Commit
·
a66b2ca
1
Parent(s):
a60d228
Training in progress, step 56000
Browse files- adapter_model.bin +1 -1
- {checkpoint-52000 → checkpoint-55000/adapter_model}/README.md +0 -0
- {checkpoint-52000 → checkpoint-55000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-52000 → checkpoint-55000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-52000/adapter_model → checkpoint-56000}/README.md +0 -0
- {checkpoint-52000/adapter_model → checkpoint-56000}/adapter_config.json +0 -0
- {checkpoint-52000/adapter_model → checkpoint-56000}/adapter_model.bin +1 -1
- {checkpoint-52000 → checkpoint-56000}/optimizer.pt +1 -1
- {checkpoint-52000 → checkpoint-56000}/rng_state.pth +1 -1
- {checkpoint-52000 → checkpoint-56000}/scheduler.pt +1 -1
- {checkpoint-52000 → checkpoint-56000}/trainer_state.json +277 -5
- {checkpoint-52000 → checkpoint-56000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d61b2ab661f17f1b28e07a6ea4c559efd2487b69440f512fbda902147b2007f
|
| 3 |
size 500897101
|
{checkpoint-52000 → checkpoint-55000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-52000 → checkpoint-55000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-52000 → checkpoint-55000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f8b0f0db95ea8131359795995cdc710c82f988052688d4cb6fe2ddae5cacce5
|
| 3 |
size 500897101
|
{checkpoint-52000/adapter_model → checkpoint-56000}/README.md
RENAMED
|
File without changes
|
{checkpoint-52000/adapter_model → checkpoint-56000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-52000/adapter_model → checkpoint-56000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d61b2ab661f17f1b28e07a6ea4c559efd2487b69440f512fbda902147b2007f
|
| 3 |
size 500897101
|
{checkpoint-52000 → checkpoint-56000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0d0299416431a6687f29eb725bd8536e5bc5512ff27981755266d125bd960dc
|
| 3 |
size 1001723453
|
{checkpoint-52000 → checkpoint-56000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d223168e1cf1a8cbe3b356c3a0cf2b7c1b147eab84d53ec37dea66d1618867f6
|
| 3 |
size 14575
|
{checkpoint-52000 → checkpoint-56000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40f07a3bf2b8b7e85bd7ec32b459bd8eba34e3ffd70129884ee8cac79708a84f
|
| 3 |
size 627
|
{checkpoint-52000 → checkpoint-56000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -3542,11 +3542,283 @@
|
|
| 3542 |
"eval_samples_per_second": 0.443,
|
| 3543 |
"eval_steps_per_second": 0.443,
|
| 3544 |
"step": 52000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3545 |
}
|
| 3546 |
],
|
| 3547 |
"max_steps": 80463,
|
| 3548 |
"num_train_epochs": 3,
|
| 3549 |
-
"total_flos": 1.
|
| 3550 |
"trial_name": null,
|
| 3551 |
"trial_params": null
|
| 3552 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.49594032764434814,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-56000",
|
| 4 |
+
"epoch": 2.087916185078856,
|
| 5 |
+
"global_step": 56000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 3542 |
"eval_samples_per_second": 0.443,
|
| 3543 |
"eval_steps_per_second": 0.443,
|
| 3544 |
"step": 52000
|
| 3545 |
+
},
|
| 3546 |
+
{
|
| 3547 |
+
"epoch": 1.94,
|
| 3548 |
+
"learning_rate": 5.531372663056664e-05,
|
| 3549 |
+
"loss": 0.4523,
|
| 3550 |
+
"step": 52100
|
| 3551 |
+
},
|
| 3552 |
+
{
|
| 3553 |
+
"epoch": 1.95,
|
| 3554 |
+
"learning_rate": 5.496473670323052e-05,
|
| 3555 |
+
"loss": 0.4434,
|
| 3556 |
+
"step": 52200
|
| 3557 |
+
},
|
| 3558 |
+
{
|
| 3559 |
+
"epoch": 1.95,
|
| 3560 |
+
"learning_rate": 5.461643347647335e-05,
|
| 3561 |
+
"loss": 0.4239,
|
| 3562 |
+
"step": 52300
|
| 3563 |
+
},
|
| 3564 |
+
{
|
| 3565 |
+
"epoch": 1.95,
|
| 3566 |
+
"learning_rate": 5.4268822261245023e-05,
|
| 3567 |
+
"loss": 0.4581,
|
| 3568 |
+
"step": 52400
|
| 3569 |
+
},
|
| 3570 |
+
{
|
| 3571 |
+
"epoch": 1.96,
|
| 3572 |
+
"learning_rate": 5.392190835794369e-05,
|
| 3573 |
+
"loss": 0.4297,
|
| 3574 |
+
"step": 52500
|
| 3575 |
+
},
|
| 3576 |
+
{
|
| 3577 |
+
"epoch": 1.96,
|
| 3578 |
+
"learning_rate": 5.357569705633465e-05,
|
| 3579 |
+
"loss": 0.4431,
|
| 3580 |
+
"step": 52600
|
| 3581 |
+
},
|
| 3582 |
+
{
|
| 3583 |
+
"epoch": 1.96,
|
| 3584 |
+
"learning_rate": 5.3230193635470136e-05,
|
| 3585 |
+
"loss": 0.4406,
|
| 3586 |
+
"step": 52700
|
| 3587 |
+
},
|
| 3588 |
+
{
|
| 3589 |
+
"epoch": 1.97,
|
| 3590 |
+
"learning_rate": 5.288540336360836e-05,
|
| 3591 |
+
"loss": 0.4734,
|
| 3592 |
+
"step": 52800
|
| 3593 |
+
},
|
| 3594 |
+
{
|
| 3595 |
+
"epoch": 1.97,
|
| 3596 |
+
"learning_rate": 5.254133149813349e-05,
|
| 3597 |
+
"loss": 0.4449,
|
| 3598 |
+
"step": 52900
|
| 3599 |
+
},
|
| 3600 |
+
{
|
| 3601 |
+
"epoch": 1.98,
|
| 3602 |
+
"learning_rate": 5.2197983285475315e-05,
|
| 3603 |
+
"loss": 0.4519,
|
| 3604 |
+
"step": 53000
|
| 3605 |
+
},
|
| 3606 |
+
{
|
| 3607 |
+
"epoch": 1.98,
|
| 3608 |
+
"eval_loss": 0.4982399046421051,
|
| 3609 |
+
"eval_runtime": 1234.2642,
|
| 3610 |
+
"eval_samples_per_second": 0.439,
|
| 3611 |
+
"eval_steps_per_second": 0.439,
|
| 3612 |
+
"step": 53000
|
| 3613 |
+
},
|
| 3614 |
+
{
|
| 3615 |
+
"epoch": 1.98,
|
| 3616 |
+
"learning_rate": 5.185536396102946e-05,
|
| 3617 |
+
"loss": 0.4352,
|
| 3618 |
+
"step": 53100
|
| 3619 |
+
},
|
| 3620 |
+
{
|
| 3621 |
+
"epoch": 1.98,
|
| 3622 |
+
"learning_rate": 5.1513478749077274e-05,
|
| 3623 |
+
"loss": 0.4801,
|
| 3624 |
+
"step": 53200
|
| 3625 |
+
},
|
| 3626 |
+
{
|
| 3627 |
+
"epoch": 1.99,
|
| 3628 |
+
"learning_rate": 5.1172332862706376e-05,
|
| 3629 |
+
"loss": 0.4315,
|
| 3630 |
+
"step": 53300
|
| 3631 |
+
},
|
| 3632 |
+
{
|
| 3633 |
+
"epoch": 1.99,
|
| 3634 |
+
"learning_rate": 5.0831931503731065e-05,
|
| 3635 |
+
"loss": 0.4443,
|
| 3636 |
+
"step": 53400
|
| 3637 |
+
},
|
| 3638 |
+
{
|
| 3639 |
+
"epoch": 1.99,
|
| 3640 |
+
"learning_rate": 5.049227986261302e-05,
|
| 3641 |
+
"loss": 0.4597,
|
| 3642 |
+
"step": 53500
|
| 3643 |
+
},
|
| 3644 |
+
{
|
| 3645 |
+
"epoch": 2.0,
|
| 3646 |
+
"learning_rate": 5.0153383118382355e-05,
|
| 3647 |
+
"loss": 0.4519,
|
| 3648 |
+
"step": 53600
|
| 3649 |
+
},
|
| 3650 |
+
{
|
| 3651 |
+
"epoch": 2.0,
|
| 3652 |
+
"learning_rate": 4.9815246438558264e-05,
|
| 3653 |
+
"loss": 0.4314,
|
| 3654 |
+
"step": 53700
|
| 3655 |
+
},
|
| 3656 |
+
{
|
| 3657 |
+
"epoch": 2.01,
|
| 3658 |
+
"learning_rate": 4.9477874979070474e-05,
|
| 3659 |
+
"loss": 0.3999,
|
| 3660 |
+
"step": 53800
|
| 3661 |
+
},
|
| 3662 |
+
{
|
| 3663 |
+
"epoch": 2.01,
|
| 3664 |
+
"learning_rate": 4.914127388418062e-05,
|
| 3665 |
+
"loss": 0.3937,
|
| 3666 |
+
"step": 53900
|
| 3667 |
+
},
|
| 3668 |
+
{
|
| 3669 |
+
"epoch": 2.01,
|
| 3670 |
+
"learning_rate": 4.880544828640372e-05,
|
| 3671 |
+
"loss": 0.3597,
|
| 3672 |
+
"step": 54000
|
| 3673 |
+
},
|
| 3674 |
+
{
|
| 3675 |
+
"epoch": 2.01,
|
| 3676 |
+
"eval_loss": 0.49871668219566345,
|
| 3677 |
+
"eval_runtime": 1230.6045,
|
| 3678 |
+
"eval_samples_per_second": 0.44,
|
| 3679 |
+
"eval_steps_per_second": 0.44,
|
| 3680 |
+
"step": 54000
|
| 3681 |
+
},
|
| 3682 |
+
{
|
| 3683 |
+
"epoch": 2.02,
|
| 3684 |
+
"learning_rate": 4.8470403306430056e-05,
|
| 3685 |
+
"loss": 0.4075,
|
| 3686 |
+
"step": 54100
|
| 3687 |
+
},
|
| 3688 |
+
{
|
| 3689 |
+
"epoch": 2.02,
|
| 3690 |
+
"learning_rate": 4.813614405304693e-05,
|
| 3691 |
+
"loss": 0.383,
|
| 3692 |
+
"step": 54200
|
| 3693 |
+
},
|
| 3694 |
+
{
|
| 3695 |
+
"epoch": 2.02,
|
| 3696 |
+
"learning_rate": 4.780267562306081e-05,
|
| 3697 |
+
"loss": 0.3833,
|
| 3698 |
+
"step": 54300
|
| 3699 |
+
},
|
| 3700 |
+
{
|
| 3701 |
+
"epoch": 2.03,
|
| 3702 |
+
"learning_rate": 4.7470003101219664e-05,
|
| 3703 |
+
"loss": 0.3828,
|
| 3704 |
+
"step": 54400
|
| 3705 |
+
},
|
| 3706 |
+
{
|
| 3707 |
+
"epoch": 2.03,
|
| 3708 |
+
"learning_rate": 4.713813156013548e-05,
|
| 3709 |
+
"loss": 0.3864,
|
| 3710 |
+
"step": 54500
|
| 3711 |
+
},
|
| 3712 |
+
{
|
| 3713 |
+
"epoch": 2.04,
|
| 3714 |
+
"learning_rate": 4.680706606020668e-05,
|
| 3715 |
+
"loss": 0.3805,
|
| 3716 |
+
"step": 54600
|
| 3717 |
+
},
|
| 3718 |
+
{
|
| 3719 |
+
"epoch": 2.04,
|
| 3720 |
+
"learning_rate": 4.64768116495413e-05,
|
| 3721 |
+
"loss": 0.3934,
|
| 3722 |
+
"step": 54700
|
| 3723 |
+
},
|
| 3724 |
+
{
|
| 3725 |
+
"epoch": 2.04,
|
| 3726 |
+
"learning_rate": 4.61473733638797e-05,
|
| 3727 |
+
"loss": 0.3842,
|
| 3728 |
+
"step": 54800
|
| 3729 |
+
},
|
| 3730 |
+
{
|
| 3731 |
+
"epoch": 2.05,
|
| 3732 |
+
"learning_rate": 4.5818756226517924e-05,
|
| 3733 |
+
"loss": 0.3783,
|
| 3734 |
+
"step": 54900
|
| 3735 |
+
},
|
| 3736 |
+
{
|
| 3737 |
+
"epoch": 2.05,
|
| 3738 |
+
"learning_rate": 4.54909652482312e-05,
|
| 3739 |
+
"loss": 0.3912,
|
| 3740 |
+
"step": 55000
|
| 3741 |
+
},
|
| 3742 |
+
{
|
| 3743 |
+
"epoch": 2.05,
|
| 3744 |
+
"eval_loss": 0.49695634841918945,
|
| 3745 |
+
"eval_runtime": 1220.9853,
|
| 3746 |
+
"eval_samples_per_second": 0.444,
|
| 3747 |
+
"eval_steps_per_second": 0.444,
|
| 3748 |
+
"step": 55000
|
| 3749 |
+
},
|
| 3750 |
+
{
|
| 3751 |
+
"epoch": 2.05,
|
| 3752 |
+
"learning_rate": 4.516400542719733e-05,
|
| 3753 |
+
"loss": 0.3941,
|
| 3754 |
+
"step": 55100
|
| 3755 |
+
},
|
| 3756 |
+
{
|
| 3757 |
+
"epoch": 2.06,
|
| 3758 |
+
"learning_rate": 4.4837881748920594e-05,
|
| 3759 |
+
"loss": 0.366,
|
| 3760 |
+
"step": 55200
|
| 3761 |
+
},
|
| 3762 |
+
{
|
| 3763 |
+
"epoch": 2.06,
|
| 3764 |
+
"learning_rate": 4.451259918615569e-05,
|
| 3765 |
+
"loss": 0.4203,
|
| 3766 |
+
"step": 55300
|
| 3767 |
+
},
|
| 3768 |
+
{
|
| 3769 |
+
"epoch": 2.07,
|
| 3770 |
+
"learning_rate": 4.418816269883204e-05,
|
| 3771 |
+
"loss": 0.3613,
|
| 3772 |
+
"step": 55400
|
| 3773 |
+
},
|
| 3774 |
+
{
|
| 3775 |
+
"epoch": 2.07,
|
| 3776 |
+
"learning_rate": 4.386457723397794e-05,
|
| 3777 |
+
"loss": 0.3825,
|
| 3778 |
+
"step": 55500
|
| 3779 |
+
},
|
| 3780 |
+
{
|
| 3781 |
+
"epoch": 2.07,
|
| 3782 |
+
"learning_rate": 4.354184772564526e-05,
|
| 3783 |
+
"loss": 0.4147,
|
| 3784 |
+
"step": 55600
|
| 3785 |
+
},
|
| 3786 |
+
{
|
| 3787 |
+
"epoch": 2.08,
|
| 3788 |
+
"learning_rate": 4.3219979094834275e-05,
|
| 3789 |
+
"loss": 0.3812,
|
| 3790 |
+
"step": 55700
|
| 3791 |
+
},
|
| 3792 |
+
{
|
| 3793 |
+
"epoch": 2.08,
|
| 3794 |
+
"learning_rate": 4.289897624941841e-05,
|
| 3795 |
+
"loss": 0.3926,
|
| 3796 |
+
"step": 55800
|
| 3797 |
+
},
|
| 3798 |
+
{
|
| 3799 |
+
"epoch": 2.08,
|
| 3800 |
+
"learning_rate": 4.257884408406968e-05,
|
| 3801 |
+
"loss": 0.4103,
|
| 3802 |
+
"step": 55900
|
| 3803 |
+
},
|
| 3804 |
+
{
|
| 3805 |
+
"epoch": 2.09,
|
| 3806 |
+
"learning_rate": 4.225958748018381e-05,
|
| 3807 |
+
"loss": 0.4009,
|
| 3808 |
+
"step": 56000
|
| 3809 |
+
},
|
| 3810 |
+
{
|
| 3811 |
+
"epoch": 2.09,
|
| 3812 |
+
"eval_loss": 0.49594032764434814,
|
| 3813 |
+
"eval_runtime": 1218.3341,
|
| 3814 |
+
"eval_samples_per_second": 0.445,
|
| 3815 |
+
"eval_steps_per_second": 0.445,
|
| 3816 |
+
"step": 56000
|
| 3817 |
}
|
| 3818 |
],
|
| 3819 |
"max_steps": 80463,
|
| 3820 |
"num_train_epochs": 3,
|
| 3821 |
+
"total_flos": 1.5701320307350487e+19,
|
| 3822 |
"trial_name": null,
|
| 3823 |
"trial_params": null
|
| 3824 |
}
|
{checkpoint-52000 → checkpoint-56000}/training_args.bin
RENAMED
|
File without changes
|