Training in progress, step 4000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 891558696
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:935c0c9e6c9bad61c25203243a2dced5424fa2c750c3b63d12e6a9555bd1d414
|
3 |
size 891558696
|
last-checkpoint/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1783272762
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5b6424579b540963ca5dff78ef0b1bc07be8ee502bd08a007f4baf7bfc0962f2
|
3 |
size 1783272762
|
last-checkpoint/rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14244
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:994ecdd24463592bb7a469b126032a8d12a6fc0731753b561060d5f56a11ebf4
|
3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:86f6fa01b61cfbdf0a99f96e2c8520500066cb2276b7fc4c693828106716e38a
|
3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./fine-tuned/checkpoint-
|
4 |
-
"epoch": 0.
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
@@ -81,9 +81,9 @@
|
|
81 |
{
|
82 |
"epoch": 0.08,
|
83 |
"eval_loss": 0.09235642850399017,
|
84 |
-
"eval_runtime":
|
85 |
-
"eval_samples_per_second":
|
86 |
-
"eval_steps_per_second": 2.
|
87 |
"step": 500
|
88 |
},
|
89 |
{
|
@@ -159,9 +159,9 @@
|
|
159 |
{
|
160 |
"epoch": 0.16,
|
161 |
"eval_loss": 0.08949962258338928,
|
162 |
-
"eval_runtime":
|
163 |
-
"eval_samples_per_second":
|
164 |
-
"eval_steps_per_second": 2.
|
165 |
"step": 1000
|
166 |
},
|
167 |
{
|
@@ -237,9 +237,9 @@
|
|
237 |
{
|
238 |
"epoch": 0.24,
|
239 |
"eval_loss": 0.08808805048465729,
|
240 |
-
"eval_runtime":
|
241 |
-
"eval_samples_per_second":
|
242 |
-
"eval_steps_per_second": 2.
|
243 |
"step": 1500
|
244 |
},
|
245 |
{
|
@@ -315,9 +315,9 @@
|
|
315 |
{
|
316 |
"epoch": 0.32,
|
317 |
"eval_loss": 0.08704760670661926,
|
318 |
-
"eval_runtime":
|
319 |
-
"eval_samples_per_second":
|
320 |
-
"eval_steps_per_second": 2.
|
321 |
"step": 2000
|
322 |
},
|
323 |
{
|
@@ -393,9 +393,9 @@
|
|
393 |
{
|
394 |
"epoch": 0.4,
|
395 |
"eval_loss": 0.08615937829017639,
|
396 |
-
"eval_runtime":
|
397 |
-
"eval_samples_per_second":
|
398 |
-
"eval_steps_per_second": 2.
|
399 |
"step": 2500
|
400 |
},
|
401 |
{
|
@@ -471,9 +471,9 @@
|
|
471 |
{
|
472 |
"epoch": 0.48,
|
473 |
"eval_loss": 0.08551913499832153,
|
474 |
-
"eval_runtime":
|
475 |
-
"eval_samples_per_second":
|
476 |
-
"eval_steps_per_second": 2.
|
477 |
"step": 3000
|
478 |
},
|
479 |
{
|
@@ -549,10 +549,88 @@
|
|
549 |
{
|
550 |
"epoch": 0.56,
|
551 |
"eval_loss": 0.08540560305118561,
|
552 |
-
"eval_runtime":
|
553 |
-
"eval_samples_per_second":
|
554 |
-
"eval_steps_per_second": 2.
|
555 |
"step": 3500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
556 |
}
|
557 |
],
|
558 |
"logging_steps": 50,
|
@@ -572,7 +650,7 @@
|
|
572 |
"attributes": {}
|
573 |
}
|
574 |
},
|
575 |
-
"total_flos": 1.
|
576 |
"train_batch_size": 8,
|
577 |
"trial_name": null,
|
578 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.08466340601444244,
|
3 |
+
"best_model_checkpoint": "./fine-tuned/checkpoint-4000",
|
4 |
+
"epoch": 0.64,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 4000,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
|
|
81 |
{
|
82 |
"epoch": 0.08,
|
83 |
"eval_loss": 0.09235642850399017,
|
84 |
+
"eval_runtime": 109.274,
|
85 |
+
"eval_samples_per_second": 18.303,
|
86 |
+
"eval_steps_per_second": 2.288,
|
87 |
"step": 500
|
88 |
},
|
89 |
{
|
|
|
159 |
{
|
160 |
"epoch": 0.16,
|
161 |
"eval_loss": 0.08949962258338928,
|
162 |
+
"eval_runtime": 109.2536,
|
163 |
+
"eval_samples_per_second": 18.306,
|
164 |
+
"eval_steps_per_second": 2.288,
|
165 |
"step": 1000
|
166 |
},
|
167 |
{
|
|
|
237 |
{
|
238 |
"epoch": 0.24,
|
239 |
"eval_loss": 0.08808805048465729,
|
240 |
+
"eval_runtime": 109.2355,
|
241 |
+
"eval_samples_per_second": 18.309,
|
242 |
+
"eval_steps_per_second": 2.289,
|
243 |
"step": 1500
|
244 |
},
|
245 |
{
|
|
|
315 |
{
|
316 |
"epoch": 0.32,
|
317 |
"eval_loss": 0.08704760670661926,
|
318 |
+
"eval_runtime": 109.4348,
|
319 |
+
"eval_samples_per_second": 18.276,
|
320 |
+
"eval_steps_per_second": 2.284,
|
321 |
"step": 2000
|
322 |
},
|
323 |
{
|
|
|
393 |
{
|
394 |
"epoch": 0.4,
|
395 |
"eval_loss": 0.08615937829017639,
|
396 |
+
"eval_runtime": 109.2621,
|
397 |
+
"eval_samples_per_second": 18.305,
|
398 |
+
"eval_steps_per_second": 2.288,
|
399 |
"step": 2500
|
400 |
},
|
401 |
{
|
|
|
471 |
{
|
472 |
"epoch": 0.48,
|
473 |
"eval_loss": 0.08551913499832153,
|
474 |
+
"eval_runtime": 109.2626,
|
475 |
+
"eval_samples_per_second": 18.305,
|
476 |
+
"eval_steps_per_second": 2.288,
|
477 |
"step": 3000
|
478 |
},
|
479 |
{
|
|
|
549 |
{
|
550 |
"epoch": 0.56,
|
551 |
"eval_loss": 0.08540560305118561,
|
552 |
+
"eval_runtime": 109.3641,
|
553 |
+
"eval_samples_per_second": 18.288,
|
554 |
+
"eval_steps_per_second": 2.286,
|
555 |
"step": 3500
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.568,
|
559 |
+
"grad_norm": 6408.47265625,
|
560 |
+
"learning_rate": 2.432e-05,
|
561 |
+
"loss": 0.0652,
|
562 |
+
"step": 3550
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.576,
|
566 |
+
"grad_norm": 5537.69287109375,
|
567 |
+
"learning_rate": 2.4240000000000002e-05,
|
568 |
+
"loss": 0.0633,
|
569 |
+
"step": 3600
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.584,
|
573 |
+
"grad_norm": 7664.20703125,
|
574 |
+
"learning_rate": 2.4160000000000002e-05,
|
575 |
+
"loss": 0.0652,
|
576 |
+
"step": 3650
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.592,
|
580 |
+
"grad_norm": 5726.9697265625,
|
581 |
+
"learning_rate": 2.408e-05,
|
582 |
+
"loss": 0.0667,
|
583 |
+
"step": 3700
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.6,
|
587 |
+
"grad_norm": 6898.275390625,
|
588 |
+
"learning_rate": 2.4e-05,
|
589 |
+
"loss": 0.0675,
|
590 |
+
"step": 3750
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.608,
|
594 |
+
"grad_norm": 9309.822265625,
|
595 |
+
"learning_rate": 2.392e-05,
|
596 |
+
"loss": 0.0668,
|
597 |
+
"step": 3800
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.616,
|
601 |
+
"grad_norm": 8566.080078125,
|
602 |
+
"learning_rate": 2.384e-05,
|
603 |
+
"loss": 0.064,
|
604 |
+
"step": 3850
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.624,
|
608 |
+
"grad_norm": 5729.54833984375,
|
609 |
+
"learning_rate": 2.3760000000000003e-05,
|
610 |
+
"loss": 0.0635,
|
611 |
+
"step": 3900
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.632,
|
615 |
+
"grad_norm": 9562.8701171875,
|
616 |
+
"learning_rate": 2.368e-05,
|
617 |
+
"loss": 0.0643,
|
618 |
+
"step": 3950
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.64,
|
622 |
+
"grad_norm": 4704.76025390625,
|
623 |
+
"learning_rate": 2.3599999999999998e-05,
|
624 |
+
"loss": 0.0649,
|
625 |
+
"step": 4000
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.64,
|
629 |
+
"eval_loss": 0.08466340601444244,
|
630 |
+
"eval_runtime": 109.2066,
|
631 |
+
"eval_samples_per_second": 18.314,
|
632 |
+
"eval_steps_per_second": 2.289,
|
633 |
+
"step": 4000
|
634 |
}
|
635 |
],
|
636 |
"logging_steps": 50,
|
|
|
650 |
"attributes": {}
|
651 |
}
|
652 |
},
|
653 |
+
"total_flos": 1.948665249792e+16,
|
654 |
"train_batch_size": 8,
|
655 |
"trial_name": null,
|
656 |
"trial_params": null
|