Model save
Browse files- README.md +12 -6
- adapter_model.safetensors +1 -1
- all_results.json +10 -10
- eval_results.json +5 -5
- train_results.json +6 -6
- trainer_state.json +590 -600
README.md
CHANGED
@@ -1,10 +1,15 @@
|
|
1 |
---
|
2 |
base_model: hllj/mistral-vi-math
|
3 |
tags:
|
|
|
|
|
4 |
- generated_from_trainer
|
|
|
|
|
5 |
model-index:
|
6 |
- name: sft-mistral-v1-clean-valid
|
7 |
results: []
|
|
|
8 |
---
|
9 |
|
10 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -12,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
12 |
|
13 |
# sft-mistral-v1-clean-valid
|
14 |
|
15 |
-
This model is a fine-tuned version of [hllj/mistral-vi-math](https://huggingface.co/hllj/mistral-vi-math) on
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- Loss: 0.
|
18 |
|
19 |
## Model description
|
20 |
|
@@ -48,12 +53,13 @@ The following hyperparameters were used during training:
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|
|
51 |
-
| 0.
|
52 |
|
53 |
|
54 |
### Framework versions
|
55 |
|
56 |
-
-
|
57 |
-
-
|
58 |
-
-
|
|
|
59 |
- Tokenizers 0.15.0
|
|
|
1 |
---
|
2 |
base_model: hllj/mistral-vi-math
|
3 |
tags:
|
4 |
+
- trl
|
5 |
+
- sft
|
6 |
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- generator
|
9 |
model-index:
|
10 |
- name: sft-mistral-v1-clean-valid
|
11 |
results: []
|
12 |
+
library_name: peft
|
13 |
---
|
14 |
|
15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
17 |
|
18 |
# sft-mistral-v1-clean-valid
|
19 |
|
20 |
+
This model is a fine-tuned version of [hllj/mistral-vi-math](https://huggingface.co/hllj/mistral-vi-math) on the generator dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.5101
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
53 |
|
54 |
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
|:-------------:|:-----:|:----:|:---------------:|
|
56 |
+
| 0.2676 | 1.05 | 1000 | 0.5086 |
|
57 |
|
58 |
|
59 |
### Framework versions
|
60 |
|
61 |
+
- PEFT 0.5.0
|
62 |
+
- Transformers 4.36.2
|
63 |
+
- Pytorch 2.1.2+cu121
|
64 |
+
- Datasets 2.16.0
|
65 |
- Tokenizers 0.15.0
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 872450448
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cde042ccf32ffa4b57b1b9c1dcb5e326931a6822190e834eea9e7bbedb9b468b
|
3 |
size 872450448
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss": 0.
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 140,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
-
"train_loss": 0.
|
9 |
-
"train_runtime":
|
10 |
-
"train_samples":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second":
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_loss": 0.5100870728492737,
|
4 |
+
"eval_runtime": 12.2379,
|
5 |
"eval_samples": 140,
|
6 |
+
"eval_samples_per_second": 3.595,
|
7 |
+
"eval_steps_per_second": 0.899,
|
8 |
+
"train_loss": 0.3227183536165639,
|
9 |
+
"train_runtime": 6710.5312,
|
10 |
+
"train_samples": 8517,
|
11 |
+
"train_samples_per_second": 1.132,
|
12 |
+
"train_steps_per_second": 0.283
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss": 0.
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 140,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_loss": 0.5100870728492737,
|
4 |
+
"eval_runtime": 12.2379,
|
5 |
"eval_samples": 140,
|
6 |
+
"eval_samples_per_second": 3.595,
|
7 |
+
"eval_steps_per_second": 0.899
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"train_loss": 0.3227183536165639,
|
4 |
+
"train_runtime": 6710.5312,
|
5 |
+
"train_samples": 8517,
|
6 |
+
"train_samples_per_second": 1.132,
|
7 |
+
"train_steps_per_second": 0.283
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,1194 +1,1184 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 1000,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0,
|
13 |
-
"learning_rate":
|
14 |
-
"loss": 0.
|
15 |
"step": 1
|
16 |
},
|
17 |
{
|
18 |
-
"epoch": 0.
|
19 |
-
"learning_rate":
|
20 |
-
"loss": 0.
|
21 |
"step": 10
|
22 |
},
|
23 |
{
|
24 |
-
"epoch": 0.
|
25 |
-
"learning_rate":
|
26 |
-
"loss": 0.
|
27 |
"step": 20
|
28 |
},
|
29 |
{
|
30 |
-
"epoch": 0.
|
31 |
-
"learning_rate":
|
32 |
-
"loss": 0.
|
33 |
"step": 30
|
34 |
},
|
35 |
{
|
36 |
-
"epoch": 0.
|
37 |
-
"learning_rate":
|
38 |
-
"loss": 0.
|
39 |
"step": 40
|
40 |
},
|
41 |
{
|
42 |
-
"epoch": 0.
|
43 |
-
"learning_rate":
|
44 |
-
"loss": 0.
|
45 |
"step": 50
|
46 |
},
|
47 |
{
|
48 |
-
"epoch": 0.
|
49 |
-
"learning_rate":
|
50 |
-
"loss": 0.
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
-
"epoch": 0.
|
55 |
-
"learning_rate":
|
56 |
-
"loss": 0.
|
57 |
"step": 70
|
58 |
},
|
59 |
{
|
60 |
-
"epoch": 0.
|
61 |
-
"learning_rate":
|
62 |
-
"loss": 0.
|
63 |
"step": 80
|
64 |
},
|
65 |
{
|
66 |
-
"epoch": 0.
|
67 |
-
"learning_rate":
|
68 |
-
"loss": 0.
|
69 |
"step": 90
|
70 |
},
|
71 |
{
|
72 |
-
"epoch": 0.
|
73 |
-
"learning_rate":
|
74 |
-
"loss": 0.
|
75 |
"step": 100
|
76 |
},
|
77 |
{
|
78 |
-
"epoch": 0.
|
79 |
-
"learning_rate":
|
80 |
-
"loss": 0.
|
81 |
"step": 110
|
82 |
},
|
83 |
{
|
84 |
-
"epoch": 0.
|
85 |
-
"learning_rate":
|
86 |
-
"loss": 0.
|
87 |
"step": 120
|
88 |
},
|
89 |
{
|
90 |
-
"epoch": 0.
|
91 |
-
"learning_rate":
|
92 |
-
"loss": 0.
|
93 |
"step": 130
|
94 |
},
|
95 |
{
|
96 |
-
"epoch": 0.
|
97 |
-
"learning_rate":
|
98 |
-
"loss": 0.
|
99 |
"step": 140
|
100 |
},
|
101 |
{
|
102 |
-
"epoch": 0.
|
103 |
-
"learning_rate": 2.
|
104 |
-
"loss": 0.
|
105 |
"step": 150
|
106 |
},
|
107 |
{
|
108 |
-
"epoch": 0.
|
109 |
-
"learning_rate": 2.
|
110 |
-
"loss": 0.
|
111 |
"step": 160
|
112 |
},
|
113 |
{
|
114 |
-
"epoch": 0.
|
115 |
-
"learning_rate": 2.
|
116 |
-
"loss": 0.
|
117 |
"step": 170
|
118 |
},
|
119 |
{
|
120 |
-
"epoch": 0.
|
121 |
-
"learning_rate": 2.
|
122 |
-
"loss": 0.
|
123 |
"step": 180
|
124 |
},
|
125 |
{
|
126 |
-
"epoch": 0.
|
127 |
-
"learning_rate": 2.
|
128 |
-
"loss": 0.
|
129 |
"step": 190
|
130 |
},
|
131 |
{
|
132 |
-
"epoch": 0.
|
133 |
-
"learning_rate": 2.
|
134 |
-
"loss": 0.
|
135 |
"step": 200
|
136 |
},
|
137 |
{
|
138 |
-
"epoch": 0.
|
139 |
-
"learning_rate": 2.
|
140 |
-
"loss": 0.
|
141 |
"step": 210
|
142 |
},
|
143 |
{
|
144 |
-
"epoch": 0.
|
145 |
-
"learning_rate": 2.
|
146 |
-
"loss": 0.
|
147 |
"step": 220
|
148 |
},
|
149 |
{
|
150 |
-
"epoch": 0.
|
151 |
-
"learning_rate": 2.
|
152 |
-
"loss": 0.
|
153 |
"step": 230
|
154 |
},
|
155 |
{
|
156 |
-
"epoch": 0.
|
157 |
-
"learning_rate": 2.
|
158 |
-
"loss": 0.
|
159 |
"step": 240
|
160 |
},
|
161 |
{
|
162 |
-
"epoch": 0.
|
163 |
-
"learning_rate": 2.
|
164 |
-
"loss": 0.
|
165 |
"step": 250
|
166 |
},
|
167 |
{
|
168 |
-
"epoch": 0.
|
169 |
-
"learning_rate": 2.
|
170 |
-
"loss": 0.
|
171 |
"step": 260
|
172 |
},
|
173 |
{
|
174 |
-
"epoch": 0.
|
175 |
-
"learning_rate": 2.
|
176 |
-
"loss": 0.
|
177 |
"step": 270
|
178 |
},
|
179 |
{
|
180 |
-
"epoch": 0.
|
181 |
-
"learning_rate": 2.
|
182 |
-
"loss": 0.
|
183 |
"step": 280
|
184 |
},
|
185 |
{
|
186 |
-
"epoch": 0.
|
187 |
-
"learning_rate": 2.
|
188 |
-
"loss": 0.
|
189 |
"step": 290
|
190 |
},
|
191 |
{
|
192 |
-
"epoch": 0.
|
193 |
-
"learning_rate": 2.
|
194 |
-
"loss": 0.
|
195 |
"step": 300
|
196 |
},
|
197 |
{
|
198 |
-
"epoch": 0.
|
199 |
-
"learning_rate": 2.
|
200 |
-
"loss": 0.
|
201 |
"step": 310
|
202 |
},
|
203 |
{
|
204 |
-
"epoch": 0.
|
205 |
-
"learning_rate": 2.
|
206 |
-
"loss": 0.
|
207 |
"step": 320
|
208 |
},
|
209 |
{
|
210 |
-
"epoch": 0.
|
211 |
-
"learning_rate": 2.
|
212 |
-
"loss": 0.
|
213 |
"step": 330
|
214 |
},
|
215 |
{
|
216 |
-
"epoch": 0.
|
217 |
-
"learning_rate": 2.
|
218 |
-
"loss": 0.
|
219 |
"step": 340
|
220 |
},
|
221 |
{
|
222 |
-
"epoch": 0.
|
223 |
-
"learning_rate": 2.
|
224 |
-
"loss": 0.
|
225 |
"step": 350
|
226 |
},
|
227 |
{
|
228 |
-
"epoch": 0.
|
229 |
-
"learning_rate": 2.
|
230 |
-
"loss": 0.
|
231 |
"step": 360
|
232 |
},
|
233 |
{
|
234 |
-
"epoch": 0.
|
235 |
-
"learning_rate": 2.
|
236 |
-
"loss": 0.
|
237 |
"step": 370
|
238 |
},
|
239 |
{
|
240 |
-
"epoch": 0.
|
241 |
-
"learning_rate": 2.
|
242 |
-
"loss": 0.
|
243 |
"step": 380
|
244 |
},
|
245 |
{
|
246 |
-
"epoch": 0.
|
247 |
-
"learning_rate": 2.
|
248 |
-
"loss": 0.
|
249 |
"step": 390
|
250 |
},
|
251 |
{
|
252 |
-
"epoch": 0.
|
253 |
-
"learning_rate": 2.
|
254 |
-
"loss": 0.
|
255 |
"step": 400
|
256 |
},
|
257 |
{
|
258 |
-
"epoch": 0.
|
259 |
-
"learning_rate": 2.
|
260 |
-
"loss": 0.
|
261 |
"step": 410
|
262 |
},
|
263 |
{
|
264 |
-
"epoch": 0.
|
265 |
-
"learning_rate": 2.
|
266 |
-
"loss": 0.
|
267 |
"step": 420
|
268 |
},
|
269 |
{
|
270 |
-
"epoch": 0.
|
271 |
-
"learning_rate": 2.
|
272 |
-
"loss": 0.
|
273 |
"step": 430
|
274 |
},
|
275 |
{
|
276 |
-
"epoch": 0.
|
277 |
-
"learning_rate": 2.
|
278 |
-
"loss": 0.
|
279 |
"step": 440
|
280 |
},
|
281 |
{
|
282 |
-
"epoch": 0.
|
283 |
-
"learning_rate": 2.
|
284 |
-
"loss": 0.
|
285 |
"step": 450
|
286 |
},
|
287 |
{
|
288 |
-
"epoch": 0.
|
289 |
-
"learning_rate": 2.
|
290 |
-
"loss": 0.
|
291 |
"step": 460
|
292 |
},
|
293 |
{
|
294 |
-
"epoch": 0.
|
295 |
-
"learning_rate": 2.
|
296 |
-
"loss": 0.
|
297 |
"step": 470
|
298 |
},
|
299 |
{
|
300 |
-
"epoch": 0.
|
301 |
-
"learning_rate": 2.
|
302 |
-
"loss": 0.
|
303 |
"step": 480
|
304 |
},
|
305 |
{
|
306 |
-
"epoch": 0.
|
307 |
-
"learning_rate": 2.
|
308 |
-
"loss": 0.
|
309 |
"step": 490
|
310 |
},
|
311 |
{
|
312 |
-
"epoch": 0.
|
313 |
-
"learning_rate": 2.
|
314 |
-
"loss": 0.
|
315 |
"step": 500
|
316 |
},
|
317 |
{
|
318 |
-
"epoch": 0.
|
319 |
-
"learning_rate": 2.
|
320 |
-
"loss": 0.
|
321 |
"step": 510
|
322 |
},
|
323 |
{
|
324 |
-
"epoch": 0.
|
325 |
-
"learning_rate": 2.
|
326 |
-
"loss": 0.
|
327 |
"step": 520
|
328 |
},
|
329 |
{
|
330 |
-
"epoch": 0.
|
331 |
-
"learning_rate": 2.
|
332 |
-
"loss": 0.
|
333 |
"step": 530
|
334 |
},
|
335 |
{
|
336 |
-
"epoch": 0.
|
337 |
-
"learning_rate": 2.
|
338 |
-
"loss": 0.
|
339 |
"step": 540
|
340 |
},
|
341 |
{
|
342 |
-
"epoch": 0.
|
343 |
-
"learning_rate": 2.
|
344 |
-
"loss": 0.
|
345 |
"step": 550
|
346 |
},
|
347 |
{
|
348 |
-
"epoch": 0.
|
349 |
-
"learning_rate": 2.
|
350 |
-
"loss": 0.
|
351 |
"step": 560
|
352 |
},
|
353 |
{
|
354 |
-
"epoch": 0.
|
355 |
-
"learning_rate": 2.
|
356 |
-
"loss": 0.
|
357 |
"step": 570
|
358 |
},
|
359 |
{
|
360 |
-
"epoch": 0.
|
361 |
-
"learning_rate": 2.
|
362 |
-
"loss": 0.
|
363 |
"step": 580
|
364 |
},
|
365 |
{
|
366 |
-
"epoch": 0.
|
367 |
-
"learning_rate": 2.
|
368 |
-
"loss": 0.
|
369 |
"step": 590
|
370 |
},
|
371 |
{
|
372 |
-
"epoch": 0.
|
373 |
-
"learning_rate": 2.
|
374 |
-
"loss": 0.
|
375 |
"step": 600
|
376 |
},
|
377 |
{
|
378 |
-
"epoch": 0.
|
379 |
-
"learning_rate": 2.
|
380 |
-
"loss": 0.
|
381 |
"step": 610
|
382 |
},
|
383 |
{
|
384 |
-
"epoch": 0.
|
385 |
-
"learning_rate": 2.
|
386 |
-
"loss": 0.
|
387 |
"step": 620
|
388 |
},
|
389 |
{
|
390 |
-
"epoch": 0.
|
391 |
-
"learning_rate": 2.
|
392 |
-
"loss": 0.
|
393 |
"step": 630
|
394 |
},
|
395 |
{
|
396 |
-
"epoch": 0.
|
397 |
-
"learning_rate": 2.
|
398 |
-
"loss": 0.
|
399 |
"step": 640
|
400 |
},
|
401 |
{
|
402 |
-
"epoch": 0.
|
403 |
-
"learning_rate": 2.
|
404 |
-
"loss": 0.
|
405 |
"step": 650
|
406 |
},
|
407 |
{
|
408 |
-
"epoch": 0.
|
409 |
-
"learning_rate": 2.
|
410 |
-
"loss": 0.
|
411 |
"step": 660
|
412 |
},
|
413 |
{
|
414 |
-
"epoch": 0.
|
415 |
-
"learning_rate": 2.
|
416 |
-
"loss": 0.
|
417 |
"step": 670
|
418 |
},
|
419 |
{
|
420 |
-
"epoch": 0.
|
421 |
-
"learning_rate": 2.
|
422 |
-
"loss": 0.
|
423 |
"step": 680
|
424 |
},
|
425 |
{
|
426 |
-
"epoch": 0.
|
427 |
-
"learning_rate": 2.
|
428 |
-
"loss": 0.
|
429 |
"step": 690
|
430 |
},
|
431 |
{
|
432 |
-
"epoch": 0.
|
433 |
-
"learning_rate": 2.
|
434 |
-
"loss": 0.
|
435 |
"step": 700
|
436 |
},
|
437 |
{
|
438 |
-
"epoch": 0.
|
439 |
-
"learning_rate": 2.
|
440 |
-
"loss": 0.
|
441 |
"step": 710
|
442 |
},
|
443 |
{
|
444 |
-
"epoch": 0.
|
445 |
-
"learning_rate": 2.
|
446 |
-
"loss": 0.
|
447 |
"step": 720
|
448 |
},
|
449 |
{
|
450 |
-
"epoch": 0.
|
451 |
-
"learning_rate": 2.
|
452 |
-
"loss": 0.
|
453 |
"step": 730
|
454 |
},
|
455 |
{
|
456 |
-
"epoch": 0.
|
457 |
-
"learning_rate": 2.
|
458 |
-
"loss": 0.
|
459 |
"step": 740
|
460 |
},
|
461 |
{
|
462 |
-
"epoch": 0.
|
463 |
-
"learning_rate": 2.
|
464 |
-
"loss": 0.
|
465 |
"step": 750
|
466 |
},
|
467 |
{
|
468 |
-
"epoch": 0.
|
469 |
-
"learning_rate": 2.
|
470 |
-
"loss": 0.
|
471 |
"step": 760
|
472 |
},
|
473 |
{
|
474 |
-
"epoch": 0.
|
475 |
-
"learning_rate": 2.
|
476 |
-
"loss": 0.
|
477 |
"step": 770
|
478 |
},
|
479 |
{
|
480 |
-
"epoch": 0.
|
481 |
-
"learning_rate": 2.
|
482 |
-
"loss": 0.
|
483 |
"step": 780
|
484 |
},
|
485 |
{
|
486 |
-
"epoch": 0.
|
487 |
-
"learning_rate": 2.
|
488 |
-
"loss": 0.
|
489 |
"step": 790
|
490 |
},
|
491 |
{
|
492 |
-
"epoch": 0.
|
493 |
-
"learning_rate": 2.
|
494 |
-
"loss": 0.
|
495 |
"step": 800
|
496 |
},
|
497 |
{
|
498 |
-
"epoch": 0.
|
499 |
-
"learning_rate":
|
500 |
-
"loss": 0.
|
501 |
"step": 810
|
502 |
},
|
503 |
{
|
504 |
-
"epoch": 0.
|
505 |
-
"learning_rate":
|
506 |
-
"loss": 0.
|
507 |
"step": 820
|
508 |
},
|
509 |
{
|
510 |
-
"epoch": 0.
|
511 |
-
"learning_rate":
|
512 |
-
"loss": 0.
|
513 |
"step": 830
|
514 |
},
|
515 |
{
|
516 |
-
"epoch": 0.
|
517 |
-
"learning_rate":
|
518 |
-
"loss": 0.
|
519 |
"step": 840
|
520 |
},
|
521 |
{
|
522 |
-
"epoch": 0.
|
523 |
-
"learning_rate":
|
524 |
-
"loss": 0.
|
525 |
"step": 850
|
526 |
},
|
527 |
{
|
528 |
-
"epoch": 0.
|
529 |
-
"learning_rate":
|
530 |
-
"loss": 0.
|
531 |
"step": 860
|
532 |
},
|
533 |
{
|
534 |
-
"epoch": 0.
|
535 |
-
"learning_rate":
|
536 |
-
"loss": 0.
|
537 |
"step": 870
|
538 |
},
|
539 |
{
|
540 |
-
"epoch": 0.
|
541 |
-
"learning_rate":
|
542 |
-
"loss": 0.
|
543 |
"step": 880
|
544 |
},
|
545 |
{
|
546 |
-
"epoch": 0.
|
547 |
-
"learning_rate":
|
548 |
-
"loss": 0.
|
549 |
"step": 890
|
550 |
},
|
551 |
{
|
552 |
-
"epoch": 0.
|
553 |
-
"learning_rate":
|
554 |
-
"loss": 0.
|
555 |
"step": 900
|
556 |
},
|
557 |
{
|
558 |
-
"epoch": 0.
|
559 |
-
"learning_rate":
|
560 |
-
"loss": 0.
|
561 |
"step": 910
|
562 |
},
|
563 |
{
|
564 |
-
"epoch": 0.
|
565 |
-
"learning_rate":
|
566 |
-
"loss": 0.
|
567 |
"step": 920
|
568 |
},
|
569 |
{
|
570 |
-
"epoch": 0.
|
571 |
-
"learning_rate":
|
572 |
-
"loss": 0.
|
573 |
"step": 930
|
574 |
},
|
575 |
{
|
576 |
-
"epoch": 0.
|
577 |
-
"learning_rate":
|
578 |
-
"loss": 0.
|
579 |
"step": 940
|
580 |
},
|
581 |
{
|
582 |
-
"epoch": 0
|
583 |
-
"learning_rate":
|
584 |
-
"loss": 0.
|
585 |
"step": 950
|
586 |
},
|
587 |
{
|
588 |
-
"epoch":
|
589 |
-
"learning_rate":
|
590 |
-
"loss": 0.
|
591 |
"step": 960
|
592 |
},
|
593 |
{
|
594 |
-
"epoch": 1.
|
595 |
-
"learning_rate":
|
596 |
-
"loss": 0.
|
597 |
"step": 970
|
598 |
},
|
599 |
{
|
600 |
-
"epoch": 1.
|
601 |
-
"learning_rate":
|
602 |
-
"loss": 0.
|
603 |
"step": 980
|
604 |
},
|
605 |
{
|
606 |
-
"epoch": 1.
|
607 |
-
"learning_rate":
|
608 |
-
"loss": 0.
|
609 |
"step": 990
|
610 |
},
|
611 |
{
|
612 |
-
"epoch": 1.
|
613 |
-
"learning_rate":
|
614 |
-
"loss": 0.
|
615 |
"step": 1000
|
616 |
},
|
617 |
{
|
618 |
-
"epoch": 1.
|
619 |
-
"eval_loss": 0.
|
620 |
-
"eval_runtime":
|
621 |
-
"eval_samples_per_second":
|
622 |
-
"eval_steps_per_second":
|
623 |
"step": 1000
|
624 |
},
|
625 |
{
|
626 |
-
"epoch": 1.
|
627 |
-
"learning_rate":
|
628 |
-
"loss": 0.
|
629 |
"step": 1010
|
630 |
},
|
631 |
{
|
632 |
-
"epoch": 1.
|
633 |
-
"learning_rate":
|
634 |
-
"loss": 0.
|
635 |
"step": 1020
|
636 |
},
|
637 |
{
|
638 |
-
"epoch": 1.
|
639 |
-
"learning_rate":
|
640 |
-
"loss": 0.
|
641 |
"step": 1030
|
642 |
},
|
643 |
{
|
644 |
-
"epoch": 1.
|
645 |
-
"learning_rate":
|
646 |
-
"loss": 0.
|
647 |
"step": 1040
|
648 |
},
|
649 |
{
|
650 |
-
"epoch": 1.
|
651 |
-
"learning_rate":
|
652 |
-
"loss": 0.
|
653 |
"step": 1050
|
654 |
},
|
655 |
{
|
656 |
-
"epoch": 1.
|
657 |
-
"learning_rate":
|
658 |
-
"loss": 0.
|
659 |
"step": 1060
|
660 |
},
|
661 |
{
|
662 |
-
"epoch": 1.
|
663 |
-
"learning_rate":
|
664 |
-
"loss": 0.
|
665 |
"step": 1070
|
666 |
},
|
667 |
{
|
668 |
-
"epoch": 1.
|
669 |
-
"learning_rate":
|
670 |
-
"loss": 0.
|
671 |
"step": 1080
|
672 |
},
|
673 |
{
|
674 |
-
"epoch": 1.
|
675 |
-
"learning_rate":
|
676 |
-
"loss": 0.
|
677 |
"step": 1090
|
678 |
},
|
679 |
{
|
680 |
-
"epoch": 1.
|
681 |
-
"learning_rate":
|
682 |
-
"loss": 0.
|
683 |
"step": 1100
|
684 |
},
|
685 |
{
|
686 |
-
"epoch": 1.
|
687 |
-
"learning_rate":
|
688 |
-
"loss": 0.
|
689 |
"step": 1110
|
690 |
},
|
691 |
{
|
692 |
-
"epoch": 1.
|
693 |
-
"learning_rate":
|
694 |
-
"loss": 0.
|
695 |
"step": 1120
|
696 |
},
|
697 |
{
|
698 |
-
"epoch": 1.
|
699 |
-
"learning_rate":
|
700 |
-
"loss": 0.
|
701 |
"step": 1130
|
702 |
},
|
703 |
{
|
704 |
-
"epoch": 1.
|
705 |
-
"learning_rate":
|
706 |
-
"loss": 0.
|
707 |
"step": 1140
|
708 |
},
|
709 |
{
|
710 |
-
"epoch": 1.
|
711 |
-
"learning_rate":
|
712 |
-
"loss": 0.
|
713 |
"step": 1150
|
714 |
},
|
715 |
{
|
716 |
-
"epoch": 1.
|
717 |
-
"learning_rate":
|
718 |
-
"loss": 0.
|
719 |
"step": 1160
|
720 |
},
|
721 |
{
|
722 |
-
"epoch": 1.
|
723 |
-
"learning_rate":
|
724 |
-
"loss": 0.
|
725 |
"step": 1170
|
726 |
},
|
727 |
{
|
728 |
-
"epoch": 1.
|
729 |
-
"learning_rate":
|
730 |
-
"loss": 0.
|
731 |
"step": 1180
|
732 |
},
|
733 |
{
|
734 |
-
"epoch": 1.
|
735 |
-
"learning_rate":
|
736 |
-
"loss": 0.
|
737 |
"step": 1190
|
738 |
},
|
739 |
{
|
740 |
-
"epoch": 1.
|
741 |
-
"learning_rate":
|
742 |
-
"loss": 0.
|
743 |
"step": 1200
|
744 |
},
|
745 |
{
|
746 |
-
"epoch": 1.
|
747 |
-
"learning_rate":
|
748 |
-
"loss": 0.
|
749 |
"step": 1210
|
750 |
},
|
751 |
{
|
752 |
-
"epoch": 1.
|
753 |
-
"learning_rate":
|
754 |
-
"loss": 0.
|
755 |
"step": 1220
|
756 |
},
|
757 |
{
|
758 |
-
"epoch": 1.
|
759 |
-
"learning_rate":
|
760 |
-
"loss": 0.
|
761 |
"step": 1230
|
762 |
},
|
763 |
{
|
764 |
-
"epoch": 1.
|
765 |
-
"learning_rate":
|
766 |
-
"loss": 0.
|
767 |
"step": 1240
|
768 |
},
|
769 |
{
|
770 |
-
"epoch": 1.
|
771 |
-
"learning_rate":
|
772 |
-
"loss": 0.
|
773 |
"step": 1250
|
774 |
},
|
775 |
{
|
776 |
-
"epoch": 1.
|
777 |
-
"learning_rate":
|
778 |
-
"loss": 0.
|
779 |
"step": 1260
|
780 |
},
|
781 |
{
|
782 |
-
"epoch": 1.
|
783 |
-
"learning_rate":
|
784 |
-
"loss": 0.
|
785 |
"step": 1270
|
786 |
},
|
787 |
{
|
788 |
-
"epoch": 1.
|
789 |
-
"learning_rate":
|
790 |
-
"loss": 0.
|
791 |
"step": 1280
|
792 |
},
|
793 |
{
|
794 |
-
"epoch": 1.
|
795 |
-
"learning_rate":
|
796 |
-
"loss": 0.
|
797 |
"step": 1290
|
798 |
},
|
799 |
{
|
800 |
-
"epoch": 1.
|
801 |
-
"learning_rate":
|
802 |
-
"loss": 0.
|
803 |
"step": 1300
|
804 |
},
|
805 |
{
|
806 |
-
"epoch": 1.
|
807 |
-
"learning_rate":
|
808 |
-
"loss": 0.
|
809 |
"step": 1310
|
810 |
},
|
811 |
{
|
812 |
-
"epoch": 1.
|
813 |
-
"learning_rate":
|
814 |
-
"loss": 0.
|
815 |
"step": 1320
|
816 |
},
|
817 |
{
|
818 |
-
"epoch": 1.
|
819 |
-
"learning_rate":
|
820 |
-
"loss": 0.
|
821 |
"step": 1330
|
822 |
},
|
823 |
{
|
824 |
-
"epoch": 1.
|
825 |
-
"learning_rate":
|
826 |
-
"loss": 0.
|
827 |
"step": 1340
|
828 |
},
|
829 |
{
|
830 |
-
"epoch": 1.
|
831 |
-
"learning_rate":
|
832 |
-
"loss": 0.
|
833 |
"step": 1350
|
834 |
},
|
835 |
{
|
836 |
-
"epoch": 1.
|
837 |
-
"learning_rate":
|
838 |
-
"loss": 0.
|
839 |
"step": 1360
|
840 |
},
|
841 |
{
|
842 |
-
"epoch": 1.
|
843 |
-
"learning_rate":
|
844 |
-
"loss": 0.
|
845 |
"step": 1370
|
846 |
},
|
847 |
{
|
848 |
-
"epoch": 1.
|
849 |
-
"learning_rate":
|
850 |
-
"loss": 0.
|
851 |
"step": 1380
|
852 |
},
|
853 |
{
|
854 |
-
"epoch": 1.
|
855 |
-
"learning_rate":
|
856 |
-
"loss": 0.
|
857 |
"step": 1390
|
858 |
},
|
859 |
{
|
860 |
-
"epoch": 1.
|
861 |
-
"learning_rate":
|
862 |
-
"loss": 0.
|
863 |
"step": 1400
|
864 |
},
|
865 |
{
|
866 |
-
"epoch": 1.
|
867 |
-
"learning_rate":
|
868 |
-
"loss": 0.
|
869 |
"step": 1410
|
870 |
},
|
871 |
{
|
872 |
-
"epoch": 1.
|
873 |
-
"learning_rate":
|
874 |
-
"loss": 0.
|
875 |
"step": 1420
|
876 |
},
|
877 |
{
|
878 |
-
"epoch": 1.
|
879 |
-
"learning_rate":
|
880 |
-
"loss": 0.
|
881 |
"step": 1430
|
882 |
},
|
883 |
{
|
884 |
-
"epoch": 1.
|
885 |
-
"learning_rate":
|
886 |
-
"loss": 0.
|
887 |
"step": 1440
|
888 |
},
|
889 |
{
|
890 |
-
"epoch": 1.
|
891 |
-
"learning_rate":
|
892 |
-
"loss": 0.
|
893 |
"step": 1450
|
894 |
},
|
895 |
{
|
896 |
-
"epoch": 1.
|
897 |
-
"learning_rate":
|
898 |
-
"loss": 0.
|
899 |
"step": 1460
|
900 |
},
|
901 |
{
|
902 |
-
"epoch": 1.
|
903 |
-
"learning_rate":
|
904 |
-
"loss": 0.
|
905 |
"step": 1470
|
906 |
},
|
907 |
{
|
908 |
-
"epoch": 1.
|
909 |
-
"learning_rate":
|
910 |
-
"loss": 0.
|
911 |
"step": 1480
|
912 |
},
|
913 |
{
|
914 |
-
"epoch": 1.
|
915 |
-
"learning_rate":
|
916 |
-
"loss": 0.
|
917 |
"step": 1490
|
918 |
},
|
919 |
{
|
920 |
-
"epoch": 1.
|
921 |
-
"learning_rate":
|
922 |
-
"loss": 0.
|
923 |
"step": 1500
|
924 |
},
|
925 |
{
|
926 |
-
"epoch": 1.
|
927 |
-
"learning_rate":
|
928 |
-
"loss": 0.
|
929 |
"step": 1510
|
930 |
},
|
931 |
{
|
932 |
-
"epoch": 1.
|
933 |
-
"learning_rate":
|
934 |
-
"loss": 0.
|
935 |
"step": 1520
|
936 |
},
|
937 |
{
|
938 |
-
"epoch": 1.
|
939 |
-
"learning_rate":
|
940 |
-
"loss": 0.
|
941 |
"step": 1530
|
942 |
},
|
943 |
{
|
944 |
-
"epoch": 1.
|
945 |
-
"learning_rate": 2.
|
946 |
-
"loss": 0.
|
947 |
"step": 1540
|
948 |
},
|
949 |
{
|
950 |
-
"epoch": 1.
|
951 |
-
"learning_rate": 2.
|
952 |
-
"loss": 0.
|
953 |
"step": 1550
|
954 |
},
|
955 |
{
|
956 |
-
"epoch": 1.
|
957 |
-
"learning_rate": 2.
|
958 |
-
"loss": 0.
|
959 |
"step": 1560
|
960 |
},
|
961 |
{
|
962 |
-
"epoch": 1.
|
963 |
-
"learning_rate": 2.
|
964 |
-
"loss": 0.
|
965 |
"step": 1570
|
966 |
},
|
967 |
{
|
968 |
-
"epoch": 1.
|
969 |
-
"learning_rate": 2.
|
970 |
-
"loss": 0.
|
971 |
"step": 1580
|
972 |
},
|
973 |
{
|
974 |
-
"epoch": 1.
|
975 |
-
"learning_rate": 2.
|
976 |
-
"loss": 0.
|
977 |
"step": 1590
|
978 |
},
|
979 |
{
|
980 |
-
"epoch": 1.
|
981 |
-
"learning_rate": 2.
|
982 |
-
"loss": 0.
|
983 |
"step": 1600
|
984 |
},
|
985 |
{
|
986 |
-
"epoch": 1.
|
987 |
-
"learning_rate":
|
988 |
-
"loss": 0.
|
989 |
"step": 1610
|
990 |
},
|
991 |
{
|
992 |
-
"epoch": 1.
|
993 |
-
"learning_rate":
|
994 |
-
"loss": 0.
|
995 |
"step": 1620
|
996 |
},
|
997 |
{
|
998 |
-
"epoch": 1.
|
999 |
-
"learning_rate":
|
1000 |
-
"loss": 0.
|
1001 |
"step": 1630
|
1002 |
},
|
1003 |
{
|
1004 |
-
"epoch": 1.
|
1005 |
-
"learning_rate":
|
1006 |
-
"loss": 0.
|
1007 |
"step": 1640
|
1008 |
},
|
1009 |
{
|
1010 |
-
"epoch": 1.
|
1011 |
-
"learning_rate":
|
1012 |
-
"loss": 0.
|
1013 |
"step": 1650
|
1014 |
},
|
1015 |
{
|
1016 |
-
"epoch": 1.
|
1017 |
-
"learning_rate":
|
1018 |
-
"loss": 0.
|
1019 |
"step": 1660
|
1020 |
},
|
1021 |
{
|
1022 |
-
"epoch": 1.
|
1023 |
-
"learning_rate":
|
1024 |
-
"loss": 0.
|
1025 |
"step": 1670
|
1026 |
},
|
1027 |
{
|
1028 |
-
"epoch": 1.
|
1029 |
-
"learning_rate":
|
1030 |
-
"loss": 0.
|
1031 |
"step": 1680
|
1032 |
},
|
1033 |
{
|
1034 |
-
"epoch": 1.
|
1035 |
-
"learning_rate":
|
1036 |
-
"loss": 0.
|
1037 |
"step": 1690
|
1038 |
},
|
1039 |
{
|
1040 |
-
"epoch": 1.
|
1041 |
-
"learning_rate":
|
1042 |
-
"loss": 0.
|
1043 |
"step": 1700
|
1044 |
},
|
1045 |
{
|
1046 |
-
"epoch": 1.
|
1047 |
-
"learning_rate":
|
1048 |
-
"loss": 0.
|
1049 |
"step": 1710
|
1050 |
},
|
1051 |
{
|
1052 |
-
"epoch": 1.
|
1053 |
-
"learning_rate":
|
1054 |
-
"loss": 0.
|
1055 |
"step": 1720
|
1056 |
},
|
1057 |
{
|
1058 |
-
"epoch": 1.
|
1059 |
-
"learning_rate":
|
1060 |
-
"loss": 0.
|
1061 |
"step": 1730
|
1062 |
},
|
1063 |
{
|
1064 |
-
"epoch": 1.
|
1065 |
-
"learning_rate":
|
1066 |
-
"loss": 0.
|
1067 |
"step": 1740
|
1068 |
},
|
1069 |
{
|
1070 |
-
"epoch": 1.
|
1071 |
-
"learning_rate":
|
1072 |
-
"loss": 0.
|
1073 |
"step": 1750
|
1074 |
},
|
1075 |
{
|
1076 |
-
"epoch": 1.
|
1077 |
-
"learning_rate":
|
1078 |
-
"loss": 0.
|
1079 |
"step": 1760
|
1080 |
},
|
1081 |
{
|
1082 |
-
"epoch": 1.
|
1083 |
-
"learning_rate":
|
1084 |
-
"loss": 0.
|
1085 |
"step": 1770
|
1086 |
},
|
1087 |
{
|
1088 |
-
"epoch": 1.
|
1089 |
-
"learning_rate":
|
1090 |
-
"loss": 0.
|
1091 |
"step": 1780
|
1092 |
},
|
1093 |
{
|
1094 |
-
"epoch": 1.
|
1095 |
-
"learning_rate": 2.
|
1096 |
-
"loss": 0.
|
1097 |
"step": 1790
|
1098 |
},
|
1099 |
{
|
1100 |
-
"epoch": 1.
|
1101 |
-
"learning_rate": 2.
|
1102 |
-
"loss": 0.
|
1103 |
"step": 1800
|
1104 |
},
|
1105 |
{
|
1106 |
-
"epoch": 1.
|
1107 |
-
"learning_rate":
|
1108 |
-
"loss": 0.
|
1109 |
"step": 1810
|
1110 |
},
|
1111 |
{
|
1112 |
-
"epoch": 1.
|
1113 |
-
"learning_rate":
|
1114 |
-
"loss": 0.
|
1115 |
"step": 1820
|
1116 |
},
|
1117 |
{
|
1118 |
-
"epoch": 1.
|
1119 |
-
"learning_rate": 1.
|
1120 |
-
"loss": 0.
|
1121 |
"step": 1830
|
1122 |
},
|
1123 |
{
|
1124 |
-
"epoch": 1.
|
1125 |
-
"learning_rate":
|
1126 |
-
"loss": 0.
|
1127 |
"step": 1840
|
1128 |
},
|
1129 |
{
|
1130 |
-
"epoch": 1.
|
1131 |
-
"learning_rate":
|
1132 |
-
"loss": 0.
|
1133 |
"step": 1850
|
1134 |
},
|
1135 |
{
|
1136 |
-
"epoch": 1.
|
1137 |
-
"learning_rate":
|
1138 |
-
"loss": 0.
|
1139 |
"step": 1860
|
1140 |
},
|
1141 |
{
|
1142 |
-
"epoch": 1.
|
1143 |
-
"learning_rate":
|
1144 |
-
"loss": 0.
|
1145 |
"step": 1870
|
1146 |
},
|
1147 |
{
|
1148 |
-
"epoch": 1.
|
1149 |
-
"learning_rate": 1.
|
1150 |
-
"loss": 0.
|
1151 |
"step": 1880
|
1152 |
},
|
1153 |
{
|
1154 |
-
"epoch": 1.
|
1155 |
-
"learning_rate":
|
1156 |
-
"loss": 0.
|
1157 |
"step": 1890
|
1158 |
},
|
1159 |
{
|
1160 |
-
"epoch":
|
1161 |
-
"learning_rate":
|
1162 |
-
"loss": 0.
|
1163 |
"step": 1900
|
1164 |
},
|
1165 |
{
|
1166 |
-
"epoch":
|
1167 |
-
"
|
1168 |
-
"
|
1169 |
-
"
|
1170 |
-
|
1171 |
-
|
1172 |
-
"
|
1173 |
-
"learning_rate": 1.900112269921026e-05,
|
1174 |
-
"loss": 0.2695,
|
1175 |
-
"step": 1920
|
1176 |
-
},
|
1177 |
-
{
|
1178 |
-
"epoch": 1.44,
|
1179 |
-
"step": 1922,
|
1180 |
-
"total_flos": 3.4599459174219776e+17,
|
1181 |
-
"train_loss": 0.3275021193364904,
|
1182 |
-
"train_runtime": 3966.2406,
|
1183 |
-
"train_samples_per_second": 4.365,
|
1184 |
-
"train_steps_per_second": 1.092
|
1185 |
}
|
1186 |
],
|
1187 |
"logging_steps": 10,
|
1188 |
-
"max_steps":
|
|
|
1189 |
"num_train_epochs": 2,
|
1190 |
"save_steps": 1000,
|
1191 |
-
"total_flos": 3.
|
|
|
1192 |
"trial_name": null,
|
1193 |
"trial_params": null
|
1194 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
"eval_steps": 1000,
|
6 |
+
"global_step": 1900,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.0,
|
13 |
+
"learning_rate": 3.1578947368421055e-07,
|
14 |
+
"loss": 0.7568,
|
15 |
"step": 1
|
16 |
},
|
17 |
{
|
18 |
+
"epoch": 0.01,
|
19 |
+
"learning_rate": 3.157894736842105e-06,
|
20 |
+
"loss": 0.7081,
|
21 |
"step": 10
|
22 |
},
|
23 |
{
|
24 |
+
"epoch": 0.02,
|
25 |
+
"learning_rate": 6.31578947368421e-06,
|
26 |
+
"loss": 0.6806,
|
27 |
"step": 20
|
28 |
},
|
29 |
{
|
30 |
+
"epoch": 0.03,
|
31 |
+
"learning_rate": 9.473684210526315e-06,
|
32 |
+
"loss": 0.6123,
|
33 |
"step": 30
|
34 |
},
|
35 |
{
|
36 |
+
"epoch": 0.04,
|
37 |
+
"learning_rate": 1.263157894736842e-05,
|
38 |
+
"loss": 0.4982,
|
39 |
"step": 40
|
40 |
},
|
41 |
{
|
42 |
+
"epoch": 0.05,
|
43 |
+
"learning_rate": 1.5789473684210526e-05,
|
44 |
+
"loss": 0.4448,
|
45 |
"step": 50
|
46 |
},
|
47 |
{
|
48 |
+
"epoch": 0.06,
|
49 |
+
"learning_rate": 1.894736842105263e-05,
|
50 |
+
"loss": 0.4243,
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
+
"epoch": 0.07,
|
55 |
+
"learning_rate": 2.2105263157894736e-05,
|
56 |
+
"loss": 0.3961,
|
57 |
"step": 70
|
58 |
},
|
59 |
{
|
60 |
+
"epoch": 0.08,
|
61 |
+
"learning_rate": 2.526315789473684e-05,
|
62 |
+
"loss": 0.3963,
|
63 |
"step": 80
|
64 |
},
|
65 |
{
|
66 |
+
"epoch": 0.09,
|
67 |
+
"learning_rate": 2.8421052631578946e-05,
|
68 |
+
"loss": 0.3795,
|
69 |
"step": 90
|
70 |
},
|
71 |
{
|
72 |
+
"epoch": 0.11,
|
73 |
+
"learning_rate": 2.9999432005848255e-05,
|
74 |
+
"loss": 0.3786,
|
75 |
"step": 100
|
76 |
},
|
77 |
{
|
78 |
+
"epoch": 0.12,
|
79 |
+
"learning_rate": 2.99948883107249e-05,
|
80 |
+
"loss": 0.3962,
|
81 |
"step": 110
|
82 |
},
|
83 |
{
|
84 |
+
"epoch": 0.13,
|
85 |
+
"learning_rate": 2.9985802296874666e-05,
|
86 |
+
"loss": 0.3566,
|
87 |
"step": 120
|
88 |
},
|
89 |
{
|
90 |
+
"epoch": 0.14,
|
91 |
+
"learning_rate": 2.9972176716673562e-05,
|
92 |
+
"loss": 0.3817,
|
93 |
"step": 130
|
94 |
},
|
95 |
{
|
96 |
+
"epoch": 0.15,
|
97 |
+
"learning_rate": 2.9954015697643372e-05,
|
98 |
+
"loss": 0.3644,
|
99 |
"step": 140
|
100 |
},
|
101 |
{
|
102 |
+
"epoch": 0.16,
|
103 |
+
"learning_rate": 2.9931324741201325e-05,
|
104 |
+
"loss": 0.3497,
|
105 |
"step": 150
|
106 |
},
|
107 |
{
|
108 |
+
"epoch": 0.17,
|
109 |
+
"learning_rate": 2.9904110720993565e-05,
|
110 |
+
"loss": 0.3601,
|
111 |
"step": 160
|
112 |
},
|
113 |
{
|
114 |
+
"epoch": 0.18,
|
115 |
+
"learning_rate": 2.987238188081299e-05,
|
116 |
+
"loss": 0.3877,
|
117 |
"step": 170
|
118 |
},
|
119 |
{
|
120 |
+
"epoch": 0.19,
|
121 |
+
"learning_rate": 2.983614783210197e-05,
|
122 |
+
"loss": 0.3585,
|
123 |
"step": 180
|
124 |
},
|
125 |
{
|
126 |
+
"epoch": 0.2,
|
127 |
+
"learning_rate": 2.9795419551040836e-05,
|
128 |
+
"loss": 0.3498,
|
129 |
"step": 190
|
130 |
},
|
131 |
{
|
132 |
+
"epoch": 0.21,
|
133 |
+
"learning_rate": 2.9750209375222893e-05,
|
134 |
+
"loss": 0.3618,
|
135 |
"step": 200
|
136 |
},
|
137 |
{
|
138 |
+
"epoch": 0.22,
|
139 |
+
"learning_rate": 2.97005309999171e-05,
|
140 |
+
"loss": 0.3496,
|
141 |
"step": 210
|
142 |
},
|
143 |
{
|
144 |
+
"epoch": 0.23,
|
145 |
+
"learning_rate": 2.964639947391939e-05,
|
146 |
+
"loss": 0.3483,
|
147 |
"step": 220
|
148 |
},
|
149 |
{
|
150 |
+
"epoch": 0.24,
|
151 |
+
"learning_rate": 2.958783119499408e-05,
|
152 |
+
"loss": 0.3488,
|
153 |
"step": 230
|
154 |
},
|
155 |
{
|
156 |
+
"epoch": 0.25,
|
157 |
+
"learning_rate": 2.9524843904906528e-05,
|
158 |
+
"loss": 0.3405,
|
159 |
"step": 240
|
160 |
},
|
161 |
{
|
162 |
+
"epoch": 0.26,
|
163 |
+
"learning_rate": 2.9457456684048772e-05,
|
164 |
+
"loss": 0.3441,
|
165 |
"step": 250
|
166 |
},
|
167 |
{
|
168 |
+
"epoch": 0.27,
|
169 |
+
"learning_rate": 2.938568994565956e-05,
|
170 |
+
"loss": 0.3638,
|
171 |
"step": 260
|
172 |
},
|
173 |
{
|
174 |
+
"epoch": 0.28,
|
175 |
+
"learning_rate": 2.9309565429640724e-05,
|
176 |
+
"loss": 0.3695,
|
177 |
"step": 270
|
178 |
},
|
179 |
{
|
180 |
+
"epoch": 0.29,
|
181 |
+
"learning_rate": 2.9229106195971603e-05,
|
182 |
+
"loss": 0.3372,
|
183 |
"step": 280
|
184 |
},
|
185 |
{
|
186 |
+
"epoch": 0.31,
|
187 |
+
"learning_rate": 2.9144336617723625e-05,
|
188 |
+
"loss": 0.354,
|
189 |
"step": 290
|
190 |
},
|
191 |
{
|
192 |
+
"epoch": 0.32,
|
193 |
+
"learning_rate": 2.90552823736771e-05,
|
194 |
+
"loss": 0.358,
|
195 |
"step": 300
|
196 |
},
|
197 |
{
|
198 |
+
"epoch": 0.33,
|
199 |
+
"learning_rate": 2.8961970440542496e-05,
|
200 |
+
"loss": 0.3473,
|
201 |
"step": 310
|
202 |
},
|
203 |
{
|
204 |
+
"epoch": 0.34,
|
205 |
+
"learning_rate": 2.8864429084788534e-05,
|
206 |
+
"loss": 0.3379,
|
207 |
"step": 320
|
208 |
},
|
209 |
{
|
210 |
+
"epoch": 0.35,
|
211 |
+
"learning_rate": 2.8762687854079563e-05,
|
212 |
+
"loss": 0.3794,
|
213 |
"step": 330
|
214 |
},
|
215 |
{
|
216 |
+
"epoch": 0.36,
|
217 |
+
"learning_rate": 2.8656777568324878e-05,
|
218 |
+
"loss": 0.3475,
|
219 |
"step": 340
|
220 |
},
|
221 |
{
|
222 |
+
"epoch": 0.37,
|
223 |
+
"learning_rate": 2.8546730310342593e-05,
|
224 |
+
"loss": 0.3655,
|
225 |
"step": 350
|
226 |
},
|
227 |
{
|
228 |
+
"epoch": 0.38,
|
229 |
+
"learning_rate": 2.8432579416140984e-05,
|
230 |
+
"loss": 0.3354,
|
231 |
"step": 360
|
232 |
},
|
233 |
{
|
234 |
+
"epoch": 0.39,
|
235 |
+
"learning_rate": 2.8314359464820184e-05,
|
236 |
+
"loss": 0.3448,
|
237 |
"step": 370
|
238 |
},
|
239 |
{
|
240 |
+
"epoch": 0.4,
|
241 |
+
"learning_rate": 2.8192106268097336e-05,
|
242 |
+
"loss": 0.3565,
|
243 |
"step": 380
|
244 |
},
|
245 |
{
|
246 |
+
"epoch": 0.41,
|
247 |
+
"learning_rate": 2.8065856859458346e-05,
|
248 |
+
"loss": 0.3436,
|
249 |
"step": 390
|
250 |
},
|
251 |
{
|
252 |
+
"epoch": 0.42,
|
253 |
+
"learning_rate": 2.7935649482939533e-05,
|
254 |
+
"loss": 0.3509,
|
255 |
"step": 400
|
256 |
},
|
257 |
{
|
258 |
+
"epoch": 0.43,
|
259 |
+
"learning_rate": 2.7801523581542563e-05,
|
260 |
+
"loss": 0.3312,
|
261 |
"step": 410
|
262 |
},
|
263 |
{
|
264 |
+
"epoch": 0.44,
|
265 |
+
"learning_rate": 2.766351978528622e-05,
|
266 |
+
"loss": 0.3445,
|
267 |
"step": 420
|
268 |
},
|
269 |
{
|
270 |
+
"epoch": 0.45,
|
271 |
+
"learning_rate": 2.7521679898898567e-05,
|
272 |
+
"loss": 0.3374,
|
273 |
"step": 430
|
274 |
},
|
275 |
{
|
276 |
+
"epoch": 0.46,
|
277 |
+
"learning_rate": 2.737604688915327e-05,
|
278 |
+
"loss": 0.3438,
|
279 |
"step": 440
|
280 |
},
|
281 |
{
|
282 |
+
"epoch": 0.47,
|
283 |
+
"learning_rate": 2.72266648718539e-05,
|
284 |
+
"loss": 0.3307,
|
285 |
"step": 450
|
286 |
},
|
287 |
{
|
288 |
+
"epoch": 0.48,
|
289 |
+
"learning_rate": 2.7073579098470196e-05,
|
290 |
+
"loss": 0.3344,
|
291 |
"step": 460
|
292 |
},
|
293 |
{
|
294 |
+
"epoch": 0.49,
|
295 |
+
"learning_rate": 2.6916835942430292e-05,
|
296 |
+
"loss": 0.325,
|
297 |
"step": 470
|
298 |
},
|
299 |
{
|
300 |
+
"epoch": 0.51,
|
301 |
+
"learning_rate": 2.6756482885073032e-05,
|
302 |
+
"loss": 0.3296,
|
303 |
"step": 480
|
304 |
},
|
305 |
{
|
306 |
+
"epoch": 0.52,
|
307 |
+
"learning_rate": 2.6592568501264746e-05,
|
308 |
+
"loss": 0.3536,
|
309 |
"step": 490
|
310 |
},
|
311 |
{
|
312 |
+
"epoch": 0.53,
|
313 |
+
"learning_rate": 2.6425142444684735e-05,
|
314 |
+
"loss": 0.3272,
|
315 |
"step": 500
|
316 |
},
|
317 |
{
|
318 |
+
"epoch": 0.54,
|
319 |
+
"learning_rate": 2.6254255432783933e-05,
|
320 |
+
"loss": 0.3547,
|
321 |
"step": 510
|
322 |
},
|
323 |
{
|
324 |
+
"epoch": 0.55,
|
325 |
+
"learning_rate": 2.6079959231421347e-05,
|
326 |
+
"loss": 0.3313,
|
327 |
"step": 520
|
328 |
},
|
329 |
{
|
330 |
+
"epoch": 0.56,
|
331 |
+
"learning_rate": 2.5902306639182952e-05,
|
332 |
+
"loss": 0.343,
|
333 |
"step": 530
|
334 |
},
|
335 |
{
|
336 |
+
"epoch": 0.57,
|
337 |
+
"learning_rate": 2.5721351471387666e-05,
|
338 |
+
"loss": 0.3129,
|
339 |
"step": 540
|
340 |
},
|
341 |
{
|
342 |
+
"epoch": 0.58,
|
343 |
+
"learning_rate": 2.5537148543785385e-05,
|
344 |
+
"loss": 0.3505,
|
345 |
"step": 550
|
346 |
},
|
347 |
{
|
348 |
+
"epoch": 0.59,
|
349 |
+
"learning_rate": 2.534975365595196e-05,
|
350 |
+
"loss": 0.3523,
|
351 |
"step": 560
|
352 |
},
|
353 |
{
|
354 |
+
"epoch": 0.6,
|
355 |
+
"learning_rate": 2.5159223574386117e-05,
|
356 |
+
"loss": 0.3699,
|
357 |
"step": 570
|
358 |
},
|
359 |
{
|
360 |
+
"epoch": 0.61,
|
361 |
+
"learning_rate": 2.496561601531353e-05,
|
362 |
+
"loss": 0.315,
|
363 |
"step": 580
|
364 |
},
|
365 |
{
|
366 |
+
"epoch": 0.62,
|
367 |
+
"learning_rate": 2.4768989627203123e-05,
|
368 |
+
"loss": 0.3382,
|
369 |
"step": 590
|
370 |
},
|
371 |
{
|
372 |
+
"epoch": 0.63,
|
373 |
+
"learning_rate": 2.4569403973001045e-05,
|
374 |
+
"loss": 0.3387,
|
375 |
"step": 600
|
376 |
},
|
377 |
{
|
378 |
+
"epoch": 0.64,
|
379 |
+
"learning_rate": 2.436691951208758e-05,
|
380 |
+
"loss": 0.3286,
|
381 |
"step": 610
|
382 |
},
|
383 |
{
|
384 |
+
"epoch": 0.65,
|
385 |
+
"learning_rate": 2.4161597581962526e-05,
|
386 |
+
"loss": 0.3096,
|
387 |
"step": 620
|
388 |
},
|
389 |
{
|
390 |
+
"epoch": 0.66,
|
391 |
+
"learning_rate": 2.395350037966456e-05,
|
392 |
+
"loss": 0.3384,
|
393 |
"step": 630
|
394 |
},
|
395 |
{
|
396 |
+
"epoch": 0.67,
|
397 |
+
"learning_rate": 2.3742690942930235e-05,
|
398 |
+
"loss": 0.3284,
|
399 |
"step": 640
|
400 |
},
|
401 |
{
|
402 |
+
"epoch": 0.68,
|
403 |
+
"learning_rate": 2.3529233131098313e-05,
|
404 |
+
"loss": 0.3295,
|
405 |
"step": 650
|
406 |
},
|
407 |
{
|
408 |
+
"epoch": 0.69,
|
409 |
+
"learning_rate": 2.33131916057652e-05,
|
410 |
+
"loss": 0.3164,
|
411 |
"step": 660
|
412 |
},
|
413 |
{
|
414 |
+
"epoch": 0.71,
|
415 |
+
"learning_rate": 2.309463181119736e-05,
|
416 |
+
"loss": 0.3597,
|
417 |
"step": 670
|
418 |
},
|
419 |
{
|
420 |
+
"epoch": 0.72,
|
421 |
+
"learning_rate": 2.287361995450667e-05,
|
422 |
+
"loss": 0.3239,
|
423 |
"step": 680
|
424 |
},
|
425 |
{
|
426 |
+
"epoch": 0.73,
|
427 |
+
"learning_rate": 2.2650222985594634e-05,
|
428 |
+
"loss": 0.3298,
|
429 |
"step": 690
|
430 |
},
|
431 |
{
|
432 |
+
"epoch": 0.74,
|
433 |
+
"learning_rate": 2.2424508576871623e-05,
|
434 |
+
"loss": 0.3176,
|
435 |
"step": 700
|
436 |
},
|
437 |
{
|
438 |
+
"epoch": 0.75,
|
439 |
+
"learning_rate": 2.219654510275728e-05,
|
440 |
+
"loss": 0.3344,
|
441 |
"step": 710
|
442 |
},
|
443 |
{
|
444 |
+
"epoch": 0.76,
|
445 |
+
"learning_rate": 2.1966401618968194e-05,
|
446 |
+
"loss": 0.3518,
|
447 |
"step": 720
|
448 |
},
|
449 |
{
|
450 |
+
"epoch": 0.77,
|
451 |
+
"learning_rate": 2.173414784159925e-05,
|
452 |
+
"loss": 0.3323,
|
453 |
"step": 730
|
454 |
},
|
455 |
{
|
456 |
+
"epoch": 0.78,
|
457 |
+
"learning_rate": 2.149985412600492e-05,
|
458 |
+
"loss": 0.3343,
|
459 |
"step": 740
|
460 |
},
|
461 |
{
|
462 |
+
"epoch": 0.79,
|
463 |
+
"learning_rate": 2.1263591445486895e-05,
|
464 |
+
"loss": 0.3383,
|
465 |
"step": 750
|
466 |
},
|
467 |
{
|
468 |
+
"epoch": 0.8,
|
469 |
+
"learning_rate": 2.1025431369794546e-05,
|
470 |
+
"loss": 0.3135,
|
471 |
"step": 760
|
472 |
},
|
473 |
{
|
474 |
+
"epoch": 0.81,
|
475 |
+
"learning_rate": 2.0785446043444677e-05,
|
476 |
+
"loss": 0.3278,
|
477 |
"step": 770
|
478 |
},
|
479 |
{
|
480 |
+
"epoch": 0.82,
|
481 |
+
"learning_rate": 2.0543708163867204e-05,
|
482 |
+
"loss": 0.3148,
|
483 |
"step": 780
|
484 |
},
|
485 |
{
|
486 |
+
"epoch": 0.83,
|
487 |
+
"learning_rate": 2.0300290959383318e-05,
|
488 |
+
"loss": 0.3511,
|
489 |
"step": 790
|
490 |
},
|
491 |
{
|
492 |
+
"epoch": 0.84,
|
493 |
+
"learning_rate": 2.0055268167022835e-05,
|
494 |
+
"loss": 0.3146,
|
495 |
"step": 800
|
496 |
},
|
497 |
{
|
498 |
+
"epoch": 0.85,
|
499 |
+
"learning_rate": 1.9808714010187425e-05,
|
500 |
+
"loss": 0.3321,
|
501 |
"step": 810
|
502 |
},
|
503 |
{
|
504 |
+
"epoch": 0.86,
|
505 |
+
"learning_rate": 1.9560703176166565e-05,
|
506 |
+
"loss": 0.3373,
|
507 |
"step": 820
|
508 |
},
|
509 |
{
|
510 |
+
"epoch": 0.87,
|
511 |
+
"learning_rate": 1.931131079351289e-05,
|
512 |
+
"loss": 0.32,
|
513 |
"step": 830
|
514 |
},
|
515 |
{
|
516 |
+
"epoch": 0.88,
|
517 |
+
"learning_rate": 1.9060612409283946e-05,
|
518 |
+
"loss": 0.3106,
|
519 |
"step": 840
|
520 |
},
|
521 |
{
|
522 |
+
"epoch": 0.89,
|
523 |
+
"learning_rate": 1.8808683966157132e-05,
|
524 |
+
"loss": 0.3167,
|
525 |
"step": 850
|
526 |
},
|
527 |
{
|
528 |
+
"epoch": 0.91,
|
529 |
+
"learning_rate": 1.8555601779424778e-05,
|
530 |
+
"loss": 0.2993,
|
531 |
"step": 860
|
532 |
},
|
533 |
{
|
534 |
+
"epoch": 0.92,
|
535 |
+
"learning_rate": 1.8301442513876406e-05,
|
536 |
+
"loss": 0.3405,
|
537 |
"step": 870
|
538 |
},
|
539 |
{
|
540 |
+
"epoch": 0.93,
|
541 |
+
"learning_rate": 1.804628316057508e-05,
|
542 |
+
"loss": 0.3346,
|
543 |
"step": 880
|
544 |
},
|
545 |
{
|
546 |
+
"epoch": 0.94,
|
547 |
+
"learning_rate": 1.779020101353492e-05,
|
548 |
+
"loss": 0.3319,
|
549 |
"step": 890
|
550 |
},
|
551 |
{
|
552 |
+
"epoch": 0.95,
|
553 |
+
"learning_rate": 1.7533273646306857e-05,
|
554 |
+
"loss": 0.3087,
|
555 |
"step": 900
|
556 |
},
|
557 |
{
|
558 |
+
"epoch": 0.96,
|
559 |
+
"learning_rate": 1.7275578888479714e-05,
|
560 |
+
"loss": 0.316,
|
561 |
"step": 910
|
562 |
},
|
563 |
{
|
564 |
+
"epoch": 0.97,
|
565 |
+
"learning_rate": 1.7017194802103705e-05,
|
566 |
+
"loss": 0.3054,
|
567 |
"step": 920
|
568 |
},
|
569 |
{
|
570 |
+
"epoch": 0.98,
|
571 |
+
"learning_rate": 1.6758199658043538e-05,
|
572 |
+
"loss": 0.3255,
|
573 |
"step": 930
|
574 |
},
|
575 |
{
|
576 |
+
"epoch": 0.99,
|
577 |
+
"learning_rate": 1.6498671912268256e-05,
|
578 |
+
"loss": 0.3175,
|
579 |
"step": 940
|
580 |
},
|
581 |
{
|
582 |
+
"epoch": 1.0,
|
583 |
+
"learning_rate": 1.623869018208499e-05,
|
584 |
+
"loss": 0.3424,
|
585 |
"step": 950
|
586 |
},
|
587 |
{
|
588 |
+
"epoch": 1.01,
|
589 |
+
"learning_rate": 1.5978333222323858e-05,
|
590 |
+
"loss": 0.2825,
|
591 |
"step": 960
|
592 |
},
|
593 |
{
|
594 |
+
"epoch": 1.02,
|
595 |
+
"learning_rate": 1.571767990148122e-05,
|
596 |
+
"loss": 0.2886,
|
597 |
"step": 970
|
598 |
},
|
599 |
{
|
600 |
+
"epoch": 1.03,
|
601 |
+
"learning_rate": 1.5456809177828444e-05,
|
602 |
+
"loss": 0.3196,
|
603 |
"step": 980
|
604 |
},
|
605 |
{
|
606 |
+
"epoch": 1.04,
|
607 |
+
"learning_rate": 1.5195800075493542e-05,
|
608 |
+
"loss": 0.3178,
|
609 |
"step": 990
|
610 |
},
|
611 |
{
|
612 |
+
"epoch": 1.05,
|
613 |
+
"learning_rate": 1.4934731660522817e-05,
|
614 |
+
"loss": 0.2676,
|
615 |
"step": 1000
|
616 |
},
|
617 |
{
|
618 |
+
"epoch": 1.05,
|
619 |
+
"eval_loss": 0.508576512336731,
|
620 |
+
"eval_runtime": 12.2399,
|
621 |
+
"eval_samples_per_second": 3.595,
|
622 |
+
"eval_steps_per_second": 0.899,
|
623 |
"step": 1000
|
624 |
},
|
625 |
{
|
626 |
+
"epoch": 1.06,
|
627 |
+
"learning_rate": 1.4673683016929805e-05,
|
628 |
+
"loss": 0.2803,
|
629 |
"step": 1010
|
630 |
},
|
631 |
{
|
632 |
+
"epoch": 1.07,
|
633 |
+
"learning_rate": 1.441273322273884e-05,
|
634 |
+
"loss": 0.319,
|
635 |
"step": 1020
|
636 |
},
|
637 |
{
|
638 |
+
"epoch": 1.08,
|
639 |
+
"learning_rate": 1.4151961326030314e-05,
|
640 |
+
"loss": 0.2965,
|
641 |
"step": 1030
|
642 |
},
|
643 |
{
|
644 |
+
"epoch": 1.09,
|
645 |
+
"learning_rate": 1.3891446320995143e-05,
|
646 |
+
"loss": 0.3067,
|
647 |
"step": 1040
|
648 |
},
|
649 |
{
|
650 |
+
"epoch": 1.11,
|
651 |
+
"learning_rate": 1.3631267124005453e-05,
|
652 |
+
"loss": 0.3072,
|
653 |
"step": 1050
|
654 |
},
|
655 |
{
|
656 |
+
"epoch": 1.12,
|
657 |
+
"learning_rate": 1.337150254970891e-05,
|
658 |
+
"loss": 0.2963,
|
659 |
"step": 1060
|
660 |
},
|
661 |
{
|
662 |
+
"epoch": 1.13,
|
663 |
+
"learning_rate": 1.3112231287153798e-05,
|
664 |
+
"loss": 0.3063,
|
665 |
"step": 1070
|
666 |
},
|
667 |
{
|
668 |
+
"epoch": 1.14,
|
669 |
+
"learning_rate": 1.28535318759522e-05,
|
670 |
+
"loss": 0.2732,
|
671 |
"step": 1080
|
672 |
},
|
673 |
{
|
674 |
+
"epoch": 1.15,
|
675 |
+
"learning_rate": 1.2595482682488443e-05,
|
676 |
+
"loss": 0.2874,
|
677 |
"step": 1090
|
678 |
},
|
679 |
{
|
680 |
+
"epoch": 1.16,
|
681 |
+
"learning_rate": 1.2338161876179964e-05,
|
682 |
+
"loss": 0.2998,
|
683 |
"step": 1100
|
684 |
},
|
685 |
{
|
686 |
+
"epoch": 1.17,
|
687 |
+
"learning_rate": 1.2081647405797923e-05,
|
688 |
+
"loss": 0.2849,
|
689 |
"step": 1110
|
690 |
},
|
691 |
{
|
692 |
+
"epoch": 1.18,
|
693 |
+
"learning_rate": 1.1826016975854563e-05,
|
694 |
+
"loss": 0.2735,
|
695 |
"step": 1120
|
696 |
},
|
697 |
{
|
698 |
+
"epoch": 1.19,
|
699 |
+
"learning_rate": 1.1571348023064662e-05,
|
700 |
+
"loss": 0.2937,
|
701 |
"step": 1130
|
702 |
},
|
703 |
{
|
704 |
+
"epoch": 1.2,
|
705 |
+
"learning_rate": 1.1317717692888014e-05,
|
706 |
+
"loss": 0.2993,
|
707 |
"step": 1140
|
708 |
},
|
709 |
{
|
710 |
+
"epoch": 1.21,
|
711 |
+
"learning_rate": 1.1065202816160213e-05,
|
712 |
+
"loss": 0.2933,
|
713 |
"step": 1150
|
714 |
},
|
715 |
{
|
716 |
+
"epoch": 1.22,
|
717 |
+
"learning_rate": 1.081387988581869e-05,
|
718 |
+
"loss": 0.285,
|
719 |
"step": 1160
|
720 |
},
|
721 |
{
|
722 |
+
"epoch": 1.23,
|
723 |
+
"learning_rate": 1.0563825033731146e-05,
|
724 |
+
"loss": 0.2896,
|
725 |
"step": 1170
|
726 |
},
|
727 |
{
|
728 |
+
"epoch": 1.24,
|
729 |
+
"learning_rate": 1.031511400763332e-05,
|
730 |
+
"loss": 0.2874,
|
731 |
"step": 1180
|
732 |
},
|
733 |
{
|
734 |
+
"epoch": 1.25,
|
735 |
+
"learning_rate": 1.0067822148183194e-05,
|
736 |
+
"loss": 0.2851,
|
737 |
"step": 1190
|
738 |
},
|
739 |
{
|
740 |
+
"epoch": 1.26,
|
741 |
+
"learning_rate": 9.822024366138397e-06,
|
742 |
+
"loss": 0.2925,
|
743 |
"step": 1200
|
744 |
},
|
745 |
{
|
746 |
+
"epoch": 1.27,
|
747 |
+
"learning_rate": 9.577795119663966e-06,
|
748 |
+
"loss": 0.2843,
|
749 |
"step": 1210
|
750 |
},
|
751 |
{
|
752 |
+
"epoch": 1.28,
|
753 |
+
"learning_rate": 9.335208391777106e-06,
|
754 |
+
"loss": 0.2884,
|
755 |
"step": 1220
|
756 |
},
|
757 |
{
|
758 |
+
"epoch": 1.29,
|
759 |
+
"learning_rate": 9.094337667935942e-06,
|
760 |
+
"loss": 0.269,
|
761 |
"step": 1230
|
762 |
},
|
763 |
{
|
764 |
+
"epoch": 1.31,
|
765 |
+
"learning_rate": 8.855255913778949e-06,
|
766 |
+
"loss": 0.2849,
|
767 |
"step": 1240
|
768 |
},
|
769 |
{
|
770 |
+
"epoch": 1.32,
|
771 |
+
"learning_rate": 8.618035553021925e-06,
|
772 |
+
"loss": 0.304,
|
773 |
"step": 1250
|
774 |
},
|
775 |
{
|
776 |
+
"epoch": 1.33,
|
777 |
+
"learning_rate": 8.382748445519008e-06,
|
778 |
+
"loss": 0.3049,
|
779 |
"step": 1260
|
780 |
},
|
781 |
{
|
782 |
+
"epoch": 1.34,
|
783 |
+
"learning_rate": 8.149465865494633e-06,
|
784 |
+
"loss": 0.2999,
|
785 |
"step": 1270
|
786 |
},
|
787 |
{
|
788 |
+
"epoch": 1.35,
|
789 |
+
"learning_rate": 7.918258479952763e-06,
|
790 |
+
"loss": 0.2835,
|
791 |
"step": 1280
|
792 |
},
|
793 |
{
|
794 |
+
"epoch": 1.36,
|
795 |
+
"learning_rate": 7.689196327270171e-06,
|
796 |
+
"loss": 0.3167,
|
797 |
"step": 1290
|
798 |
},
|
799 |
{
|
800 |
+
"epoch": 1.37,
|
801 |
+
"learning_rate": 7.462348795980088e-06,
|
802 |
+
"loss": 0.2842,
|
803 |
"step": 1300
|
804 |
},
|
805 |
{
|
806 |
+
"epoch": 1.38,
|
807 |
+
"learning_rate": 7.237784603752705e-06,
|
808 |
+
"loss": 0.2909,
|
809 |
"step": 1310
|
810 |
},
|
811 |
{
|
812 |
+
"epoch": 1.39,
|
813 |
+
"learning_rate": 7.015571776578922e-06,
|
814 |
+
"loss": 0.2881,
|
815 |
"step": 1320
|
816 |
},
|
817 |
{
|
818 |
+
"epoch": 1.4,
|
819 |
+
"learning_rate": 6.795777628163599e-06,
|
820 |
+
"loss": 0.2796,
|
821 |
"step": 1330
|
822 |
},
|
823 |
{
|
824 |
+
"epoch": 1.41,
|
825 |
+
"learning_rate": 6.578468739534602e-06,
|
826 |
+
"loss": 0.3056,
|
827 |
"step": 1340
|
828 |
},
|
829 |
{
|
830 |
+
"epoch": 1.42,
|
831 |
+
"learning_rate": 6.363710938873759e-06,
|
832 |
+
"loss": 0.2987,
|
833 |
"step": 1350
|
834 |
},
|
835 |
{
|
836 |
+
"epoch": 1.43,
|
837 |
+
"learning_rate": 6.151569281575925e-06,
|
838 |
+
"loss": 0.288,
|
839 |
"step": 1360
|
840 |
},
|
841 |
{
|
842 |
+
"epoch": 1.44,
|
843 |
+
"learning_rate": 5.942108030542074e-06,
|
844 |
+
"loss": 0.2954,
|
845 |
"step": 1370
|
846 |
},
|
847 |
{
|
848 |
+
"epoch": 1.45,
|
849 |
+
"learning_rate": 5.735390636712514e-06,
|
850 |
+
"loss": 0.2837,
|
851 |
"step": 1380
|
852 |
},
|
853 |
{
|
854 |
+
"epoch": 1.46,
|
855 |
+
"learning_rate": 5.531479719846038e-06,
|
856 |
+
"loss": 0.3055,
|
857 |
"step": 1390
|
858 |
},
|
859 |
{
|
860 |
+
"epoch": 1.47,
|
861 |
+
"learning_rate": 5.330437049550868e-06,
|
862 |
+
"loss": 0.313,
|
863 |
"step": 1400
|
864 |
},
|
865 |
{
|
866 |
+
"epoch": 1.48,
|
867 |
+
"learning_rate": 5.132323526573126e-06,
|
868 |
+
"loss": 0.2966,
|
869 |
"step": 1410
|
870 |
},
|
871 |
{
|
872 |
+
"epoch": 1.49,
|
873 |
+
"learning_rate": 4.937199164348521e-06,
|
874 |
+
"loss": 0.2741,
|
875 |
"step": 1420
|
876 |
},
|
877 |
{
|
878 |
+
"epoch": 1.51,
|
879 |
+
"learning_rate": 4.745123070822786e-06,
|
880 |
+
"loss": 0.2973,
|
881 |
"step": 1430
|
882 |
},
|
883 |
{
|
884 |
+
"epoch": 1.52,
|
885 |
+
"learning_rate": 4.556153430546451e-06,
|
886 |
+
"loss": 0.281,
|
887 |
"step": 1440
|
888 |
},
|
889 |
{
|
890 |
+
"epoch": 1.53,
|
891 |
+
"learning_rate": 4.370347487049313e-06,
|
892 |
+
"loss": 0.2905,
|
893 |
"step": 1450
|
894 |
},
|
895 |
{
|
896 |
+
"epoch": 1.54,
|
897 |
+
"learning_rate": 4.187761525499973e-06,
|
898 |
+
"loss": 0.2806,
|
899 |
"step": 1460
|
900 |
},
|
901 |
{
|
902 |
+
"epoch": 1.55,
|
903 |
+
"learning_rate": 4.008450855655675e-06,
|
904 |
+
"loss": 0.2716,
|
905 |
"step": 1470
|
906 |
},
|
907 |
{
|
908 |
+
"epoch": 1.56,
|
909 |
+
"learning_rate": 3.83246979510764e-06,
|
910 |
+
"loss": 0.2834,
|
911 |
"step": 1480
|
912 |
},
|
913 |
{
|
914 |
+
"epoch": 1.57,
|
915 |
+
"learning_rate": 3.676977737529078e-06,
|
916 |
+
"loss": 0.2809,
|
917 |
"step": 1490
|
918 |
},
|
919 |
{
|
920 |
+
"epoch": 1.58,
|
921 |
+
"learning_rate": 3.5074689542164895e-06,
|
922 |
+
"loss": 0.2845,
|
923 |
"step": 1500
|
924 |
},
|
925 |
{
|
926 |
+
"epoch": 1.59,
|
927 |
+
"learning_rate": 3.341441539881574e-06,
|
928 |
+
"loss": 0.3001,
|
929 |
"step": 1510
|
930 |
},
|
931 |
{
|
932 |
+
"epoch": 1.6,
|
933 |
+
"learning_rate": 3.1789457882922753e-06,
|
934 |
+
"loss": 0.2941,
|
935 |
"step": 1520
|
936 |
},
|
937 |
{
|
938 |
+
"epoch": 1.61,
|
939 |
+
"learning_rate": 3.020030923389471e-06,
|
940 |
+
"loss": 0.2917,
|
941 |
"step": 1530
|
942 |
},
|
943 |
{
|
944 |
+
"epoch": 1.62,
|
945 |
+
"learning_rate": 2.86474508437579e-06,
|
946 |
+
"loss": 0.293,
|
947 |
"step": 1540
|
948 |
},
|
949 |
{
|
950 |
+
"epoch": 1.63,
|
951 |
+
"learning_rate": 2.7131353111330843e-06,
|
952 |
+
"loss": 0.2941,
|
953 |
"step": 1550
|
954 |
},
|
955 |
{
|
956 |
+
"epoch": 1.64,
|
957 |
+
"learning_rate": 2.565247529972901e-06,
|
958 |
+
"loss": 0.2787,
|
959 |
"step": 1560
|
960 |
},
|
961 |
{
|
962 |
+
"epoch": 1.65,
|
963 |
+
"learning_rate": 2.4211265397242854e-06,
|
964 |
+
"loss": 0.2899,
|
965 |
"step": 1570
|
966 |
},
|
967 |
{
|
968 |
+
"epoch": 1.66,
|
969 |
+
"learning_rate": 2.280815998163083e-06,
|
970 |
+
"loss": 0.2814,
|
971 |
"step": 1580
|
972 |
},
|
973 |
{
|
974 |
+
"epoch": 1.67,
|
975 |
+
"learning_rate": 2.144358408786986e-06,
|
976 |
+
"loss": 0.2786,
|
977 |
"step": 1590
|
978 |
},
|
979 |
{
|
980 |
+
"epoch": 1.68,
|
981 |
+
"learning_rate": 2.011795107940138e-06,
|
982 |
+
"loss": 0.2916,
|
983 |
"step": 1600
|
984 |
},
|
985 |
{
|
986 |
+
"epoch": 1.69,
|
987 |
+
"learning_rate": 1.8831662522913594e-06,
|
988 |
+
"loss": 0.2834,
|
989 |
"step": 1610
|
990 |
},
|
991 |
{
|
992 |
+
"epoch": 1.71,
|
993 |
+
"learning_rate": 1.7585108066697136e-06,
|
994 |
+
"loss": 0.2735,
|
995 |
"step": 1620
|
996 |
},
|
997 |
{
|
998 |
+
"epoch": 1.72,
|
999 |
+
"learning_rate": 1.6378665322611002e-06,
|
1000 |
+
"loss": 0.3039,
|
1001 |
"step": 1630
|
1002 |
},
|
1003 |
{
|
1004 |
+
"epoch": 1.73,
|
1005 |
+
"learning_rate": 1.521269975169471e-06,
|
1006 |
+
"loss": 0.2872,
|
1007 |
"step": 1640
|
1008 |
},
|
1009 |
{
|
1010 |
+
"epoch": 1.74,
|
1011 |
+
"learning_rate": 1.408756455346114e-06,
|
1012 |
+
"loss": 0.2863,
|
1013 |
"step": 1650
|
1014 |
},
|
1015 |
{
|
1016 |
+
"epoch": 1.75,
|
1017 |
+
"learning_rate": 1.3003600558903927e-06,
|
1018 |
+
"loss": 0.2854,
|
1019 |
"step": 1660
|
1020 |
},
|
1021 |
{
|
1022 |
+
"epoch": 1.76,
|
1023 |
+
"learning_rate": 1.196113612725116e-06,
|
1024 |
+
"loss": 0.2727,
|
1025 |
"step": 1670
|
1026 |
},
|
1027 |
{
|
1028 |
+
"epoch": 1.77,
|
1029 |
+
"learning_rate": 1.0960487046497524e-06,
|
1030 |
+
"loss": 0.2763,
|
1031 |
"step": 1680
|
1032 |
},
|
1033 |
{
|
1034 |
+
"epoch": 1.78,
|
1035 |
+
"learning_rate": 1.000195643774431e-06,
|
1036 |
+
"loss": 0.2962,
|
1037 |
"step": 1690
|
1038 |
},
|
1039 |
{
|
1040 |
+
"epoch": 1.79,
|
1041 |
+
"learning_rate": 9.085834663376629e-07,
|
1042 |
+
"loss": 0.2888,
|
1043 |
"step": 1700
|
1044 |
},
|
1045 |
{
|
1046 |
+
"epoch": 1.8,
|
1047 |
+
"learning_rate": 8.212399239105534e-07,
|
1048 |
+
"loss": 0.2839,
|
1049 |
"step": 1710
|
1050 |
},
|
1051 |
{
|
1052 |
+
"epoch": 1.81,
|
1053 |
+
"learning_rate": 7.381914749901752e-07,
|
1054 |
+
"loss": 0.2789,
|
1055 |
"step": 1720
|
1056 |
},
|
1057 |
{
|
1058 |
+
"epoch": 1.82,
|
1059 |
+
"learning_rate": 6.594632769846353e-07,
|
1060 |
+
"loss": 0.2772,
|
1061 |
"step": 1730
|
1062 |
},
|
1063 |
{
|
1064 |
+
"epoch": 1.83,
|
1065 |
+
"learning_rate": 5.850791785922849e-07,
|
1066 |
+
"loss": 0.278,
|
1067 |
"step": 1740
|
1068 |
},
|
1069 |
{
|
1070 |
+
"epoch": 1.84,
|
1071 |
+
"learning_rate": 5.150617125773633e-07,
|
1072 |
+
"loss": 0.2878,
|
1073 |
"step": 1750
|
1074 |
},
|
1075 |
{
|
1076 |
+
"epoch": 1.85,
|
1077 |
+
"learning_rate": 4.494320889442749e-07,
|
1078 |
+
"loss": 0.2734,
|
1079 |
"step": 1760
|
1080 |
},
|
1081 |
{
|
1082 |
+
"epoch": 1.86,
|
1083 |
+
"learning_rate": 3.882101885125539e-07,
|
1084 |
+
"loss": 0.2826,
|
1085 |
"step": 1770
|
1086 |
},
|
1087 |
{
|
1088 |
+
"epoch": 1.87,
|
1089 |
+
"learning_rate": 3.3141455689448266e-07,
|
1090 |
+
"loss": 0.2875,
|
1091 |
"step": 1780
|
1092 |
},
|
1093 |
{
|
1094 |
+
"epoch": 1.88,
|
1095 |
+
"learning_rate": 2.790623988771712e-07,
|
1096 |
+
"loss": 0.2898,
|
1097 |
"step": 1790
|
1098 |
},
|
1099 |
{
|
1100 |
+
"epoch": 1.89,
|
1101 |
+
"learning_rate": 2.3116957321080102e-07,
|
1102 |
+
"loss": 0.2919,
|
1103 |
"step": 1800
|
1104 |
},
|
1105 |
{
|
1106 |
+
"epoch": 1.91,
|
1107 |
+
"learning_rate": 1.8775058780463094e-07,
|
1108 |
+
"loss": 0.2778,
|
1109 |
"step": 1810
|
1110 |
},
|
1111 |
{
|
1112 |
+
"epoch": 1.92,
|
1113 |
+
"learning_rate": 1.4881859533218466e-07,
|
1114 |
+
"loss": 0.3179,
|
1115 |
"step": 1820
|
1116 |
},
|
1117 |
{
|
1118 |
+
"epoch": 1.93,
|
1119 |
+
"learning_rate": 1.1438538924699094e-07,
|
1120 |
+
"loss": 0.2771,
|
1121 |
"step": 1830
|
1122 |
},
|
1123 |
{
|
1124 |
+
"epoch": 1.94,
|
1125 |
+
"learning_rate": 8.446140021006132e-08,
|
1126 |
+
"loss": 0.2817,
|
1127 |
"step": 1840
|
1128 |
},
|
1129 |
{
|
1130 |
+
"epoch": 1.95,
|
1131 |
+
"learning_rate": 5.9055692930179426e-08,
|
1132 |
+
"loss": 0.2903,
|
1133 |
"step": 1850
|
1134 |
},
|
1135 |
{
|
1136 |
+
"epoch": 1.96,
|
1137 |
+
"learning_rate": 3.8175963417980685e-08,
|
1138 |
+
"loss": 0.2934,
|
1139 |
"step": 1860
|
1140 |
},
|
1141 |
{
|
1142 |
+
"epoch": 1.97,
|
1143 |
+
"learning_rate": 2.1828536654647235e-08,
|
1144 |
+
"loss": 0.2952,
|
1145 |
"step": 1870
|
1146 |
},
|
1147 |
{
|
1148 |
+
"epoch": 1.98,
|
1149 |
+
"learning_rate": 1.0018364675912217e-08,
|
1150 |
+
"loss": 0.2881,
|
1151 |
"step": 1880
|
1152 |
},
|
1153 |
{
|
1154 |
+
"epoch": 1.99,
|
1155 |
+
"learning_rate": 2.7490250719663933e-09,
|
1156 |
+
"loss": 0.2701,
|
1157 |
"step": 1890
|
1158 |
},
|
1159 |
{
|
1160 |
+
"epoch": 2.0,
|
1161 |
+
"learning_rate": 2.2719903721712954e-11,
|
1162 |
+
"loss": 0.2955,
|
1163 |
"step": 1900
|
1164 |
},
|
1165 |
{
|
1166 |
+
"epoch": 2.0,
|
1167 |
+
"step": 1900,
|
1168 |
+
"total_flos": 3.4203213408659046e+17,
|
1169 |
+
"train_loss": 0.3227183536165639,
|
1170 |
+
"train_runtime": 6710.5312,
|
1171 |
+
"train_samples_per_second": 1.132,
|
1172 |
+
"train_steps_per_second": 0.283
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1173 |
}
|
1174 |
],
|
1175 |
"logging_steps": 10,
|
1176 |
+
"max_steps": 1900,
|
1177 |
+
"num_input_tokens_seen": 0,
|
1178 |
"num_train_epochs": 2,
|
1179 |
"save_steps": 1000,
|
1180 |
+
"total_flos": 3.4203213408659046e+17,
|
1181 |
+
"train_batch_size": 4,
|
1182 |
"trial_name": null,
|
1183 |
"trial_params": null
|
1184 |
}
|