Training in progress, step 21000, checkpoint
Browse files
checkpoint-21000/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1711848436
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8d88e4391b49b7f47122e6579b3d3acfbc9c067ea039fb9eabac0ab6772aff48
|
3 |
size 1711848436
|
checkpoint-21000/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3424043887
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8b8ba3e1769c9a5eb1594683a4c35c2c714234c4df59ce37892e4d1d138efb08
|
3 |
size 3424043887
|
checkpoint-21000/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:551bd9fefe21a0210b8f009f7e379124ea37132a2e134634bfee8fc680d70fd3
|
3 |
size 14503
|
checkpoint-21000/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 sha256:259cafb41160652c3d54806748b10cd7a9019d0dd9321fe4c5ddeee2dc391a9c
|
3 |
size 623
|
checkpoint-21000/trainer_state.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 21000,
|
7 |
"is_hyper_param_search": false,
|
@@ -9,638 +9,638 @@
|
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
-
"epoch": 0.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 4.
|
15 |
-
"loss": 0.
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
-
"epoch": 0.
|
20 |
-
"eval_loss": 1.
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second":
|
23 |
-
"eval_steps_per_second":
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
-
"epoch": 0.
|
28 |
-
"grad_norm":
|
29 |
-
"learning_rate": 4.
|
30 |
-
"loss": 0.
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
-
"epoch": 0.
|
35 |
-
"eval_loss": 0.
|
36 |
-
"eval_runtime":
|
37 |
-
"eval_samples_per_second":
|
38 |
-
"eval_steps_per_second":
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
-
"epoch": 0.
|
43 |
-
"grad_norm":
|
44 |
-
"learning_rate": 4.
|
45 |
-
"loss": 0.
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
-
"epoch": 0.
|
50 |
-
"eval_loss": 0.
|
51 |
-
"eval_runtime":
|
52 |
-
"eval_samples_per_second": 15.
|
53 |
-
"eval_steps_per_second": 1.
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
-
"epoch": 0.
|
58 |
-
"grad_norm":
|
59 |
-
"learning_rate": 4.
|
60 |
-
"loss": 0.
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
-
"epoch": 0.
|
65 |
-
"eval_loss": 0.
|
66 |
-
"eval_runtime":
|
67 |
-
"eval_samples_per_second": 15.
|
68 |
-
"eval_steps_per_second": 1.
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
-
"epoch": 0.
|
73 |
-
"grad_norm":
|
74 |
-
"learning_rate": 4.
|
75 |
-
"loss": 0.
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
-
"epoch": 0.
|
80 |
-
"eval_loss": 0.
|
81 |
-
"eval_runtime":
|
82 |
-
"eval_samples_per_second": 15.
|
83 |
-
"eval_steps_per_second": 1.
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
-
"epoch": 0.
|
88 |
-
"grad_norm":
|
89 |
-
"learning_rate": 4.
|
90 |
-
"loss": 0.
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
-
"epoch": 0.
|
95 |
-
"eval_loss": 0.
|
96 |
-
"eval_runtime":
|
97 |
-
"eval_samples_per_second": 15.
|
98 |
-
"eval_steps_per_second": 1.
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
-
"epoch": 0.
|
103 |
-
"grad_norm":
|
104 |
-
"learning_rate": 4.
|
105 |
-
"loss": 0.
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
-
"epoch": 0.
|
110 |
-
"eval_loss": 0.
|
111 |
-
"eval_runtime":
|
112 |
-
"eval_samples_per_second": 15.
|
113 |
-
"eval_steps_per_second": 1.
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
-
"epoch": 0.
|
118 |
-
"grad_norm":
|
119 |
-
"learning_rate": 4.
|
120 |
-
"loss": 0.
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
-
"epoch": 0.
|
125 |
-
"eval_loss": 0.
|
126 |
-
"eval_runtime":
|
127 |
-
"eval_samples_per_second": 15.
|
128 |
-
"eval_steps_per_second": 1.
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
-
"epoch": 0.
|
133 |
-
"grad_norm":
|
134 |
-
"learning_rate":
|
135 |
-
"loss": 0.
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
-
"epoch": 0.
|
140 |
-
"eval_loss": 0.
|
141 |
-
"eval_runtime":
|
142 |
-
"eval_samples_per_second": 15.
|
143 |
-
"eval_steps_per_second": 1.
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
-
"epoch": 0.
|
148 |
-
"grad_norm":
|
149 |
-
"learning_rate":
|
150 |
-
"loss": 0.
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
-
"epoch": 0.
|
155 |
-
"eval_loss": 0.
|
156 |
-
"eval_runtime":
|
157 |
-
"eval_samples_per_second":
|
158 |
-
"eval_steps_per_second":
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
-
"epoch":
|
163 |
-
"grad_norm":
|
164 |
-
"learning_rate":
|
165 |
-
"loss": 0.
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
-
"epoch":
|
170 |
-
"eval_loss": 0.
|
171 |
-
"eval_runtime":
|
172 |
-
"eval_samples_per_second": 15.
|
173 |
-
"eval_steps_per_second": 1.
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
-
"epoch":
|
178 |
-
"grad_norm": 0.
|
179 |
-
"learning_rate":
|
180 |
-
"loss": 0.
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
-
"epoch":
|
185 |
-
"eval_loss": 0.
|
186 |
-
"eval_runtime":
|
187 |
-
"eval_samples_per_second":
|
188 |
-
"eval_steps_per_second":
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
-
"epoch":
|
193 |
-
"grad_norm":
|
194 |
-
"learning_rate":
|
195 |
-
"loss": 0.
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
-
"epoch":
|
200 |
-
"eval_loss": 0.
|
201 |
-
"eval_runtime":
|
202 |
-
"eval_samples_per_second": 15.
|
203 |
-
"eval_steps_per_second": 1.
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
-
"epoch":
|
208 |
-
"grad_norm":
|
209 |
-
"learning_rate":
|
210 |
-
"loss": 0.
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
-
"epoch":
|
215 |
-
"eval_loss": 0.
|
216 |
-
"eval_runtime":
|
217 |
-
"eval_samples_per_second":
|
218 |
-
"eval_steps_per_second":
|
219 |
"step": 7000
|
220 |
},
|
221 |
{
|
222 |
-
"epoch":
|
223 |
-
"grad_norm":
|
224 |
-
"learning_rate": 3.
|
225 |
-
"loss": 0.
|
226 |
"step": 7500
|
227 |
},
|
228 |
{
|
229 |
-
"epoch":
|
230 |
-
"eval_loss": 0.
|
231 |
-
"eval_runtime":
|
232 |
-
"eval_samples_per_second":
|
233 |
-
"eval_steps_per_second":
|
234 |
"step": 7500
|
235 |
},
|
236 |
{
|
237 |
-
"epoch":
|
238 |
-
"grad_norm":
|
239 |
-
"learning_rate": 3.
|
240 |
-
"loss": 0.
|
241 |
"step": 8000
|
242 |
},
|
243 |
{
|
244 |
-
"epoch":
|
245 |
-
"eval_loss": 0.
|
246 |
-
"eval_runtime":
|
247 |
-
"eval_samples_per_second": 15.
|
248 |
-
"eval_steps_per_second": 1.
|
249 |
"step": 8000
|
250 |
},
|
251 |
{
|
252 |
-
"epoch":
|
253 |
-
"grad_norm":
|
254 |
-
"learning_rate": 3.
|
255 |
-
"loss": 0.
|
256 |
"step": 8500
|
257 |
},
|
258 |
{
|
259 |
-
"epoch":
|
260 |
-
"eval_loss": 0.
|
261 |
-
"eval_runtime":
|
262 |
-
"eval_samples_per_second":
|
263 |
-
"eval_steps_per_second":
|
264 |
"step": 8500
|
265 |
},
|
266 |
{
|
267 |
-
"epoch": 1.
|
268 |
-
"grad_norm":
|
269 |
-
"learning_rate":
|
270 |
-
"loss": 0.
|
271 |
"step": 9000
|
272 |
},
|
273 |
{
|
274 |
-
"epoch": 1.
|
275 |
-
"eval_loss": 0.
|
276 |
-
"eval_runtime":
|
277 |
-
"eval_samples_per_second": 15.
|
278 |
"eval_steps_per_second": 1.98,
|
279 |
"step": 9000
|
280 |
},
|
281 |
{
|
282 |
-
"epoch": 1.
|
283 |
-
"grad_norm":
|
284 |
-
"learning_rate":
|
285 |
-
"loss": 0.
|
286 |
"step": 9500
|
287 |
},
|
288 |
{
|
289 |
-
"epoch": 1.
|
290 |
-
"eval_loss": 0.
|
291 |
-
"eval_runtime":
|
292 |
-
"eval_samples_per_second": 15.
|
293 |
-
"eval_steps_per_second": 1.
|
294 |
"step": 9500
|
295 |
},
|
296 |
{
|
297 |
-
"epoch": 1.
|
298 |
-
"grad_norm": 0.
|
299 |
-
"learning_rate":
|
300 |
-
"loss": 0.
|
301 |
"step": 10000
|
302 |
},
|
303 |
{
|
304 |
-
"epoch": 1.
|
305 |
-
"eval_loss": 0.
|
306 |
-
"eval_runtime":
|
307 |
-
"eval_samples_per_second":
|
308 |
-
"eval_steps_per_second":
|
309 |
"step": 10000
|
310 |
},
|
311 |
{
|
312 |
-
"epoch": 1.
|
313 |
-
"grad_norm":
|
314 |
-
"learning_rate":
|
315 |
-
"loss": 0.
|
316 |
"step": 10500
|
317 |
},
|
318 |
{
|
319 |
-
"epoch": 1.
|
320 |
-
"eval_loss": 0.
|
321 |
-
"eval_runtime":
|
322 |
-
"eval_samples_per_second": 15.
|
323 |
-
"eval_steps_per_second": 1.
|
324 |
"step": 10500
|
325 |
},
|
326 |
{
|
327 |
-
"epoch":
|
328 |
-
"grad_norm":
|
329 |
-
"learning_rate":
|
330 |
-
"loss": 0.
|
331 |
"step": 11000
|
332 |
},
|
333 |
{
|
334 |
-
"epoch":
|
335 |
-
"eval_loss": 0.
|
336 |
-
"eval_runtime":
|
337 |
-
"eval_samples_per_second": 15.
|
338 |
-
"eval_steps_per_second": 1.
|
339 |
"step": 11000
|
340 |
},
|
341 |
{
|
342 |
-
"epoch":
|
343 |
-
"grad_norm":
|
344 |
-
"learning_rate":
|
345 |
-
"loss": 0.
|
346 |
"step": 11500
|
347 |
},
|
348 |
{
|
349 |
-
"epoch":
|
350 |
-
"eval_loss": 0.
|
351 |
-
"eval_runtime":
|
352 |
-
"eval_samples_per_second":
|
353 |
-
"eval_steps_per_second":
|
354 |
"step": 11500
|
355 |
},
|
356 |
{
|
357 |
-
"epoch":
|
358 |
-
"grad_norm":
|
359 |
-
"learning_rate":
|
360 |
-
"loss": 0.
|
361 |
"step": 12000
|
362 |
},
|
363 |
{
|
364 |
-
"epoch":
|
365 |
-
"eval_loss": 0.
|
366 |
-
"eval_runtime":
|
367 |
-
"eval_samples_per_second": 15.
|
368 |
-
"eval_steps_per_second": 1.
|
369 |
"step": 12000
|
370 |
},
|
371 |
{
|
372 |
-
"epoch":
|
373 |
-
"grad_norm":
|
374 |
-
"learning_rate":
|
375 |
-
"loss": 0.
|
376 |
"step": 12500
|
377 |
},
|
378 |
{
|
379 |
-
"epoch":
|
380 |
-
"eval_loss": 0.
|
381 |
-
"eval_runtime":
|
382 |
-
"eval_samples_per_second": 15.
|
383 |
-
"eval_steps_per_second": 1.
|
384 |
"step": 12500
|
385 |
},
|
386 |
{
|
387 |
-
"epoch":
|
388 |
-
"grad_norm":
|
389 |
-
"learning_rate":
|
390 |
-
"loss": 0.
|
391 |
"step": 13000
|
392 |
},
|
393 |
{
|
394 |
-
"epoch":
|
395 |
-
"eval_loss": 0.
|
396 |
-
"eval_runtime":
|
397 |
-
"eval_samples_per_second": 15.
|
398 |
-
"eval_steps_per_second": 1.
|
399 |
"step": 13000
|
400 |
},
|
401 |
{
|
402 |
-
"epoch":
|
403 |
-
"grad_norm": 0.
|
404 |
-
"learning_rate":
|
405 |
-
"loss": 0.
|
406 |
"step": 13500
|
407 |
},
|
408 |
{
|
409 |
-
"epoch":
|
410 |
-
"eval_loss": 0.
|
411 |
-
"eval_runtime":
|
412 |
-
"eval_samples_per_second": 15.
|
413 |
-
"eval_steps_per_second": 1.
|
414 |
"step": 13500
|
415 |
},
|
416 |
{
|
417 |
-
"epoch":
|
418 |
-
"grad_norm":
|
419 |
-
"learning_rate":
|
420 |
-
"loss": 0.
|
421 |
"step": 14000
|
422 |
},
|
423 |
{
|
424 |
-
"epoch":
|
425 |
-
"eval_loss": 0.
|
426 |
-
"eval_runtime":
|
427 |
-
"eval_samples_per_second":
|
428 |
-
"eval_steps_per_second":
|
429 |
"step": 14000
|
430 |
},
|
431 |
{
|
432 |
-
"epoch":
|
433 |
-
"grad_norm":
|
434 |
-
"learning_rate":
|
435 |
-
"loss": 0.
|
436 |
"step": 14500
|
437 |
},
|
438 |
{
|
439 |
-
"epoch":
|
440 |
-
"eval_loss": 0.
|
441 |
-
"eval_runtime":
|
442 |
-
"eval_samples_per_second":
|
443 |
-
"eval_steps_per_second":
|
444 |
"step": 14500
|
445 |
},
|
446 |
{
|
447 |
-
"epoch":
|
448 |
-
"grad_norm":
|
449 |
-
"learning_rate":
|
450 |
-
"loss": 0.
|
451 |
"step": 15000
|
452 |
},
|
453 |
{
|
454 |
-
"epoch":
|
455 |
-
"eval_loss": 0.
|
456 |
-
"eval_runtime":
|
457 |
-
"eval_samples_per_second": 15.
|
458 |
-
"eval_steps_per_second": 1.
|
459 |
"step": 15000
|
460 |
},
|
461 |
{
|
462 |
-
"epoch":
|
463 |
-
"grad_norm":
|
464 |
-
"learning_rate":
|
465 |
-
"loss": 0.
|
466 |
"step": 15500
|
467 |
},
|
468 |
{
|
469 |
-
"epoch":
|
470 |
-
"eval_loss": 0.
|
471 |
-
"eval_runtime":
|
472 |
-
"eval_samples_per_second":
|
473 |
-
"eval_steps_per_second":
|
474 |
"step": 15500
|
475 |
},
|
476 |
{
|
477 |
-
"epoch":
|
478 |
-
"grad_norm":
|
479 |
-
"learning_rate":
|
480 |
-
"loss": 0.
|
481 |
"step": 16000
|
482 |
},
|
483 |
{
|
484 |
-
"epoch":
|
485 |
-
"eval_loss": 0.
|
486 |
-
"eval_runtime":
|
487 |
-
"eval_samples_per_second":
|
488 |
-
"eval_steps_per_second":
|
489 |
"step": 16000
|
490 |
},
|
491 |
{
|
492 |
-
"epoch":
|
493 |
-
"grad_norm":
|
494 |
-
"learning_rate":
|
495 |
-
"loss": 0.
|
496 |
"step": 16500
|
497 |
},
|
498 |
{
|
499 |
-
"epoch":
|
500 |
-
"eval_loss": 0.
|
501 |
-
"eval_runtime":
|
502 |
-
"eval_samples_per_second": 15.
|
503 |
-
"eval_steps_per_second": 1.
|
504 |
"step": 16500
|
505 |
},
|
506 |
{
|
507 |
-
"epoch":
|
508 |
-
"grad_norm":
|
509 |
-
"learning_rate":
|
510 |
-
"loss": 0.
|
511 |
"step": 17000
|
512 |
},
|
513 |
{
|
514 |
-
"epoch":
|
515 |
-
"eval_loss": 0.
|
516 |
-
"eval_runtime":
|
517 |
-
"eval_samples_per_second":
|
518 |
-
"eval_steps_per_second":
|
519 |
"step": 17000
|
520 |
},
|
521 |
{
|
522 |
-
"epoch":
|
523 |
-
"grad_norm": 0.
|
524 |
-
"learning_rate":
|
525 |
-
"loss": 0.
|
526 |
"step": 17500
|
527 |
},
|
528 |
{
|
529 |
-
"epoch":
|
530 |
-
"eval_loss": 0.
|
531 |
-
"eval_runtime":
|
532 |
-
"eval_samples_per_second":
|
533 |
-
"eval_steps_per_second":
|
534 |
"step": 17500
|
535 |
},
|
536 |
{
|
537 |
-
"epoch":
|
538 |
-
"grad_norm":
|
539 |
-
"learning_rate":
|
540 |
-
"loss": 0.
|
541 |
"step": 18000
|
542 |
},
|
543 |
{
|
544 |
-
"epoch":
|
545 |
-
"eval_loss": 0.
|
546 |
-
"eval_runtime":
|
547 |
-
"eval_samples_per_second": 15.
|
548 |
-
"eval_steps_per_second": 1.
|
549 |
"step": 18000
|
550 |
},
|
551 |
{
|
552 |
-
"epoch":
|
553 |
-
"grad_norm":
|
554 |
-
"learning_rate":
|
555 |
-
"loss": 0.
|
556 |
"step": 18500
|
557 |
},
|
558 |
{
|
559 |
-
"epoch":
|
560 |
-
"eval_loss": 0.
|
561 |
-
"eval_runtime":
|
562 |
-
"eval_samples_per_second":
|
563 |
-
"eval_steps_per_second":
|
564 |
"step": 18500
|
565 |
},
|
566 |
{
|
567 |
-
"epoch":
|
568 |
-
"grad_norm":
|
569 |
-
"learning_rate":
|
570 |
-
"loss": 0.
|
571 |
"step": 19000
|
572 |
},
|
573 |
{
|
574 |
-
"epoch":
|
575 |
-
"eval_loss": 0.
|
576 |
-
"eval_runtime":
|
577 |
-
"eval_samples_per_second":
|
578 |
-
"eval_steps_per_second":
|
579 |
"step": 19000
|
580 |
},
|
581 |
{
|
582 |
-
"epoch":
|
583 |
-
"grad_norm":
|
584 |
-
"learning_rate":
|
585 |
-
"loss": 0.
|
586 |
"step": 19500
|
587 |
},
|
588 |
{
|
589 |
-
"epoch":
|
590 |
-
"eval_loss": 0.
|
591 |
-
"eval_runtime":
|
592 |
-
"eval_samples_per_second": 15.
|
593 |
-
"eval_steps_per_second": 1.
|
594 |
"step": 19500
|
595 |
},
|
596 |
{
|
597 |
-
"epoch":
|
598 |
-
"grad_norm":
|
599 |
-
"learning_rate":
|
600 |
-
"loss": 0.
|
601 |
"step": 20000
|
602 |
},
|
603 |
{
|
604 |
-
"epoch":
|
605 |
-
"eval_loss": 0.
|
606 |
-
"eval_runtime":
|
607 |
-
"eval_samples_per_second":
|
608 |
-
"eval_steps_per_second":
|
609 |
"step": 20000
|
610 |
},
|
611 |
{
|
612 |
-
"epoch":
|
613 |
-
"grad_norm":
|
614 |
-
"learning_rate": 2.
|
615 |
-
"loss": 0.
|
616 |
"step": 20500
|
617 |
},
|
618 |
{
|
619 |
-
"epoch":
|
620 |
-
"eval_loss": 0.
|
621 |
-
"eval_runtime":
|
622 |
-
"eval_samples_per_second":
|
623 |
-
"eval_steps_per_second":
|
624 |
"step": 20500
|
625 |
},
|
626 |
{
|
627 |
-
"epoch":
|
628 |
-
"grad_norm": 1.
|
629 |
-
"learning_rate":
|
630 |
-
"loss": 0.
|
631 |
"step": 21000
|
632 |
},
|
633 |
{
|
634 |
-
"epoch":
|
635 |
-
"eval_loss": 0.
|
636 |
-
"eval_runtime":
|
637 |
-
"eval_samples_per_second": 15.
|
638 |
-
"eval_steps_per_second": 1.
|
639 |
"step": 21000
|
640 |
}
|
641 |
],
|
642 |
"logging_steps": 500,
|
643 |
-
"max_steps":
|
644 |
"num_input_tokens_seen": 0,
|
645 |
"num_train_epochs": 4,
|
646 |
"save_steps": 500,
|
@@ -656,7 +656,7 @@
|
|
656 |
"attributes": {}
|
657 |
}
|
658 |
},
|
659 |
-
"total_flos":
|
660 |
"train_batch_size": 2,
|
661 |
"trial_name": null,
|
662 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.5568962693214417,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-18500",
|
4 |
+
"epoch": 2.3640661938534278,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 21000,
|
7 |
"is_hyper_param_search": false,
|
|
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.056287290329843524,
|
13 |
+
"grad_norm": 150.1642303466797,
|
14 |
+
"learning_rate": 4.929640887087696e-07,
|
15 |
+
"loss": 0.3486,
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.056287290329843524,
|
20 |
+
"eval_loss": 1.1265727281570435,
|
21 |
+
"eval_runtime": 123.7806,
|
22 |
+
"eval_samples_per_second": 15.948,
|
23 |
+
"eval_steps_per_second": 1.995,
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
+
"epoch": 0.11257458065968705,
|
28 |
+
"grad_norm": 366.6913146972656,
|
29 |
+
"learning_rate": 4.859281774175391e-07,
|
30 |
+
"loss": 0.2733,
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
+
"epoch": 0.11257458065968705,
|
35 |
+
"eval_loss": 0.9742079377174377,
|
36 |
+
"eval_runtime": 125.8047,
|
37 |
+
"eval_samples_per_second": 15.691,
|
38 |
+
"eval_steps_per_second": 1.963,
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
+
"epoch": 0.16886187098953057,
|
43 |
+
"grad_norm": 555.5016479492188,
|
44 |
+
"learning_rate": 4.788922661263087e-07,
|
45 |
+
"loss": 0.1851,
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
+
"epoch": 0.16886187098953057,
|
50 |
+
"eval_loss": 0.9162159562110901,
|
51 |
+
"eval_runtime": 126.7606,
|
52 |
+
"eval_samples_per_second": 15.573,
|
53 |
+
"eval_steps_per_second": 1.949,
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
+
"epoch": 0.2251491613193741,
|
58 |
+
"grad_norm": 0.423260897397995,
|
59 |
+
"learning_rate": 4.7185635483507824e-07,
|
60 |
+
"loss": 0.1973,
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
+
"epoch": 0.2251491613193741,
|
65 |
+
"eval_loss": 0.8716973662376404,
|
66 |
+
"eval_runtime": 127.6955,
|
67 |
+
"eval_samples_per_second": 15.459,
|
68 |
+
"eval_steps_per_second": 1.934,
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
+
"epoch": 0.2814364516492176,
|
73 |
+
"grad_norm": 6.840139389038086,
|
74 |
+
"learning_rate": 4.6482044354384774e-07,
|
75 |
+
"loss": 0.1881,
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
+
"epoch": 0.2814364516492176,
|
80 |
+
"eval_loss": 0.830047070980072,
|
81 |
+
"eval_runtime": 128.0771,
|
82 |
+
"eval_samples_per_second": 15.413,
|
83 |
+
"eval_steps_per_second": 1.929,
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
+
"epoch": 0.33772374197906113,
|
88 |
+
"grad_norm": 729.8377685546875,
|
89 |
+
"learning_rate": 4.577845322526173e-07,
|
90 |
+
"loss": 0.1695,
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
+
"epoch": 0.33772374197906113,
|
95 |
+
"eval_loss": 0.8265627026557922,
|
96 |
+
"eval_runtime": 128.052,
|
97 |
+
"eval_samples_per_second": 15.416,
|
98 |
+
"eval_steps_per_second": 1.929,
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
+
"epoch": 0.39401103230890466,
|
103 |
+
"grad_norm": 7.96999938756926e-06,
|
104 |
+
"learning_rate": 4.507486209613869e-07,
|
105 |
+
"loss": 0.155,
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
+
"epoch": 0.39401103230890466,
|
110 |
+
"eval_loss": 0.7721038460731506,
|
111 |
+
"eval_runtime": 128.1871,
|
112 |
+
"eval_samples_per_second": 15.399,
|
113 |
+
"eval_steps_per_second": 1.927,
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
+
"epoch": 0.4502983226387482,
|
118 |
+
"grad_norm": 0.006555848754942417,
|
119 |
+
"learning_rate": 4.4371270967015645e-07,
|
120 |
+
"loss": 0.1216,
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
+
"epoch": 0.4502983226387482,
|
125 |
+
"eval_loss": 0.7490401864051819,
|
126 |
+
"eval_runtime": 127.5516,
|
127 |
+
"eval_samples_per_second": 15.476,
|
128 |
+
"eval_steps_per_second": 1.936,
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
+
"epoch": 0.5065856129685917,
|
133 |
+
"grad_norm": 0.0004515685432124883,
|
134 |
+
"learning_rate": 4.36676798378926e-07,
|
135 |
+
"loss": 0.1531,
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
+
"epoch": 0.5065856129685917,
|
140 |
+
"eval_loss": 0.7533182501792908,
|
141 |
+
"eval_runtime": 127.4797,
|
142 |
+
"eval_samples_per_second": 15.485,
|
143 |
+
"eval_steps_per_second": 1.938,
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
+
"epoch": 0.5628729032984352,
|
148 |
+
"grad_norm": 4.4633176003117114e-05,
|
149 |
+
"learning_rate": 4.2964088708769556e-07,
|
150 |
+
"loss": 0.1154,
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
+
"epoch": 0.5628729032984352,
|
155 |
+
"eval_loss": 0.747530460357666,
|
156 |
+
"eval_runtime": 127.8675,
|
157 |
+
"eval_samples_per_second": 15.438,
|
158 |
+
"eval_steps_per_second": 1.932,
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
+
"epoch": 0.6191601936282788,
|
163 |
+
"grad_norm": 4.158839702606201,
|
164 |
+
"learning_rate": 4.2260497579646517e-07,
|
165 |
+
"loss": 0.1407,
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
+
"epoch": 0.6191601936282788,
|
170 |
+
"eval_loss": 0.724854588508606,
|
171 |
+
"eval_runtime": 125.1629,
|
172 |
+
"eval_samples_per_second": 15.771,
|
173 |
+
"eval_steps_per_second": 1.973,
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
+
"epoch": 0.6754474839581223,
|
178 |
+
"grad_norm": 0.748630166053772,
|
179 |
+
"learning_rate": 4.155690645052347e-07,
|
180 |
+
"loss": 0.1447,
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
+
"epoch": 0.6754474839581223,
|
185 |
+
"eval_loss": 0.7091771960258484,
|
186 |
+
"eval_runtime": 125.2549,
|
187 |
+
"eval_samples_per_second": 15.76,
|
188 |
+
"eval_steps_per_second": 1.972,
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
+
"epoch": 0.7317347742879657,
|
193 |
+
"grad_norm": 2.8120924980612472e-05,
|
194 |
+
"learning_rate": 4.0853315321400427e-07,
|
195 |
+
"loss": 0.1146,
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
+
"epoch": 0.7317347742879657,
|
200 |
+
"eval_loss": 0.7105826735496521,
|
201 |
+
"eval_runtime": 127.7786,
|
202 |
+
"eval_samples_per_second": 15.449,
|
203 |
+
"eval_steps_per_second": 1.933,
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
+
"epoch": 0.7880220646178093,
|
208 |
+
"grad_norm": 1.4741635823156685e-07,
|
209 |
+
"learning_rate": 4.014972419227738e-07,
|
210 |
+
"loss": 0.171,
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
+
"epoch": 0.7880220646178093,
|
215 |
+
"eval_loss": 0.6978534460067749,
|
216 |
+
"eval_runtime": 125.2567,
|
217 |
+
"eval_samples_per_second": 15.76,
|
218 |
+
"eval_steps_per_second": 1.972,
|
219 |
"step": 7000
|
220 |
},
|
221 |
{
|
222 |
+
"epoch": 0.8443093549476528,
|
223 |
+
"grad_norm": 0.002975167240947485,
|
224 |
+
"learning_rate": 3.944613306315434e-07,
|
225 |
+
"loss": 0.1584,
|
226 |
"step": 7500
|
227 |
},
|
228 |
{
|
229 |
+
"epoch": 0.8443093549476528,
|
230 |
+
"eval_loss": 0.6923494935035706,
|
231 |
+
"eval_runtime": 125.3384,
|
232 |
+
"eval_samples_per_second": 15.749,
|
233 |
+
"eval_steps_per_second": 1.971,
|
234 |
"step": 7500
|
235 |
},
|
236 |
{
|
237 |
+
"epoch": 0.9005966452774964,
|
238 |
+
"grad_norm": 9.225498797604814e-05,
|
239 |
+
"learning_rate": 3.8742541934031293e-07,
|
240 |
+
"loss": 0.1384,
|
241 |
"step": 8000
|
242 |
},
|
243 |
{
|
244 |
+
"epoch": 0.9005966452774964,
|
245 |
+
"eval_loss": 0.6872902512550354,
|
246 |
+
"eval_runtime": 128.114,
|
247 |
+
"eval_samples_per_second": 15.408,
|
248 |
+
"eval_steps_per_second": 1.928,
|
249 |
"step": 8000
|
250 |
},
|
251 |
{
|
252 |
+
"epoch": 0.9568839356073399,
|
253 |
+
"grad_norm": 1.9474915902151224e-08,
|
254 |
+
"learning_rate": 3.803895080490825e-07,
|
255 |
+
"loss": 0.1243,
|
256 |
"step": 8500
|
257 |
},
|
258 |
{
|
259 |
+
"epoch": 0.9568839356073399,
|
260 |
+
"eval_loss": 0.6788680553436279,
|
261 |
+
"eval_runtime": 125.0525,
|
262 |
+
"eval_samples_per_second": 15.785,
|
263 |
+
"eval_steps_per_second": 1.975,
|
264 |
"step": 8500
|
265 |
},
|
266 |
{
|
267 |
+
"epoch": 1.0131712259371835,
|
268 |
+
"grad_norm": 5.165647506713867,
|
269 |
+
"learning_rate": 3.7335359675785204e-07,
|
270 |
+
"loss": 0.1102,
|
271 |
"step": 9000
|
272 |
},
|
273 |
{
|
274 |
+
"epoch": 1.0131712259371835,
|
275 |
+
"eval_loss": 0.6788864135742188,
|
276 |
+
"eval_runtime": 124.7777,
|
277 |
+
"eval_samples_per_second": 15.82,
|
278 |
"eval_steps_per_second": 1.98,
|
279 |
"step": 9000
|
280 |
},
|
281 |
{
|
282 |
+
"epoch": 1.069458516267027,
|
283 |
+
"grad_norm": 0.0005840375670231879,
|
284 |
+
"learning_rate": 3.6631768546662164e-07,
|
285 |
+
"loss": 0.0523,
|
286 |
"step": 9500
|
287 |
},
|
288 |
{
|
289 |
+
"epoch": 1.069458516267027,
|
290 |
+
"eval_loss": 0.6628187894821167,
|
291 |
+
"eval_runtime": 127.8018,
|
292 |
+
"eval_samples_per_second": 15.446,
|
293 |
+
"eval_steps_per_second": 1.933,
|
294 |
"step": 9500
|
295 |
},
|
296 |
{
|
297 |
+
"epoch": 1.1257458065968704,
|
298 |
+
"grad_norm": 0.001191094284877181,
|
299 |
+
"learning_rate": 3.592817741753912e-07,
|
300 |
+
"loss": 0.0617,
|
301 |
"step": 10000
|
302 |
},
|
303 |
{
|
304 |
+
"epoch": 1.1257458065968704,
|
305 |
+
"eval_loss": 0.6507942080497742,
|
306 |
+
"eval_runtime": 125.484,
|
307 |
+
"eval_samples_per_second": 15.731,
|
308 |
+
"eval_steps_per_second": 1.968,
|
309 |
"step": 10000
|
310 |
},
|
311 |
{
|
312 |
+
"epoch": 1.1820330969267139,
|
313 |
+
"grad_norm": 0.0003092484548687935,
|
314 |
+
"learning_rate": 3.5224586288416075e-07,
|
315 |
+
"loss": 0.0598,
|
316 |
"step": 10500
|
317 |
},
|
318 |
{
|
319 |
+
"epoch": 1.1820330969267139,
|
320 |
+
"eval_loss": 0.6480408906936646,
|
321 |
+
"eval_runtime": 125.7661,
|
322 |
+
"eval_samples_per_second": 15.696,
|
323 |
+
"eval_steps_per_second": 1.964,
|
324 |
"step": 10500
|
325 |
},
|
326 |
{
|
327 |
+
"epoch": 1.2383203872565574,
|
328 |
+
"grad_norm": 5.508670710696606e-07,
|
329 |
+
"learning_rate": 3.452099515929303e-07,
|
330 |
+
"loss": 0.0782,
|
331 |
"step": 11000
|
332 |
},
|
333 |
{
|
334 |
+
"epoch": 1.2383203872565574,
|
335 |
+
"eval_loss": 0.6403237581253052,
|
336 |
+
"eval_runtime": 128.3947,
|
337 |
+
"eval_samples_per_second": 15.374,
|
338 |
+
"eval_steps_per_second": 1.924,
|
339 |
"step": 11000
|
340 |
},
|
341 |
{
|
342 |
+
"epoch": 1.294607677586401,
|
343 |
+
"grad_norm": 0.1470593810081482,
|
344 |
+
"learning_rate": 3.3817404030169986e-07,
|
345 |
+
"loss": 0.0616,
|
346 |
"step": 11500
|
347 |
},
|
348 |
{
|
349 |
+
"epoch": 1.294607677586401,
|
350 |
+
"eval_loss": 0.6376460790634155,
|
351 |
+
"eval_runtime": 125.5391,
|
352 |
+
"eval_samples_per_second": 15.724,
|
353 |
+
"eval_steps_per_second": 1.968,
|
354 |
"step": 11500
|
355 |
},
|
356 |
{
|
357 |
+
"epoch": 1.3508949679162445,
|
358 |
+
"grad_norm": 1.8021532014245167e-05,
|
359 |
+
"learning_rate": 3.3113812901046946e-07,
|
360 |
+
"loss": 0.0723,
|
361 |
"step": 12000
|
362 |
},
|
363 |
{
|
364 |
+
"epoch": 1.3508949679162445,
|
365 |
+
"eval_loss": 0.6378474831581116,
|
366 |
+
"eval_runtime": 125.5914,
|
367 |
+
"eval_samples_per_second": 15.718,
|
368 |
+
"eval_steps_per_second": 1.967,
|
369 |
"step": 12000
|
370 |
},
|
371 |
{
|
372 |
+
"epoch": 1.407182258246088,
|
373 |
+
"grad_norm": 0.00013483635848388076,
|
374 |
+
"learning_rate": 3.2410221771923896e-07,
|
375 |
+
"loss": 0.0459,
|
376 |
"step": 12500
|
377 |
},
|
378 |
{
|
379 |
+
"epoch": 1.407182258246088,
|
380 |
+
"eval_loss": 0.6378260850906372,
|
381 |
+
"eval_runtime": 128.2725,
|
382 |
+
"eval_samples_per_second": 15.389,
|
383 |
+
"eval_steps_per_second": 1.926,
|
384 |
"step": 12500
|
385 |
},
|
386 |
{
|
387 |
+
"epoch": 1.4634695485759315,
|
388 |
+
"grad_norm": 4.757418707868055e-07,
|
389 |
+
"learning_rate": 3.170663064280085e-07,
|
390 |
+
"loss": 0.0571,
|
391 |
"step": 13000
|
392 |
},
|
393 |
{
|
394 |
+
"epoch": 1.4634695485759315,
|
395 |
+
"eval_loss": 0.6366899609565735,
|
396 |
+
"eval_runtime": 125.6986,
|
397 |
+
"eval_samples_per_second": 15.704,
|
398 |
+
"eval_steps_per_second": 1.965,
|
399 |
"step": 13000
|
400 |
},
|
401 |
{
|
402 |
+
"epoch": 1.5197568389057752,
|
403 |
+
"grad_norm": 0.00028420978924259543,
|
404 |
+
"learning_rate": 3.1003039513677807e-07,
|
405 |
+
"loss": 0.0499,
|
406 |
"step": 13500
|
407 |
},
|
408 |
{
|
409 |
+
"epoch": 1.5197568389057752,
|
410 |
+
"eval_loss": 0.6286123394966125,
|
411 |
+
"eval_runtime": 127.9165,
|
412 |
+
"eval_samples_per_second": 15.432,
|
413 |
+
"eval_steps_per_second": 1.931,
|
414 |
"step": 13500
|
415 |
},
|
416 |
{
|
417 |
+
"epoch": 1.5760441292356187,
|
418 |
+
"grad_norm": 2.1999912291903456e-07,
|
419 |
+
"learning_rate": 3.0299448384554767e-07,
|
420 |
+
"loss": 0.0679,
|
421 |
"step": 14000
|
422 |
},
|
423 |
{
|
424 |
+
"epoch": 1.5760441292356187,
|
425 |
+
"eval_loss": 0.634914755821228,
|
426 |
+
"eval_runtime": 125.0396,
|
427 |
+
"eval_samples_per_second": 15.787,
|
428 |
+
"eval_steps_per_second": 1.975,
|
429 |
"step": 14000
|
430 |
},
|
431 |
{
|
432 |
+
"epoch": 1.6323314195654621,
|
433 |
+
"grad_norm": 0.00023505109129473567,
|
434 |
+
"learning_rate": 2.959585725543172e-07,
|
435 |
+
"loss": 0.0554,
|
436 |
"step": 14500
|
437 |
},
|
438 |
{
|
439 |
+
"epoch": 1.6323314195654621,
|
440 |
+
"eval_loss": 0.623322069644928,
|
441 |
+
"eval_runtime": 125.2429,
|
442 |
+
"eval_samples_per_second": 15.761,
|
443 |
+
"eval_steps_per_second": 1.972,
|
444 |
"step": 14500
|
445 |
},
|
446 |
{
|
447 |
+
"epoch": 1.6886187098953056,
|
448 |
+
"grad_norm": 11.346132278442383,
|
449 |
+
"learning_rate": 2.889226612630868e-07,
|
450 |
+
"loss": 0.0617,
|
451 |
"step": 15000
|
452 |
},
|
453 |
{
|
454 |
+
"epoch": 1.6886187098953056,
|
455 |
+
"eval_loss": 0.6115825772285461,
|
456 |
+
"eval_runtime": 127.8813,
|
457 |
+
"eval_samples_per_second": 15.436,
|
458 |
+
"eval_steps_per_second": 1.931,
|
459 |
"step": 15000
|
460 |
},
|
461 |
{
|
462 |
+
"epoch": 1.744906000225149,
|
463 |
+
"grad_norm": 308.43804931640625,
|
464 |
+
"learning_rate": 2.8188674997185633e-07,
|
465 |
+
"loss": 0.0561,
|
466 |
"step": 15500
|
467 |
},
|
468 |
{
|
469 |
+
"epoch": 1.744906000225149,
|
470 |
+
"eval_loss": 0.5951005220413208,
|
471 |
+
"eval_runtime": 125.2491,
|
472 |
+
"eval_samples_per_second": 15.761,
|
473 |
+
"eval_steps_per_second": 1.972,
|
474 |
"step": 15500
|
475 |
},
|
476 |
{
|
477 |
+
"epoch": 1.8011932905549926,
|
478 |
+
"grad_norm": 6.91558055265773e-09,
|
479 |
+
"learning_rate": 2.7485083868062594e-07,
|
480 |
+
"loss": 0.0672,
|
481 |
"step": 16000
|
482 |
},
|
483 |
{
|
484 |
+
"epoch": 1.8011932905549926,
|
485 |
+
"eval_loss": 0.5809568166732788,
|
486 |
+
"eval_runtime": 125.2557,
|
487 |
+
"eval_samples_per_second": 15.76,
|
488 |
+
"eval_steps_per_second": 1.972,
|
489 |
"step": 16000
|
490 |
},
|
491 |
{
|
492 |
+
"epoch": 1.8574805808848363,
|
493 |
+
"grad_norm": 2.2250075915053458e-07,
|
494 |
+
"learning_rate": 2.678149273893955e-07,
|
495 |
+
"loss": 0.0812,
|
496 |
"step": 16500
|
497 |
},
|
498 |
{
|
499 |
+
"epoch": 1.8574805808848363,
|
500 |
+
"eval_loss": 0.5691242218017578,
|
501 |
+
"eval_runtime": 128.3032,
|
502 |
+
"eval_samples_per_second": 15.385,
|
503 |
+
"eval_steps_per_second": 1.925,
|
504 |
"step": 16500
|
505 |
},
|
506 |
{
|
507 |
+
"epoch": 1.9137678712146797,
|
508 |
+
"grad_norm": 1.1724663972854614,
|
509 |
+
"learning_rate": 2.6077901609816504e-07,
|
510 |
+
"loss": 0.0614,
|
511 |
"step": 17000
|
512 |
},
|
513 |
{
|
514 |
+
"epoch": 1.9137678712146797,
|
515 |
+
"eval_loss": 0.5643230080604553,
|
516 |
+
"eval_runtime": 125.7191,
|
517 |
+
"eval_samples_per_second": 15.702,
|
518 |
+
"eval_steps_per_second": 1.965,
|
519 |
"step": 17000
|
520 |
},
|
521 |
{
|
522 |
+
"epoch": 1.9700551615445232,
|
523 |
+
"grad_norm": 0.00030226208036765456,
|
524 |
+
"learning_rate": 2.5374310480693454e-07,
|
525 |
+
"loss": 0.0462,
|
526 |
"step": 17500
|
527 |
},
|
528 |
{
|
529 |
+
"epoch": 1.9700551615445232,
|
530 |
+
"eval_loss": 0.5657238364219666,
|
531 |
+
"eval_runtime": 124.9923,
|
532 |
+
"eval_samples_per_second": 15.793,
|
533 |
+
"eval_steps_per_second": 1.976,
|
534 |
"step": 17500
|
535 |
},
|
536 |
{
|
537 |
+
"epoch": 2.026342451874367,
|
538 |
+
"grad_norm": 1.949754704355655e-07,
|
539 |
+
"learning_rate": 2.4670719351570415e-07,
|
540 |
+
"loss": 0.0565,
|
541 |
"step": 18000
|
542 |
},
|
543 |
{
|
544 |
+
"epoch": 2.026342451874367,
|
545 |
+
"eval_loss": 0.5607953667640686,
|
546 |
+
"eval_runtime": 127.7187,
|
547 |
+
"eval_samples_per_second": 15.456,
|
548 |
+
"eval_steps_per_second": 1.934,
|
549 |
"step": 18000
|
550 |
},
|
551 |
{
|
552 |
+
"epoch": 2.0826297422042104,
|
553 |
+
"grad_norm": 14.2212553024292,
|
554 |
+
"learning_rate": 2.396712822244737e-07,
|
555 |
+
"loss": 0.0549,
|
556 |
"step": 18500
|
557 |
},
|
558 |
{
|
559 |
+
"epoch": 2.0826297422042104,
|
560 |
+
"eval_loss": 0.5568962693214417,
|
561 |
+
"eval_runtime": 125.0503,
|
562 |
+
"eval_samples_per_second": 15.786,
|
563 |
+
"eval_steps_per_second": 1.975,
|
564 |
"step": 18500
|
565 |
},
|
566 |
{
|
567 |
+
"epoch": 2.138917032534054,
|
568 |
+
"grad_norm": 1.2045973539352417,
|
569 |
+
"learning_rate": 2.3263537093324326e-07,
|
570 |
+
"loss": 0.0245,
|
571 |
"step": 19000
|
572 |
},
|
573 |
{
|
574 |
+
"epoch": 2.138917032534054,
|
575 |
+
"eval_loss": 0.5595241785049438,
|
576 |
+
"eval_runtime": 124.8527,
|
577 |
+
"eval_samples_per_second": 15.811,
|
578 |
+
"eval_steps_per_second": 1.978,
|
579 |
"step": 19000
|
580 |
},
|
581 |
{
|
582 |
+
"epoch": 2.1952043228638973,
|
583 |
+
"grad_norm": 8.770056410867255e-06,
|
584 |
+
"learning_rate": 2.2559945964201284e-07,
|
585 |
+
"loss": 0.0428,
|
586 |
"step": 19500
|
587 |
},
|
588 |
{
|
589 |
+
"epoch": 2.1952043228638973,
|
590 |
+
"eval_loss": 0.5740877985954285,
|
591 |
+
"eval_runtime": 127.9211,
|
592 |
+
"eval_samples_per_second": 15.431,
|
593 |
+
"eval_steps_per_second": 1.931,
|
594 |
"step": 19500
|
595 |
},
|
596 |
{
|
597 |
+
"epoch": 2.251491613193741,
|
598 |
+
"grad_norm": 6.243853079013206e-08,
|
599 |
+
"learning_rate": 2.185635483507824e-07,
|
600 |
+
"loss": 0.0301,
|
601 |
"step": 20000
|
602 |
},
|
603 |
{
|
604 |
+
"epoch": 2.251491613193741,
|
605 |
+
"eval_loss": 0.5761672854423523,
|
606 |
+
"eval_runtime": 125.2829,
|
607 |
+
"eval_samples_per_second": 15.756,
|
608 |
+
"eval_steps_per_second": 1.972,
|
609 |
"step": 20000
|
610 |
},
|
611 |
{
|
612 |
+
"epoch": 2.3077789035235843,
|
613 |
+
"grad_norm": 2.299124268745345e-08,
|
614 |
+
"learning_rate": 2.1152763705955197e-07,
|
615 |
+
"loss": 0.04,
|
616 |
"step": 20500
|
617 |
},
|
618 |
{
|
619 |
+
"epoch": 2.3077789035235843,
|
620 |
+
"eval_loss": 0.5712103843688965,
|
621 |
+
"eval_runtime": 125.3797,
|
622 |
+
"eval_samples_per_second": 15.744,
|
623 |
+
"eval_steps_per_second": 1.97,
|
624 |
"step": 20500
|
625 |
},
|
626 |
{
|
627 |
+
"epoch": 2.3640661938534278,
|
628 |
+
"grad_norm": 1.971009623957798e-06,
|
629 |
+
"learning_rate": 2.044917257683215e-07,
|
630 |
+
"loss": 0.0349,
|
631 |
"step": 21000
|
632 |
},
|
633 |
{
|
634 |
+
"epoch": 2.3640661938534278,
|
635 |
+
"eval_loss": 0.5690667629241943,
|
636 |
+
"eval_runtime": 125.4733,
|
637 |
+
"eval_samples_per_second": 15.732,
|
638 |
+
"eval_steps_per_second": 1.969,
|
639 |
"step": 21000
|
640 |
}
|
641 |
],
|
642 |
"logging_steps": 500,
|
643 |
+
"max_steps": 35532,
|
644 |
"num_input_tokens_seen": 0,
|
645 |
"num_train_epochs": 4,
|
646 |
"save_steps": 500,
|
|
|
656 |
"attributes": {}
|
657 |
}
|
658 |
},
|
659 |
+
"total_flos": 7555285134811260.0,
|
660 |
"train_batch_size": 2,
|
661 |
"trial_name": null,
|
662 |
"trial_params": null
|
checkpoint-21000/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4847
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:319be2a769ef7117b6826908976f080f9082f159f4f0ad75da1a151ad29d9bff
|
3 |
size 4847
|