Training in progress, step 20000
Browse files- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/pytorch_model.bin +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/rng_state_4.pth +1 -1
- last-checkpoint/rng_state_5.pth +1 -1
- last-checkpoint/rng_state_6.pth +1 -1
- last-checkpoint/rng_state_7.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +203 -3
- pytorch_model.bin +1 -1
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 50044241
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d9c29fbdf49be880cb6493988a94306303a92a77dfa3a5685c194a3ed9741dd7
|
3 |
size 50044241
|
last-checkpoint/pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cab920afe4d207be60e60c397c80474cec3b0781a866712c960668a023bb8b59
|
3 |
size 25761253
|
last-checkpoint/rng_state_0.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_1.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_2.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_3.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_4.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_5.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_6.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/rng_state_7.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:d4daecf9b867e6b1022a006e977cc17780663921ebe9348f84c4f92fd1098fd3
|
3 |
size 14503
|
last-checkpoint/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 sha256:2fa6f3c4ec253d8129a7481e01148ec46428b7a6eb1631c7fb589fd92c25c12f
|
3 |
size 623
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch": 0.
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -206,11 +206,211 @@
|
|
206 |
"eval_samples_per_second": 1027.899,
|
207 |
"eval_steps_per_second": 16.11,
|
208 |
"step": 10000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
209 |
}
|
210 |
],
|
211 |
"max_steps": 250000,
|
212 |
"num_train_epochs": 12,
|
213 |
-
"total_flos":
|
214 |
"trial_name": null,
|
215 |
"trial_params": null
|
216 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.8922198429693077,
|
5 |
+
"global_step": 20000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
206 |
"eval_samples_per_second": 1027.899,
|
207 |
"eval_steps_per_second": 16.11,
|
208 |
"step": 10000
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 0.47,
|
212 |
+
"learning_rate": 0.0005039999999999999,
|
213 |
+
"loss": 0.6775,
|
214 |
+
"step": 10500
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 0.49,
|
218 |
+
"learning_rate": 0.0005279999999999999,
|
219 |
+
"loss": 0.6775,
|
220 |
+
"step": 11000
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 0.49,
|
224 |
+
"eval_loss": 0.6769479513168335,
|
225 |
+
"eval_runtime": 2.177,
|
226 |
+
"eval_samples_per_second": 1055.099,
|
227 |
+
"eval_steps_per_second": 16.536,
|
228 |
+
"step": 11000
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 0.51,
|
232 |
+
"learning_rate": 0.000552,
|
233 |
+
"loss": 0.6773,
|
234 |
+
"step": 11500
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.54,
|
238 |
+
"learning_rate": 0.0005759999999999999,
|
239 |
+
"loss": 0.6773,
|
240 |
+
"step": 12000
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.54,
|
244 |
+
"eval_loss": 0.6773238182067871,
|
245 |
+
"eval_runtime": 2.1281,
|
246 |
+
"eval_samples_per_second": 1079.366,
|
247 |
+
"eval_steps_per_second": 16.916,
|
248 |
+
"step": 12000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.56,
|
252 |
+
"learning_rate": 0.0006,
|
253 |
+
"loss": 0.6773,
|
254 |
+
"step": 12500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.58,
|
258 |
+
"learning_rate": 0.0005999935478721662,
|
259 |
+
"loss": 0.6774,
|
260 |
+
"step": 13000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.58,
|
264 |
+
"eval_loss": 0.677127480506897,
|
265 |
+
"eval_runtime": 2.1773,
|
266 |
+
"eval_samples_per_second": 1054.986,
|
267 |
+
"eval_steps_per_second": 16.534,
|
268 |
+
"step": 13000
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.6,
|
272 |
+
"learning_rate": 0.000599974191770902,
|
273 |
+
"loss": 0.6773,
|
274 |
+
"step": 13500
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 0.62,
|
278 |
+
"learning_rate": 0.0005999419325429058,
|
279 |
+
"loss": 0.6773,
|
280 |
+
"step": 14000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 0.62,
|
284 |
+
"eval_loss": 0.6771531105041504,
|
285 |
+
"eval_runtime": 2.1173,
|
286 |
+
"eval_samples_per_second": 1084.86,
|
287 |
+
"eval_steps_per_second": 17.003,
|
288 |
+
"step": 14000
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 0.65,
|
292 |
+
"learning_rate": 0.0005998967715993009,
|
293 |
+
"loss": 0.6773,
|
294 |
+
"step": 14500
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 0.67,
|
298 |
+
"learning_rate": 0.0005998387109155732,
|
299 |
+
"loss": 0.6773,
|
300 |
+
"step": 15000
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 0.67,
|
304 |
+
"eval_loss": 0.6771678924560547,
|
305 |
+
"eval_runtime": 2.1919,
|
306 |
+
"eval_samples_per_second": 1047.963,
|
307 |
+
"eval_steps_per_second": 16.424,
|
308 |
+
"step": 15000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.69,
|
312 |
+
"learning_rate": 0.000599767753031485,
|
313 |
+
"loss": 0.6773,
|
314 |
+
"step": 15500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.71,
|
318 |
+
"learning_rate": 0.0005996839010509641,
|
319 |
+
"loss": 0.6772,
|
320 |
+
"step": 16000
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.71,
|
324 |
+
"eval_loss": 0.6776318550109863,
|
325 |
+
"eval_runtime": 2.199,
|
326 |
+
"eval_samples_per_second": 1044.559,
|
327 |
+
"eval_steps_per_second": 16.371,
|
328 |
+
"step": 16000
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 0.74,
|
332 |
+
"learning_rate": 0.0005995871586419678,
|
333 |
+
"loss": 0.6773,
|
334 |
+
"step": 16500
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 0.76,
|
338 |
+
"learning_rate": 0.0005994775300363225,
|
339 |
+
"loss": 0.6773,
|
340 |
+
"step": 17000
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 0.76,
|
344 |
+
"eval_loss": 0.676984429359436,
|
345 |
+
"eval_runtime": 2.1946,
|
346 |
+
"eval_samples_per_second": 1046.652,
|
347 |
+
"eval_steps_per_second": 16.404,
|
348 |
+
"step": 17000
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 0.78,
|
352 |
+
"learning_rate": 0.0005993550200295384,
|
353 |
+
"loss": 0.6772,
|
354 |
+
"step": 17500
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 0.8,
|
358 |
+
"learning_rate": 0.0005992196339806002,
|
359 |
+
"loss": 0.6772,
|
360 |
+
"step": 18000
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 0.8,
|
364 |
+
"eval_loss": 0.6774880290031433,
|
365 |
+
"eval_runtime": 2.1027,
|
366 |
+
"eval_samples_per_second": 1092.415,
|
367 |
+
"eval_steps_per_second": 17.121,
|
368 |
+
"step": 18000
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.83,
|
372 |
+
"learning_rate": 0.0005990713778117324,
|
373 |
+
"loss": 0.6773,
|
374 |
+
"step": 18500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.85,
|
378 |
+
"learning_rate": 0.0005989102580081398,
|
379 |
+
"loss": 0.6772,
|
380 |
+
"step": 19000
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.85,
|
384 |
+
"eval_loss": 0.676984965801239,
|
385 |
+
"eval_runtime": 2.1686,
|
386 |
+
"eval_samples_per_second": 1059.223,
|
387 |
+
"eval_steps_per_second": 16.601,
|
388 |
+
"step": 19000
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 0.87,
|
392 |
+
"learning_rate": 0.0005987362816177249,
|
393 |
+
"loss": 0.6773,
|
394 |
+
"step": 19500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.89,
|
398 |
+
"learning_rate": 0.0005985494562507783,
|
399 |
+
"loss": 0.6774,
|
400 |
+
"step": 20000
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 0.89,
|
404 |
+
"eval_loss": 0.6769698262214661,
|
405 |
+
"eval_runtime": 2.1456,
|
406 |
+
"eval_samples_per_second": 1070.586,
|
407 |
+
"eval_steps_per_second": 16.779,
|
408 |
+
"step": 20000
|
409 |
}
|
410 |
],
|
411 |
"max_steps": 250000,
|
412 |
"num_train_epochs": 12,
|
413 |
+
"total_flos": 3.2033600573160817e+20,
|
414 |
"trial_name": null,
|
415 |
"trial_params": null
|
416 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 25761253
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cab920afe4d207be60e60c397c80474cec3b0781a866712c960668a023bb8b59
|
3 |
size 25761253
|