Training in progress, step 11000, checkpoint
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/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +353 -3
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 244223098
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f8dddf6de75a2669e45bc92f6a4ca08a65509177a3732a367cbfa1c80daacbe
|
| 3 |
size 244223098
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 381944306
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30dee1c1faa1fa9cca0bbdc3497512922f4906f4ab49d60e46fb24c934bb150d
|
| 3 |
size 381944306
|
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 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfe43fa1be8fc23eebf6d0265c9e86d27dbe1a7183ee9ff8d290496f67f7920b
|
| 3 |
size 14960
|
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 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02838e3dd99a981aed96c1e46abb129b6636bb9bdc4bb3b9d32692ead8821881
|
| 3 |
size 14960
|
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 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aed6a0b83373d2ca2f6ea1f1ac78752c4b8eb48d2f34a0bffe9748140ee5f947
|
| 3 |
size 14960
|
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 14960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0de41f811c47a09044e5ad93b32d48fbc2e808eb9859cb07a66f7923677574e
|
| 3 |
size 14960
|
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:f8fd8d6850b7427eafc7ded0e60d1d7d6419f9660dea8de7c7cbb8cd0dbd9818
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -3508,6 +3508,356 @@
|
|
| 3508 |
"learning_rate": 0.0004976542528181563,
|
| 3509 |
"loss": 24.1171,
|
| 3510 |
"step": 10000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3511 |
}
|
| 3512 |
],
|
| 3513 |
"logging_steps": 20,
|
|
@@ -3527,7 +3877,7 @@
|
|
| 3527 |
"attributes": {}
|
| 3528 |
}
|
| 3529 |
},
|
| 3530 |
-
"total_flos":
|
| 3531 |
"train_batch_size": 48,
|
| 3532 |
"trial_name": null,
|
| 3533 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.016294461660613026,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 11000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 3508 |
"learning_rate": 0.0004976542528181563,
|
| 3509 |
"loss": 24.1171,
|
| 3510 |
"step": 10000
|
| 3511 |
+
},
|
| 3512 |
+
{
|
| 3513 |
+
"epoch": 0.014842773258122048,
|
| 3514 |
+
"grad_norm": 19.625,
|
| 3515 |
+
"learning_rate": 0.0004976493138831488,
|
| 3516 |
+
"loss": 24.1736,
|
| 3517 |
+
"step": 10020
|
| 3518 |
+
},
|
| 3519 |
+
{
|
| 3520 |
+
"epoch": 0.014872399552050436,
|
| 3521 |
+
"grad_norm": 18.625,
|
| 3522 |
+
"learning_rate": 0.0004976443749481412,
|
| 3523 |
+
"loss": 24.2905,
|
| 3524 |
+
"step": 10040
|
| 3525 |
+
},
|
| 3526 |
+
{
|
| 3527 |
+
"epoch": 0.014902025845978823,
|
| 3528 |
+
"grad_norm": 18.375,
|
| 3529 |
+
"learning_rate": 0.0004976394360131336,
|
| 3530 |
+
"loss": 24.1324,
|
| 3531 |
+
"step": 10060
|
| 3532 |
+
},
|
| 3533 |
+
{
|
| 3534 |
+
"epoch": 0.014931652139907211,
|
| 3535 |
+
"grad_norm": 20.375,
|
| 3536 |
+
"learning_rate": 0.0004976344970781261,
|
| 3537 |
+
"loss": 24.1785,
|
| 3538 |
+
"step": 10080
|
| 3539 |
+
},
|
| 3540 |
+
{
|
| 3541 |
+
"epoch": 0.014961278433835597,
|
| 3542 |
+
"grad_norm": 17.625,
|
| 3543 |
+
"learning_rate": 0.0004976295581431185,
|
| 3544 |
+
"loss": 24.1872,
|
| 3545 |
+
"step": 10100
|
| 3546 |
+
},
|
| 3547 |
+
{
|
| 3548 |
+
"epoch": 0.014990904727763986,
|
| 3549 |
+
"grad_norm": 18.0,
|
| 3550 |
+
"learning_rate": 0.0004976246192081109,
|
| 3551 |
+
"loss": 24.1813,
|
| 3552 |
+
"step": 10120
|
| 3553 |
+
},
|
| 3554 |
+
{
|
| 3555 |
+
"epoch": 0.015020531021692372,
|
| 3556 |
+
"grad_norm": 21.75,
|
| 3557 |
+
"learning_rate": 0.0004976196802731033,
|
| 3558 |
+
"loss": 24.1438,
|
| 3559 |
+
"step": 10140
|
| 3560 |
+
},
|
| 3561 |
+
{
|
| 3562 |
+
"epoch": 0.01505015731562076,
|
| 3563 |
+
"grad_norm": 22.375,
|
| 3564 |
+
"learning_rate": 0.0004976147413380958,
|
| 3565 |
+
"loss": 24.1436,
|
| 3566 |
+
"step": 10160
|
| 3567 |
+
},
|
| 3568 |
+
{
|
| 3569 |
+
"epoch": 0.015079783609549147,
|
| 3570 |
+
"grad_norm": 19.5,
|
| 3571 |
+
"learning_rate": 0.0004976098024030882,
|
| 3572 |
+
"loss": 24.1394,
|
| 3573 |
+
"step": 10180
|
| 3574 |
+
},
|
| 3575 |
+
{
|
| 3576 |
+
"epoch": 0.015109409903477535,
|
| 3577 |
+
"grad_norm": 18.25,
|
| 3578 |
+
"learning_rate": 0.0004976048634680807,
|
| 3579 |
+
"loss": 24.0992,
|
| 3580 |
+
"step": 10200
|
| 3581 |
+
},
|
| 3582 |
+
{
|
| 3583 |
+
"epoch": 0.015139036197405921,
|
| 3584 |
+
"grad_norm": 15.6875,
|
| 3585 |
+
"learning_rate": 0.0004975999245330731,
|
| 3586 |
+
"loss": 24.0464,
|
| 3587 |
+
"step": 10220
|
| 3588 |
+
},
|
| 3589 |
+
{
|
| 3590 |
+
"epoch": 0.01516866249133431,
|
| 3591 |
+
"grad_norm": 17.875,
|
| 3592 |
+
"learning_rate": 0.0004975949855980656,
|
| 3593 |
+
"loss": 24.0805,
|
| 3594 |
+
"step": 10240
|
| 3595 |
+
},
|
| 3596 |
+
{
|
| 3597 |
+
"epoch": 0.015198288785262696,
|
| 3598 |
+
"grad_norm": 16.875,
|
| 3599 |
+
"learning_rate": 0.000497590046663058,
|
| 3600 |
+
"loss": 24.0985,
|
| 3601 |
+
"step": 10260
|
| 3602 |
+
},
|
| 3603 |
+
{
|
| 3604 |
+
"epoch": 0.015227915079191084,
|
| 3605 |
+
"grad_norm": 23.25,
|
| 3606 |
+
"learning_rate": 0.0004975851077280504,
|
| 3607 |
+
"loss": 24.042,
|
| 3608 |
+
"step": 10280
|
| 3609 |
+
},
|
| 3610 |
+
{
|
| 3611 |
+
"epoch": 0.01525754137311947,
|
| 3612 |
+
"grad_norm": 19.375,
|
| 3613 |
+
"learning_rate": 0.0004975801687930429,
|
| 3614 |
+
"loss": 24.0048,
|
| 3615 |
+
"step": 10300
|
| 3616 |
+
},
|
| 3617 |
+
{
|
| 3618 |
+
"epoch": 0.015287167667047859,
|
| 3619 |
+
"grad_norm": 19.75,
|
| 3620 |
+
"learning_rate": 0.0004975752298580353,
|
| 3621 |
+
"loss": 24.0158,
|
| 3622 |
+
"step": 10320
|
| 3623 |
+
},
|
| 3624 |
+
{
|
| 3625 |
+
"epoch": 0.015316793960976245,
|
| 3626 |
+
"grad_norm": 20.875,
|
| 3627 |
+
"learning_rate": 0.0004975702909230277,
|
| 3628 |
+
"loss": 23.9875,
|
| 3629 |
+
"step": 10340
|
| 3630 |
+
},
|
| 3631 |
+
{
|
| 3632 |
+
"epoch": 0.015346420254904633,
|
| 3633 |
+
"grad_norm": 20.75,
|
| 3634 |
+
"learning_rate": 0.0004975653519880202,
|
| 3635 |
+
"loss": 24.055,
|
| 3636 |
+
"step": 10360
|
| 3637 |
+
},
|
| 3638 |
+
{
|
| 3639 |
+
"epoch": 0.01537604654883302,
|
| 3640 |
+
"grad_norm": 16.125,
|
| 3641 |
+
"learning_rate": 0.0004975604130530126,
|
| 3642 |
+
"loss": 23.9887,
|
| 3643 |
+
"step": 10380
|
| 3644 |
+
},
|
| 3645 |
+
{
|
| 3646 |
+
"epoch": 0.015405672842761408,
|
| 3647 |
+
"grad_norm": 17.25,
|
| 3648 |
+
"learning_rate": 0.000497555474118005,
|
| 3649 |
+
"loss": 24.0268,
|
| 3650 |
+
"step": 10400
|
| 3651 |
+
},
|
| 3652 |
+
{
|
| 3653 |
+
"epoch": 0.015435299136689794,
|
| 3654 |
+
"grad_norm": 18.125,
|
| 3655 |
+
"learning_rate": 0.0004975505351829975,
|
| 3656 |
+
"loss": 24.0453,
|
| 3657 |
+
"step": 10420
|
| 3658 |
+
},
|
| 3659 |
+
{
|
| 3660 |
+
"epoch": 0.015464925430618183,
|
| 3661 |
+
"grad_norm": 21.625,
|
| 3662 |
+
"learning_rate": 0.0004975455962479898,
|
| 3663 |
+
"loss": 24.0189,
|
| 3664 |
+
"step": 10440
|
| 3665 |
+
},
|
| 3666 |
+
{
|
| 3667 |
+
"epoch": 0.015494551724546569,
|
| 3668 |
+
"grad_norm": 15.625,
|
| 3669 |
+
"learning_rate": 0.0004975406573129823,
|
| 3670 |
+
"loss": 23.9409,
|
| 3671 |
+
"step": 10460
|
| 3672 |
+
},
|
| 3673 |
+
{
|
| 3674 |
+
"epoch": 0.015524178018474957,
|
| 3675 |
+
"grad_norm": 18.0,
|
| 3676 |
+
"learning_rate": 0.0004975357183779747,
|
| 3677 |
+
"loss": 23.8996,
|
| 3678 |
+
"step": 10480
|
| 3679 |
+
},
|
| 3680 |
+
{
|
| 3681 |
+
"epoch": 0.015553804312403344,
|
| 3682 |
+
"grad_norm": 21.875,
|
| 3683 |
+
"learning_rate": 0.0004975307794429671,
|
| 3684 |
+
"loss": 24.0183,
|
| 3685 |
+
"step": 10500
|
| 3686 |
+
},
|
| 3687 |
+
{
|
| 3688 |
+
"epoch": 0.015583430606331732,
|
| 3689 |
+
"grad_norm": 19.0,
|
| 3690 |
+
"learning_rate": 0.0004975258405079595,
|
| 3691 |
+
"loss": 23.9392,
|
| 3692 |
+
"step": 10520
|
| 3693 |
+
},
|
| 3694 |
+
{
|
| 3695 |
+
"epoch": 0.015613056900260118,
|
| 3696 |
+
"grad_norm": 16.375,
|
| 3697 |
+
"learning_rate": 0.000497520901572952,
|
| 3698 |
+
"loss": 23.9131,
|
| 3699 |
+
"step": 10540
|
| 3700 |
+
},
|
| 3701 |
+
{
|
| 3702 |
+
"epoch": 0.015642683194188507,
|
| 3703 |
+
"grad_norm": 17.5,
|
| 3704 |
+
"learning_rate": 0.0004975159626379444,
|
| 3705 |
+
"loss": 23.8963,
|
| 3706 |
+
"step": 10560
|
| 3707 |
+
},
|
| 3708 |
+
{
|
| 3709 |
+
"epoch": 0.015672309488116893,
|
| 3710 |
+
"grad_norm": 16.625,
|
| 3711 |
+
"learning_rate": 0.0004975110237029369,
|
| 3712 |
+
"loss": 23.9361,
|
| 3713 |
+
"step": 10580
|
| 3714 |
+
},
|
| 3715 |
+
{
|
| 3716 |
+
"epoch": 0.01570193578204528,
|
| 3717 |
+
"grad_norm": 18.25,
|
| 3718 |
+
"learning_rate": 0.0004975060847679293,
|
| 3719 |
+
"loss": 23.9129,
|
| 3720 |
+
"step": 10600
|
| 3721 |
+
},
|
| 3722 |
+
{
|
| 3723 |
+
"epoch": 0.01573156207597367,
|
| 3724 |
+
"grad_norm": 19.75,
|
| 3725 |
+
"learning_rate": 0.0004975011458329218,
|
| 3726 |
+
"loss": 23.8795,
|
| 3727 |
+
"step": 10620
|
| 3728 |
+
},
|
| 3729 |
+
{
|
| 3730 |
+
"epoch": 0.015761188369902056,
|
| 3731 |
+
"grad_norm": 14.8125,
|
| 3732 |
+
"learning_rate": 0.0004974962068979142,
|
| 3733 |
+
"loss": 23.8412,
|
| 3734 |
+
"step": 10640
|
| 3735 |
+
},
|
| 3736 |
+
{
|
| 3737 |
+
"epoch": 0.015790814663830442,
|
| 3738 |
+
"grad_norm": 19.625,
|
| 3739 |
+
"learning_rate": 0.0004974912679629066,
|
| 3740 |
+
"loss": 23.8545,
|
| 3741 |
+
"step": 10660
|
| 3742 |
+
},
|
| 3743 |
+
{
|
| 3744 |
+
"epoch": 0.01582044095775883,
|
| 3745 |
+
"grad_norm": 17.875,
|
| 3746 |
+
"learning_rate": 0.0004974863290278991,
|
| 3747 |
+
"loss": 23.8848,
|
| 3748 |
+
"step": 10680
|
| 3749 |
+
},
|
| 3750 |
+
{
|
| 3751 |
+
"epoch": 0.01585006725168722,
|
| 3752 |
+
"grad_norm": 18.125,
|
| 3753 |
+
"learning_rate": 0.0004974813900928915,
|
| 3754 |
+
"loss": 23.7463,
|
| 3755 |
+
"step": 10700
|
| 3756 |
+
},
|
| 3757 |
+
{
|
| 3758 |
+
"epoch": 0.015879693545615605,
|
| 3759 |
+
"grad_norm": 17.25,
|
| 3760 |
+
"learning_rate": 0.0004974764511578839,
|
| 3761 |
+
"loss": 23.8657,
|
| 3762 |
+
"step": 10720
|
| 3763 |
+
},
|
| 3764 |
+
{
|
| 3765 |
+
"epoch": 0.01590931983954399,
|
| 3766 |
+
"grad_norm": 17.875,
|
| 3767 |
+
"learning_rate": 0.0004974715122228763,
|
| 3768 |
+
"loss": 23.7865,
|
| 3769 |
+
"step": 10740
|
| 3770 |
+
},
|
| 3771 |
+
{
|
| 3772 |
+
"epoch": 0.015938946133472378,
|
| 3773 |
+
"grad_norm": 18.875,
|
| 3774 |
+
"learning_rate": 0.0004974665732878688,
|
| 3775 |
+
"loss": 23.7971,
|
| 3776 |
+
"step": 10760
|
| 3777 |
+
},
|
| 3778 |
+
{
|
| 3779 |
+
"epoch": 0.015968572427400768,
|
| 3780 |
+
"grad_norm": 19.125,
|
| 3781 |
+
"learning_rate": 0.0004974616343528612,
|
| 3782 |
+
"loss": 23.8342,
|
| 3783 |
+
"step": 10780
|
| 3784 |
+
},
|
| 3785 |
+
{
|
| 3786 |
+
"epoch": 0.015998198721329154,
|
| 3787 |
+
"grad_norm": 16.75,
|
| 3788 |
+
"learning_rate": 0.0004974566954178537,
|
| 3789 |
+
"loss": 23.7571,
|
| 3790 |
+
"step": 10800
|
| 3791 |
+
},
|
| 3792 |
+
{
|
| 3793 |
+
"epoch": 0.01602782501525754,
|
| 3794 |
+
"grad_norm": 16.75,
|
| 3795 |
+
"learning_rate": 0.0004974517564828461,
|
| 3796 |
+
"loss": 23.8034,
|
| 3797 |
+
"step": 10820
|
| 3798 |
+
},
|
| 3799 |
+
{
|
| 3800 |
+
"epoch": 0.016057451309185927,
|
| 3801 |
+
"grad_norm": 17.25,
|
| 3802 |
+
"learning_rate": 0.0004974468175478386,
|
| 3803 |
+
"loss": 23.7763,
|
| 3804 |
+
"step": 10840
|
| 3805 |
+
},
|
| 3806 |
+
{
|
| 3807 |
+
"epoch": 0.016087077603114317,
|
| 3808 |
+
"grad_norm": 18.25,
|
| 3809 |
+
"learning_rate": 0.000497441878612831,
|
| 3810 |
+
"loss": 23.803,
|
| 3811 |
+
"step": 10860
|
| 3812 |
+
},
|
| 3813 |
+
{
|
| 3814 |
+
"epoch": 0.016116703897042704,
|
| 3815 |
+
"grad_norm": 20.875,
|
| 3816 |
+
"learning_rate": 0.0004974369396778234,
|
| 3817 |
+
"loss": 23.7222,
|
| 3818 |
+
"step": 10880
|
| 3819 |
+
},
|
| 3820 |
+
{
|
| 3821 |
+
"epoch": 0.01614633019097109,
|
| 3822 |
+
"grad_norm": 20.5,
|
| 3823 |
+
"learning_rate": 0.0004974320007428159,
|
| 3824 |
+
"loss": 23.6994,
|
| 3825 |
+
"step": 10900
|
| 3826 |
+
},
|
| 3827 |
+
{
|
| 3828 |
+
"epoch": 0.016175956484899476,
|
| 3829 |
+
"grad_norm": 15.3125,
|
| 3830 |
+
"learning_rate": 0.0004974270618078083,
|
| 3831 |
+
"loss": 23.6471,
|
| 3832 |
+
"step": 10920
|
| 3833 |
+
},
|
| 3834 |
+
{
|
| 3835 |
+
"epoch": 0.016205582778827866,
|
| 3836 |
+
"grad_norm": 15.5,
|
| 3837 |
+
"learning_rate": 0.0004974221228728007,
|
| 3838 |
+
"loss": 23.7271,
|
| 3839 |
+
"step": 10940
|
| 3840 |
+
},
|
| 3841 |
+
{
|
| 3842 |
+
"epoch": 0.016235209072756253,
|
| 3843 |
+
"grad_norm": 17.5,
|
| 3844 |
+
"learning_rate": 0.0004974171839377932,
|
| 3845 |
+
"loss": 23.6869,
|
| 3846 |
+
"step": 10960
|
| 3847 |
+
},
|
| 3848 |
+
{
|
| 3849 |
+
"epoch": 0.01626483536668464,
|
| 3850 |
+
"grad_norm": 16.75,
|
| 3851 |
+
"learning_rate": 0.0004974122450027856,
|
| 3852 |
+
"loss": 23.6976,
|
| 3853 |
+
"step": 10980
|
| 3854 |
+
},
|
| 3855 |
+
{
|
| 3856 |
+
"epoch": 0.016294461660613026,
|
| 3857 |
+
"grad_norm": 19.0,
|
| 3858 |
+
"learning_rate": 0.0004974073060677781,
|
| 3859 |
+
"loss": 23.6657,
|
| 3860 |
+
"step": 11000
|
| 3861 |
}
|
| 3862 |
],
|
| 3863 |
"logging_steps": 20,
|
|
|
|
| 3877 |
"attributes": {}
|
| 3878 |
}
|
| 3879 |
},
|
| 3880 |
+
"total_flos": 7.567919072411648e+18,
|
| 3881 |
"train_batch_size": 48,
|
| 3882 |
"trial_name": null,
|
| 3883 |
"trial_params": null
|