Training in progress, step 14000, 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 487156538
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a8af821523f7f828a05693158f90c4eb3a0034faa0c2293ad1124f5f93f2750
|
| 3 |
size 487156538
|
last-checkpoint/pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1059459406
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91553bbc60905f9dec7e2b273a7b78b1aaa7bfa9c652a40341722951f2285fa2
|
| 3 |
size 1059459406
|
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 sha256:4523bbe94ce68cf422359680d501e02156c5a468572eaddf29b6fc30a80a5c85
|
| 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 sha256:4775592d06656304b14fa76806a517bea34547605af51b0919af58d9e3ad34f6
|
| 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 sha256:5cc0d600480a1d336ef5ed5d595520ccc7fd9075dda439dbae6adbb69ff279e7
|
| 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:c2f1d1a83fbbd54a97f127d07293b97435077087576259edb01b1d629c65d3ad
|
| 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:bf353f402a8187c44395ca6064b65c6f690bca29a45070f40a2616e51dfc5dd0
|
| 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,
|
|
@@ -4558,6 +4558,356 @@
|
|
| 4558 |
"learning_rate": 0.0004969134125670214,
|
| 4559 |
"loss": 18.7422,
|
| 4560 |
"step": 13000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4561 |
}
|
| 4562 |
],
|
| 4563 |
"logging_steps": 20,
|
|
@@ -4577,7 +4927,7 @@
|
|
| 4577 |
"attributes": {}
|
| 4578 |
}
|
| 4579 |
},
|
| 4580 |
-
"total_flos":
|
| 4581 |
"train_batch_size": 48,
|
| 4582 |
"trial_name": null,
|
| 4583 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.020738405749871125,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 14000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 4558 |
"learning_rate": 0.0004969134125670214,
|
| 4559 |
"loss": 18.7422,
|
| 4560 |
"step": 13000
|
| 4561 |
+
},
|
| 4562 |
+
{
|
| 4563 |
+
"epoch": 0.019286717347380147,
|
| 4564 |
+
"grad_norm": 7.46875,
|
| 4565 |
+
"learning_rate": 0.0004969084736320138,
|
| 4566 |
+
"loss": 18.6768,
|
| 4567 |
+
"step": 13020
|
| 4568 |
+
},
|
| 4569 |
+
{
|
| 4570 |
+
"epoch": 0.019316343641308534,
|
| 4571 |
+
"grad_norm": 9.125,
|
| 4572 |
+
"learning_rate": 0.0004969035346970062,
|
| 4573 |
+
"loss": 18.7947,
|
| 4574 |
+
"step": 13040
|
| 4575 |
+
},
|
| 4576 |
+
{
|
| 4577 |
+
"epoch": 0.01934596993523692,
|
| 4578 |
+
"grad_norm": 7.15625,
|
| 4579 |
+
"learning_rate": 0.0004968985957619986,
|
| 4580 |
+
"loss": 18.8366,
|
| 4581 |
+
"step": 13060
|
| 4582 |
+
},
|
| 4583 |
+
{
|
| 4584 |
+
"epoch": 0.01937559622916531,
|
| 4585 |
+
"grad_norm": 7.15625,
|
| 4586 |
+
"learning_rate": 0.0004968936568269911,
|
| 4587 |
+
"loss": 18.7926,
|
| 4588 |
+
"step": 13080
|
| 4589 |
+
},
|
| 4590 |
+
{
|
| 4591 |
+
"epoch": 0.019405222523093697,
|
| 4592 |
+
"grad_norm": 7.84375,
|
| 4593 |
+
"learning_rate": 0.0004968887178919835,
|
| 4594 |
+
"loss": 18.8219,
|
| 4595 |
+
"step": 13100
|
| 4596 |
+
},
|
| 4597 |
+
{
|
| 4598 |
+
"epoch": 0.019434848817022083,
|
| 4599 |
+
"grad_norm": 8.0625,
|
| 4600 |
+
"learning_rate": 0.0004968837789569759,
|
| 4601 |
+
"loss": 18.7769,
|
| 4602 |
+
"step": 13120
|
| 4603 |
+
},
|
| 4604 |
+
{
|
| 4605 |
+
"epoch": 0.01946447511095047,
|
| 4606 |
+
"grad_norm": 8.375,
|
| 4607 |
+
"learning_rate": 0.0004968788400219684,
|
| 4608 |
+
"loss": 18.7364,
|
| 4609 |
+
"step": 13140
|
| 4610 |
+
},
|
| 4611 |
+
{
|
| 4612 |
+
"epoch": 0.01949410140487886,
|
| 4613 |
+
"grad_norm": 7.625,
|
| 4614 |
+
"learning_rate": 0.0004968739010869608,
|
| 4615 |
+
"loss": 18.7277,
|
| 4616 |
+
"step": 13160
|
| 4617 |
+
},
|
| 4618 |
+
{
|
| 4619 |
+
"epoch": 0.019523727698807246,
|
| 4620 |
+
"grad_norm": 8.0625,
|
| 4621 |
+
"learning_rate": 0.0004968689621519533,
|
| 4622 |
+
"loss": 18.7254,
|
| 4623 |
+
"step": 13180
|
| 4624 |
+
},
|
| 4625 |
+
{
|
| 4626 |
+
"epoch": 0.019553353992735632,
|
| 4627 |
+
"grad_norm": 7.78125,
|
| 4628 |
+
"learning_rate": 0.0004968640232169457,
|
| 4629 |
+
"loss": 18.7818,
|
| 4630 |
+
"step": 13200
|
| 4631 |
+
},
|
| 4632 |
+
{
|
| 4633 |
+
"epoch": 0.01958298028666402,
|
| 4634 |
+
"grad_norm": 8.125,
|
| 4635 |
+
"learning_rate": 0.0004968590842819381,
|
| 4636 |
+
"loss": 18.7875,
|
| 4637 |
+
"step": 13220
|
| 4638 |
+
},
|
| 4639 |
+
{
|
| 4640 |
+
"epoch": 0.01961260658059241,
|
| 4641 |
+
"grad_norm": 8.9375,
|
| 4642 |
+
"learning_rate": 0.0004968541453469306,
|
| 4643 |
+
"loss": 18.738,
|
| 4644 |
+
"step": 13240
|
| 4645 |
+
},
|
| 4646 |
+
{
|
| 4647 |
+
"epoch": 0.019642232874520795,
|
| 4648 |
+
"grad_norm": 7.4375,
|
| 4649 |
+
"learning_rate": 0.000496849206411923,
|
| 4650 |
+
"loss": 18.7297,
|
| 4651 |
+
"step": 13260
|
| 4652 |
+
},
|
| 4653 |
+
{
|
| 4654 |
+
"epoch": 0.01967185916844918,
|
| 4655 |
+
"grad_norm": 9.1875,
|
| 4656 |
+
"learning_rate": 0.0004968442674769154,
|
| 4657 |
+
"loss": 18.7477,
|
| 4658 |
+
"step": 13280
|
| 4659 |
+
},
|
| 4660 |
+
{
|
| 4661 |
+
"epoch": 0.019701485462377568,
|
| 4662 |
+
"grad_norm": 8.4375,
|
| 4663 |
+
"learning_rate": 0.0004968393285419079,
|
| 4664 |
+
"loss": 18.7385,
|
| 4665 |
+
"step": 13300
|
| 4666 |
+
},
|
| 4667 |
+
{
|
| 4668 |
+
"epoch": 0.019731111756305958,
|
| 4669 |
+
"grad_norm": 7.84375,
|
| 4670 |
+
"learning_rate": 0.0004968343896069003,
|
| 4671 |
+
"loss": 18.7936,
|
| 4672 |
+
"step": 13320
|
| 4673 |
+
},
|
| 4674 |
+
{
|
| 4675 |
+
"epoch": 0.019760738050234344,
|
| 4676 |
+
"grad_norm": 9.6875,
|
| 4677 |
+
"learning_rate": 0.0004968294506718928,
|
| 4678 |
+
"loss": 18.7235,
|
| 4679 |
+
"step": 13340
|
| 4680 |
+
},
|
| 4681 |
+
{
|
| 4682 |
+
"epoch": 0.01979036434416273,
|
| 4683 |
+
"grad_norm": 9.5,
|
| 4684 |
+
"learning_rate": 0.0004968245117368852,
|
| 4685 |
+
"loss": 18.8098,
|
| 4686 |
+
"step": 13360
|
| 4687 |
+
},
|
| 4688 |
+
{
|
| 4689 |
+
"epoch": 0.019819990638091117,
|
| 4690 |
+
"grad_norm": 9.5625,
|
| 4691 |
+
"learning_rate": 0.0004968195728018777,
|
| 4692 |
+
"loss": 18.762,
|
| 4693 |
+
"step": 13380
|
| 4694 |
+
},
|
| 4695 |
+
{
|
| 4696 |
+
"epoch": 0.019849616932019507,
|
| 4697 |
+
"grad_norm": 8.4375,
|
| 4698 |
+
"learning_rate": 0.0004968146338668701,
|
| 4699 |
+
"loss": 18.8215,
|
| 4700 |
+
"step": 13400
|
| 4701 |
+
},
|
| 4702 |
+
{
|
| 4703 |
+
"epoch": 0.019879243225947894,
|
| 4704 |
+
"grad_norm": 12.6875,
|
| 4705 |
+
"learning_rate": 0.0004968096949318625,
|
| 4706 |
+
"loss": 18.6931,
|
| 4707 |
+
"step": 13420
|
| 4708 |
+
},
|
| 4709 |
+
{
|
| 4710 |
+
"epoch": 0.01990886951987628,
|
| 4711 |
+
"grad_norm": 7.96875,
|
| 4712 |
+
"learning_rate": 0.0004968047559968548,
|
| 4713 |
+
"loss": 18.7158,
|
| 4714 |
+
"step": 13440
|
| 4715 |
+
},
|
| 4716 |
+
{
|
| 4717 |
+
"epoch": 0.019938495813804667,
|
| 4718 |
+
"grad_norm": 8.4375,
|
| 4719 |
+
"learning_rate": 0.0004967998170618474,
|
| 4720 |
+
"loss": 18.8109,
|
| 4721 |
+
"step": 13460
|
| 4722 |
+
},
|
| 4723 |
+
{
|
| 4724 |
+
"epoch": 0.019968122107733056,
|
| 4725 |
+
"grad_norm": 7.1875,
|
| 4726 |
+
"learning_rate": 0.0004967948781268397,
|
| 4727 |
+
"loss": 18.791,
|
| 4728 |
+
"step": 13480
|
| 4729 |
+
},
|
| 4730 |
+
{
|
| 4731 |
+
"epoch": 0.019997748401661443,
|
| 4732 |
+
"grad_norm": 7.15625,
|
| 4733 |
+
"learning_rate": 0.0004967899391918321,
|
| 4734 |
+
"loss": 18.7768,
|
| 4735 |
+
"step": 13500
|
| 4736 |
+
},
|
| 4737 |
+
{
|
| 4738 |
+
"epoch": 0.02002737469558983,
|
| 4739 |
+
"grad_norm": 7.8125,
|
| 4740 |
+
"learning_rate": 0.0004967850002568246,
|
| 4741 |
+
"loss": 18.7521,
|
| 4742 |
+
"step": 13520
|
| 4743 |
+
},
|
| 4744 |
+
{
|
| 4745 |
+
"epoch": 0.020057000989518216,
|
| 4746 |
+
"grad_norm": 8.25,
|
| 4747 |
+
"learning_rate": 0.000496780061321817,
|
| 4748 |
+
"loss": 18.8413,
|
| 4749 |
+
"step": 13540
|
| 4750 |
+
},
|
| 4751 |
+
{
|
| 4752 |
+
"epoch": 0.020086627283446606,
|
| 4753 |
+
"grad_norm": 24.75,
|
| 4754 |
+
"learning_rate": 0.0004967751223868095,
|
| 4755 |
+
"loss": 18.7564,
|
| 4756 |
+
"step": 13560
|
| 4757 |
+
},
|
| 4758 |
+
{
|
| 4759 |
+
"epoch": 0.020116253577374992,
|
| 4760 |
+
"grad_norm": 9.375,
|
| 4761 |
+
"learning_rate": 0.0004967701834518019,
|
| 4762 |
+
"loss": 18.7654,
|
| 4763 |
+
"step": 13580
|
| 4764 |
+
},
|
| 4765 |
+
{
|
| 4766 |
+
"epoch": 0.02014587987130338,
|
| 4767 |
+
"grad_norm": 23.625,
|
| 4768 |
+
"learning_rate": 0.0004967652445167944,
|
| 4769 |
+
"loss": 18.7145,
|
| 4770 |
+
"step": 13600
|
| 4771 |
+
},
|
| 4772 |
+
{
|
| 4773 |
+
"epoch": 0.020175506165231765,
|
| 4774 |
+
"grad_norm": 6.875,
|
| 4775 |
+
"learning_rate": 0.0004967603055817868,
|
| 4776 |
+
"loss": 18.7534,
|
| 4777 |
+
"step": 13620
|
| 4778 |
+
},
|
| 4779 |
+
{
|
| 4780 |
+
"epoch": 0.020205132459160155,
|
| 4781 |
+
"grad_norm": 8.6875,
|
| 4782 |
+
"learning_rate": 0.0004967553666467792,
|
| 4783 |
+
"loss": 18.7354,
|
| 4784 |
+
"step": 13640
|
| 4785 |
+
},
|
| 4786 |
+
{
|
| 4787 |
+
"epoch": 0.02023475875308854,
|
| 4788 |
+
"grad_norm": 7.21875,
|
| 4789 |
+
"learning_rate": 0.0004967504277117716,
|
| 4790 |
+
"loss": 18.6682,
|
| 4791 |
+
"step": 13660
|
| 4792 |
+
},
|
| 4793 |
+
{
|
| 4794 |
+
"epoch": 0.020264385047016928,
|
| 4795 |
+
"grad_norm": 8.625,
|
| 4796 |
+
"learning_rate": 0.0004967454887767641,
|
| 4797 |
+
"loss": 18.7693,
|
| 4798 |
+
"step": 13680
|
| 4799 |
+
},
|
| 4800 |
+
{
|
| 4801 |
+
"epoch": 0.020294011340945314,
|
| 4802 |
+
"grad_norm": 9.4375,
|
| 4803 |
+
"learning_rate": 0.0004967405498417565,
|
| 4804 |
+
"loss": 18.6937,
|
| 4805 |
+
"step": 13700
|
| 4806 |
+
},
|
| 4807 |
+
{
|
| 4808 |
+
"epoch": 0.020323637634873704,
|
| 4809 |
+
"grad_norm": 8.125,
|
| 4810 |
+
"learning_rate": 0.000496735610906749,
|
| 4811 |
+
"loss": 18.7447,
|
| 4812 |
+
"step": 13720
|
| 4813 |
+
},
|
| 4814 |
+
{
|
| 4815 |
+
"epoch": 0.02035326392880209,
|
| 4816 |
+
"grad_norm": 8.3125,
|
| 4817 |
+
"learning_rate": 0.0004967306719717414,
|
| 4818 |
+
"loss": 18.6705,
|
| 4819 |
+
"step": 13740
|
| 4820 |
+
},
|
| 4821 |
+
{
|
| 4822 |
+
"epoch": 0.020382890222730477,
|
| 4823 |
+
"grad_norm": 8.375,
|
| 4824 |
+
"learning_rate": 0.0004967257330367339,
|
| 4825 |
+
"loss": 18.6713,
|
| 4826 |
+
"step": 13760
|
| 4827 |
+
},
|
| 4828 |
+
{
|
| 4829 |
+
"epoch": 0.020412516516658864,
|
| 4830 |
+
"grad_norm": 9.0,
|
| 4831 |
+
"learning_rate": 0.0004967207941017263,
|
| 4832 |
+
"loss": 18.7607,
|
| 4833 |
+
"step": 13780
|
| 4834 |
+
},
|
| 4835 |
+
{
|
| 4836 |
+
"epoch": 0.020442142810587254,
|
| 4837 |
+
"grad_norm": 9.0625,
|
| 4838 |
+
"learning_rate": 0.0004967158551667187,
|
| 4839 |
+
"loss": 18.6956,
|
| 4840 |
+
"step": 13800
|
| 4841 |
+
},
|
| 4842 |
+
{
|
| 4843 |
+
"epoch": 0.02047176910451564,
|
| 4844 |
+
"grad_norm": 7.90625,
|
| 4845 |
+
"learning_rate": 0.0004967109162317111,
|
| 4846 |
+
"loss": 18.7564,
|
| 4847 |
+
"step": 13820
|
| 4848 |
+
},
|
| 4849 |
+
{
|
| 4850 |
+
"epoch": 0.020501395398444026,
|
| 4851 |
+
"grad_norm": 7.71875,
|
| 4852 |
+
"learning_rate": 0.0004967059772967036,
|
| 4853 |
+
"loss": 18.6956,
|
| 4854 |
+
"step": 13840
|
| 4855 |
+
},
|
| 4856 |
+
{
|
| 4857 |
+
"epoch": 0.020531021692372413,
|
| 4858 |
+
"grad_norm": 8.125,
|
| 4859 |
+
"learning_rate": 0.000496701038361696,
|
| 4860 |
+
"loss": 18.6879,
|
| 4861 |
+
"step": 13860
|
| 4862 |
+
},
|
| 4863 |
+
{
|
| 4864 |
+
"epoch": 0.020560647986300803,
|
| 4865 |
+
"grad_norm": 9.1875,
|
| 4866 |
+
"learning_rate": 0.0004966960994266884,
|
| 4867 |
+
"loss": 18.7128,
|
| 4868 |
+
"step": 13880
|
| 4869 |
+
},
|
| 4870 |
+
{
|
| 4871 |
+
"epoch": 0.02059027428022919,
|
| 4872 |
+
"grad_norm": 9.375,
|
| 4873 |
+
"learning_rate": 0.0004966911604916809,
|
| 4874 |
+
"loss": 18.6706,
|
| 4875 |
+
"step": 13900
|
| 4876 |
+
},
|
| 4877 |
+
{
|
| 4878 |
+
"epoch": 0.020619900574157576,
|
| 4879 |
+
"grad_norm": 8.0,
|
| 4880 |
+
"learning_rate": 0.0004966862215566733,
|
| 4881 |
+
"loss": 18.6614,
|
| 4882 |
+
"step": 13920
|
| 4883 |
+
},
|
| 4884 |
+
{
|
| 4885 |
+
"epoch": 0.020649526868085962,
|
| 4886 |
+
"grad_norm": 7.625,
|
| 4887 |
+
"learning_rate": 0.0004966812826216658,
|
| 4888 |
+
"loss": 18.7468,
|
| 4889 |
+
"step": 13940
|
| 4890 |
+
},
|
| 4891 |
+
{
|
| 4892 |
+
"epoch": 0.020679153162014352,
|
| 4893 |
+
"grad_norm": 7.40625,
|
| 4894 |
+
"learning_rate": 0.0004966763436866582,
|
| 4895 |
+
"loss": 18.7248,
|
| 4896 |
+
"step": 13960
|
| 4897 |
+
},
|
| 4898 |
+
{
|
| 4899 |
+
"epoch": 0.02070877945594274,
|
| 4900 |
+
"grad_norm": 7.4375,
|
| 4901 |
+
"learning_rate": 0.0004966714047516507,
|
| 4902 |
+
"loss": 18.7211,
|
| 4903 |
+
"step": 13980
|
| 4904 |
+
},
|
| 4905 |
+
{
|
| 4906 |
+
"epoch": 0.020738405749871125,
|
| 4907 |
+
"grad_norm": 7.21875,
|
| 4908 |
+
"learning_rate": 0.0004966664658166431,
|
| 4909 |
+
"loss": 18.6651,
|
| 4910 |
+
"step": 14000
|
| 4911 |
}
|
| 4912 |
],
|
| 4913 |
"logging_steps": 20,
|
|
|
|
| 4927 |
"attributes": {}
|
| 4928 |
}
|
| 4929 |
},
|
| 4930 |
+
"total_flos": 3.1157911970490876e+19,
|
| 4931 |
"train_batch_size": 48,
|
| 4932 |
"trial_name": null,
|
| 4933 |
"trial_params": null
|