 1.1886129228033631e-06,
|
535 |
+
"loss": 0.4421,
|
536 |
+
"step": 440
|
537 |
},
|
538 |
{
|
539 |
"epoch": 0.98,
|
540 |
+
"learning_rate": 7.89916523178351e-07,
|
541 |
+
"loss": 0.4325,
|
542 |
+
"step": 445
|
543 |
},
|
544 |
{
|
545 |
"epoch": 0.99,
|
546 |
+
"learning_rate": 4.7203068149803844e-07,
|
547 |
+
"loss": 0.476,
|
548 |
+
"step": 450
|
549 |
},
|
550 |
{
|
551 |
"epoch": 1.0,
|
552 |
+
"learning_rate": 2.3533422832716514e-07,
|
553 |
+
"loss": 0.4506,
|
554 |
+
"step": 455
|
555 |
},
|
556 |
{
|
557 |
"epoch": 1.0,
|
558 |
+
"eval_loss": 0.22213919460773468,
|
559 |
+
"eval_runtime": 29.39,
|
560 |
+
"eval_samples_per_second": 21.946,
|
561 |
+
"eval_steps_per_second": 2.756,
|
562 |
+
"step": 455
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
563 |
}
|
564 |
],
|
565 |
+
"max_steps": 1365,
|
566 |
+
"num_train_epochs": 3,
|
567 |
+
"total_flos": 474636976128000.0,
|
568 |
"trial_name": null,
|
569 |
"trial_params": null
|
570 |
}
|
training_args.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 sha256:5bc291dc42b38a7b304b7daeb4b46cfc64328ad099cb5a00588b0246f606e58b
|
3 |
+
size 3247
|