Commit
•
28a3ccf
1
Parent(s):
b4c3898
huggingartists
Browse files- README.md +3 -3
- config.json +1 -1
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +2 -2
- scheduler.pt +1 -1
- tokenizer_config.json +1 -1
- trainer_state.json +635 -7
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
| 45 |
dataset = load_dataset("huggingartists/andre-3000")
|
| 46 |
```
|
| 47 |
|
| 48 |
-
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/
|
| 49 |
|
| 50 |
## Training procedure
|
| 51 |
|
| 52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on André 3000's lyrics.
|
| 53 |
|
| 54 |
-
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/
|
| 55 |
|
| 56 |
-
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/
|
| 57 |
|
| 58 |
## How to use
|
| 59 |
|
|
|
|
| 45 |
dataset = load_dataset("huggingartists/andre-3000")
|
| 46 |
```
|
| 47 |
|
| 48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/2aj9iybn/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
|
| 49 |
|
| 50 |
## Training procedure
|
| 51 |
|
| 52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on André 3000's lyrics.
|
| 53 |
|
| 54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3lbg2cit) for full transparency and reproducibility.
|
| 55 |
|
| 56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3lbg2cit/artifacts) is logged and versioned.
|
| 57 |
|
| 58 |
## How to use
|
| 59 |
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2LMHeadModel"
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "huggingartists/andre-3000",
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2LMHeadModel"
|
evaluation.txt
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"eval_loss":
|
|
|
|
| 1 |
+
{"eval_loss": 2.7857184410095215, "eval_runtime": 3.2483, "eval_samples_per_second": 43.715, "eval_steps_per_second": 5.541, "epoch": 105.0}
|
flax_model.msgpack
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 497764120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ca2d40ae75a2d9347d2b27e73aece5aa55fffa7ae760d8a75b258044ec799c4
|
| 3 |
size 497764120
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 995604017
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f6ec27a8db3a3d8e12f1403c8e1ddd7d9d8f7c369361110364668a7d3905bcc
|
| 3 |
size 995604017
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 510403817
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:644cac333a83af4607d4d90810953f70e30bdc7159f4a9dc40f8b242c88c2770
|
| 3 |
size 510403817
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16e34fba951f86dbe52880106da1b406b7bf5d467c83625264b1d27faf3245c0
|
| 3 |
+
size 14503
|
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 sha256:4215fb2fd58e7d9127b8dee8649641af0662fd1b11cda980f7e7dc26e6aa301f
|
| 3 |
size 623
|
tokenizer_config.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "
|
|
|
|
| 1 |
+
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/andre-3000", "tokenizer_class": "GPT2Tokenizer"}
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "output/andre-3000/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -646,11 +646,639 @@
|
|
| 646 |
"eval_samples_per_second": 43.288,
|
| 647 |
"eval_steps_per_second": 5.492,
|
| 648 |
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 649 |
}
|
| 650 |
],
|
| 651 |
-
"max_steps":
|
| 652 |
-
"num_train_epochs":
|
| 653 |
-
"total_flos":
|
| 654 |
"trial_name": null,
|
| 655 |
"trial_params": null
|
| 656 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.7857184410095215,
|
| 3 |
+
"best_model_checkpoint": "output/andre-3000/checkpoint-990",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 990,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 646 |
"eval_samples_per_second": 43.288,
|
| 647 |
"eval_steps_per_second": 5.492,
|
| 648 |
"step": 500
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 5.1,
|
| 652 |
+
"learning_rate": 3.4251213229711433e-06,
|
| 653 |
+
"loss": 2.7141,
|
| 654 |
+
"step": 505
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 5.15,
|
| 658 |
+
"learning_rate": 7.625888222272204e-06,
|
| 659 |
+
"loss": 3.1031,
|
| 660 |
+
"step": 510
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 5.2,
|
| 664 |
+
"learning_rate": 1.3358460333369401e-05,
|
| 665 |
+
"loss": 2.7596,
|
| 666 |
+
"step": 515
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 5.25,
|
| 670 |
+
"learning_rate": 2.0478822703346315e-05,
|
| 671 |
+
"loss": 2.747,
|
| 672 |
+
"step": 520
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 5.3,
|
| 676 |
+
"learning_rate": 2.880809600341567e-05,
|
| 677 |
+
"loss": 2.8618,
|
| 678 |
+
"step": 525
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 5.35,
|
| 682 |
+
"learning_rate": 3.8137030375243876e-05,
|
| 683 |
+
"loss": 2.811,
|
| 684 |
+
"step": 530
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 5.4,
|
| 688 |
+
"learning_rate": 4.823126225248023e-05,
|
| 689 |
+
"loss": 2.8491,
|
| 690 |
+
"step": 535
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 5.45,
|
| 694 |
+
"learning_rate": 5.883720209445242e-05,
|
| 695 |
+
"loss": 2.5856,
|
| 696 |
+
"step": 540
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 5.51,
|
| 700 |
+
"learning_rate": 6.968840511906774e-05,
|
| 701 |
+
"loss": 2.9258,
|
| 702 |
+
"step": 545
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 5.56,
|
| 706 |
+
"learning_rate": 8.051226498795124e-05,
|
| 707 |
+
"loss": 2.9732,
|
| 708 |
+
"step": 550
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 5.61,
|
| 712 |
+
"learning_rate": 9.10368622835751e-05,
|
| 713 |
+
"loss": 2.9297,
|
| 714 |
+
"step": 555
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 5.66,
|
| 718 |
+
"learning_rate": 0.0001009977957294059,
|
| 719 |
+
"loss": 2.5333,
|
| 720 |
+
"step": 560
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 5.71,
|
| 724 |
+
"learning_rate": 0.00011014482453764393,
|
| 725 |
+
"loss": 2.7163,
|
| 726 |
+
"step": 565
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 5.76,
|
| 730 |
+
"learning_rate": 0.00011824815501400774,
|
| 731 |
+
"loss": 2.575,
|
| 732 |
+
"step": 570
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 5.81,
|
| 736 |
+
"learning_rate": 0.00012510421348608654,
|
| 737 |
+
"loss": 2.6619,
|
| 738 |
+
"step": 575
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 5.86,
|
| 742 |
+
"learning_rate": 0.0001305407605264622,
|
| 743 |
+
"loss": 2.8522,
|
| 744 |
+
"step": 580
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 5.91,
|
| 748 |
+
"learning_rate": 0.00013442121798995453,
|
| 749 |
+
"loss": 2.9609,
|
| 750 |
+
"step": 585
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 5.96,
|
| 754 |
+
"learning_rate": 0.00013664810016019257,
|
| 755 |
+
"loss": 2.7212,
|
| 756 |
+
"step": 590
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 6.0,
|
| 760 |
+
"eval_loss": 2.8283281326293945,
|
| 761 |
+
"eval_runtime": 3.2672,
|
| 762 |
+
"eval_samples_per_second": 43.462,
|
| 763 |
+
"eval_steps_per_second": 5.509,
|
| 764 |
+
"step": 594
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 6.01,
|
| 768 |
+
"learning_rate": 0.0001371654628074865,
|
| 769 |
+
"loss": 2.8262,
|
| 770 |
+
"step": 595
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 6.06,
|
| 774 |
+
"learning_rate": 0.00013596030863222166,
|
| 775 |
+
"loss": 2.45,
|
| 776 |
+
"step": 600
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 6.11,
|
| 780 |
+
"learning_rate": 0.00013306291378591335,
|
| 781 |
+
"loss": 2.7517,
|
| 782 |
+
"step": 605
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 6.16,
|
| 786 |
+
"learning_rate": 0.00012854606726698723,
|
| 787 |
+
"loss": 2.4939,
|
| 788 |
+
"step": 610
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 6.21,
|
| 792 |
+
"learning_rate": 0.00012252324229935523,
|
| 793 |
+
"loss": 2.6832,
|
| 794 |
+
"step": 615
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 6.26,
|
| 798 |
+
"learning_rate": 0.0001151457456328192,
|
| 799 |
+
"loss": 2.5246,
|
| 800 |
+
"step": 620
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 6.31,
|
| 804 |
+
"learning_rate": 0.00010659891638121517,
|
| 805 |
+
"loss": 2.8753,
|
| 806 |
+
"step": 625
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 6.36,
|
| 810 |
+
"learning_rate": 9.70974698919295e-05,
|
| 811 |
+
"loss": 2.8246,
|
| 812 |
+
"step": 630
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 6.41,
|
| 816 |
+
"learning_rate": 8.688010361913636e-05,
|
| 817 |
+
"loss": 2.5406,
|
| 818 |
+
"step": 635
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 6.46,
|
| 822 |
+
"learning_rate": 7.620350051320941e-05,
|
| 823 |
+
"loss": 2.3994,
|
| 824 |
+
"step": 640
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 6.52,
|
| 828 |
+
"learning_rate": 6.53358805744912e-05,
|
| 829 |
+
"loss": 2.4749,
|
| 830 |
+
"step": 645
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 6.57,
|
| 834 |
+
"learning_rate": 5.4550262570727935e-05,
|
| 835 |
+
"loss": 2.2935,
|
| 836 |
+
"step": 650
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 6.62,
|
| 840 |
+
"learning_rate": 4.4117605198797704e-05,
|
| 841 |
+
"loss": 2.142,
|
| 842 |
+
"step": 655
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 6.67,
|
| 846 |
+
"learning_rate": 3.429999999999998e-05,
|
| 847 |
+
"loss": 2.6776,
|
| 848 |
+
"step": 660
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 6.72,
|
| 852 |
+
"learning_rate": 2.5344087038001816e-05,
|
| 853 |
+
"loss": 2.6633,
|
| 854 |
+
"step": 665
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 6.77,
|
| 858 |
+
"learning_rate": 1.7474858752243343e-05,
|
| 859 |
+
"loss": 2.7227,
|
| 860 |
+
"step": 670
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 6.82,
|
| 864 |
+
"learning_rate": 1.0890007647780993e-05,
|
| 865 |
+
"loss": 2.7123,
|
| 866 |
+
"step": 675
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 6.87,
|
| 870 |
+
"learning_rate": 5.754959820160079e-06,
|
| 871 |
+
"loss": 2.7949,
|
| 872 |
+
"step": 680
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 6.92,
|
| 876 |
+
"learning_rate": 2.1987190842099613e-06,
|
| 877 |
+
"loss": 2.6913,
|
| 878 |
+
"step": 685
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 6.97,
|
| 882 |
+
"learning_rate": 3.106261114864048e-07,
|
| 883 |
+
"loss": 2.5163,
|
| 884 |
+
"step": 690
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 7.0,
|
| 888 |
+
"eval_loss": 2.7900161743164062,
|
| 889 |
+
"eval_runtime": 3.2724,
|
| 890 |
+
"eval_samples_per_second": 43.393,
|
| 891 |
+
"eval_steps_per_second": 5.501,
|
| 892 |
+
"step": 693
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 7.02,
|
| 896 |
+
"learning_rate": 1.3811399402871997e-07,
|
| 897 |
+
"loss": 2.5678,
|
| 898 |
+
"step": 695
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 7.07,
|
| 902 |
+
"learning_rate": 1.6855166196610614e-06,
|
| 903 |
+
"loss": 2.4848,
|
| 904 |
+
"step": 700
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 7.12,
|
| 908 |
+
"learning_rate": 4.913959795097361e-06,
|
| 909 |
+
"loss": 2.603,
|
| 910 |
+
"step": 705
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 7.17,
|
| 914 |
+
"learning_rate": 9.742337852080565e-06,
|
| 915 |
+
"loss": 2.26,
|
| 916 |
+
"step": 710
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 7.22,
|
| 920 |
+
"learning_rate": 1.604935120203803e-05,
|
| 921 |
+
"loss": 2.3438,
|
| 922 |
+
"step": 715
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 7.27,
|
| 926 |
+
"learning_rate": 2.367655365135328e-05,
|
| 927 |
+
"loss": 2.4659,
|
| 928 |
+
"step": 720
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 7.32,
|
| 932 |
+
"learning_rate": 3.243233292191947e-05,
|
| 933 |
+
"loss": 2.1382,
|
| 934 |
+
"step": 725
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 7.37,
|
| 938 |
+
"learning_rate": 4.2096724377451216e-05,
|
| 939 |
+
"loss": 2.2814,
|
| 940 |
+
"step": 730
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 7.42,
|
| 944 |
+
"learning_rate": 5.242693702405326e-05,
|
| 945 |
+
"loss": 2.212,
|
| 946 |
+
"step": 735
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 7.47,
|
| 950 |
+
"learning_rate": 6.316345295962419e-05,
|
| 951 |
+
"loss": 2.4038,
|
| 952 |
+
"step": 740
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 7.53,
|
| 956 |
+
"learning_rate": 7.403654704037569e-05,
|
| 957 |
+
"loss": 2.4106,
|
| 958 |
+
"step": 745
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 7.58,
|
| 962 |
+
"learning_rate": 8.477306297594661e-05,
|
| 963 |
+
"loss": 2.4768,
|
| 964 |
+
"step": 750
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 7.63,
|
| 968 |
+
"learning_rate": 9.510327562254866e-05,
|
| 969 |
+
"loss": 2.3888,
|
| 970 |
+
"step": 755
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 7.68,
|
| 974 |
+
"learning_rate": 0.00010476766707808043,
|
| 975 |
+
"loss": 2.4749,
|
| 976 |
+
"step": 760
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 7.73,
|
| 980 |
+
"learning_rate": 0.00011352344634864663,
|
| 981 |
+
"loss": 2.5513,
|
| 982 |
+
"step": 765
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 7.78,
|
| 986 |
+
"learning_rate": 0.00012115064879796188,
|
| 987 |
+
"loss": 2.5604,
|
| 988 |
+
"step": 770
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 7.83,
|
| 992 |
+
"learning_rate": 0.00012745766214791938,
|
| 993 |
+
"loss": 2.3697,
|
| 994 |
+
"step": 775
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 7.88,
|
| 998 |
+
"learning_rate": 0.0001322860402049026,
|
| 999 |
+
"loss": 2.3862,
|
| 1000 |
+
"step": 780
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 7.93,
|
| 1004 |
+
"learning_rate": 0.0001355144833803389,
|
| 1005 |
+
"loss": 2.323,
|
| 1006 |
+
"step": 785
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 7.98,
|
| 1010 |
+
"learning_rate": 0.00013706188600597126,
|
| 1011 |
+
"loss": 2.6325,
|
| 1012 |
+
"step": 790
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 8.0,
|
| 1016 |
+
"eval_loss": 2.807237148284912,
|
| 1017 |
+
"eval_runtime": 3.2933,
|
| 1018 |
+
"eval_samples_per_second": 43.118,
|
| 1019 |
+
"eval_steps_per_second": 5.466,
|
| 1020 |
+
"step": 792
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 8.03,
|
| 1024 |
+
"learning_rate": 0.00013688937388851358,
|
| 1025 |
+
"loss": 2.3013,
|
| 1026 |
+
"step": 795
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 8.08,
|
| 1030 |
+
"learning_rate": 0.00013500128091579002,
|
| 1031 |
+
"loss": 2.5725,
|
| 1032 |
+
"step": 800
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 8.13,
|
| 1036 |
+
"learning_rate": 0.00013144504017983998,
|
| 1037 |
+
"loss": 2.2264,
|
| 1038 |
+
"step": 805
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 8.18,
|
| 1042 |
+
"learning_rate": 0.00012630999235221907,
|
| 1043 |
+
"loss": 1.9343,
|
| 1044 |
+
"step": 810
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 8.23,
|
| 1048 |
+
"learning_rate": 0.0001197251412477569,
|
| 1049 |
+
"loss": 2.2098,
|
| 1050 |
+
"step": 815
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 8.28,
|
| 1054 |
+
"learning_rate": 0.00011185591296199847,
|
| 1055 |
+
"loss": 1.9414,
|
| 1056 |
+
"step": 820
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 8.33,
|
| 1060 |
+
"learning_rate": 0.00010289999999999993,
|
| 1061 |
+
"loss": 2.5007,
|
| 1062 |
+
"step": 825
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 8.38,
|
| 1066 |
+
"learning_rate": 9.308239480120242e-05,
|
| 1067 |
+
"loss": 2.2473,
|
| 1068 |
+
"step": 830
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 8.43,
|
| 1072 |
+
"learning_rate": 8.264973742927219e-05,
|
| 1073 |
+
"loss": 2.648,
|
| 1074 |
+
"step": 835
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 8.48,
|
| 1078 |
+
"learning_rate": 7.186411942550894e-05,
|
| 1079 |
+
"loss": 2.4495,
|
| 1080 |
+
"step": 840
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 8.54,
|
| 1084 |
+
"learning_rate": 6.0996499486790955e-05,
|
| 1085 |
+
"loss": 2.3626,
|
| 1086 |
+
"step": 845
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 8.59,
|
| 1090 |
+
"learning_rate": 5.0319896380863535e-05,
|
| 1091 |
+
"loss": 2.3541,
|
| 1092 |
+
"step": 850
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 8.64,
|
| 1096 |
+
"learning_rate": 4.0102530108070616e-05,
|
| 1097 |
+
"loss": 2.1531,
|
| 1098 |
+
"step": 855
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 8.69,
|
| 1102 |
+
"learning_rate": 3.0601083618784945e-05,
|
| 1103 |
+
"loss": 2.3576,
|
| 1104 |
+
"step": 860
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 8.74,
|
| 1108 |
+
"learning_rate": 2.20542543671809e-05,
|
| 1109 |
+
"loss": 1.9389,
|
| 1110 |
+
"step": 865
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 8.79,
|
| 1114 |
+
"learning_rate": 1.4676757700644984e-05,
|
| 1115 |
+
"loss": 2.3032,
|
| 1116 |
+
"step": 870
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 8.84,
|
| 1120 |
+
"learning_rate": 8.653932733012724e-06,
|
| 1121 |
+
"loss": 2.5286,
|
| 1122 |
+
"step": 875
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 8.89,
|
| 1126 |
+
"learning_rate": 4.137086214086705e-06,
|
| 1127 |
+
"loss": 2.1202,
|
| 1128 |
+
"step": 880
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 8.94,
|
| 1132 |
+
"learning_rate": 1.2396913677783503e-06,
|
| 1133 |
+
"loss": 2.24,
|
| 1134 |
+
"step": 885
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 8.99,
|
| 1138 |
+
"learning_rate": 3.4537192513510953e-08,
|
| 1139 |
+
"loss": 2.1317,
|
| 1140 |
+
"step": 890
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 9.0,
|
| 1144 |
+
"eval_loss": 2.797985315322876,
|
| 1145 |
+
"eval_runtime": 3.2929,
|
| 1146 |
+
"eval_samples_per_second": 43.124,
|
| 1147 |
+
"eval_steps_per_second": 5.466,
|
| 1148 |
+
"step": 891
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 9.04,
|
| 1152 |
+
"learning_rate": 5.51899839807432e-07,
|
| 1153 |
+
"loss": 2.1029,
|
| 1154 |
+
"step": 895
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 9.09,
|
| 1158 |
+
"learning_rate": 2.778782010045488e-06,
|
| 1159 |
+
"loss": 1.9706,
|
| 1160 |
+
"step": 900
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 9.14,
|
| 1164 |
+
"learning_rate": 6.6592394735377576e-06,
|
| 1165 |
+
"loss": 2.1198,
|
| 1166 |
+
"step": 905
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 9.19,
|
| 1170 |
+
"learning_rate": 1.2095786513913378e-05,
|
| 1171 |
+
"loss": 2.1092,
|
| 1172 |
+
"step": 910
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 9.24,
|
| 1176 |
+
"learning_rate": 1.895184498599217e-05,
|
| 1177 |
+
"loss": 2.1326,
|
| 1178 |
+
"step": 915
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 9.29,
|
| 1182 |
+
"learning_rate": 2.7055175462355963e-05,
|
| 1183 |
+
"loss": 2.0587,
|
| 1184 |
+
"step": 920
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 9.34,
|
| 1188 |
+
"learning_rate": 3.6202204270593973e-05,
|
| 1189 |
+
"loss": 1.9377,
|
| 1190 |
+
"step": 925
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 9.39,
|
| 1194 |
+
"learning_rate": 4.6163137716425034e-05,
|
| 1195 |
+
"loss": 2.0585,
|
| 1196 |
+
"step": 930
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 9.44,
|
| 1200 |
+
"learning_rate": 5.6687735012048636e-05,
|
| 1201 |
+
"loss": 1.9919,
|
| 1202 |
+
"step": 935
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 9.49,
|
| 1206 |
+
"learning_rate": 6.751159488093214e-05,
|
| 1207 |
+
"loss": 1.8915,
|
| 1208 |
+
"step": 940
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 9.55,
|
| 1212 |
+
"learning_rate": 7.836279790554722e-05,
|
| 1213 |
+
"loss": 1.871,
|
| 1214 |
+
"step": 945
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 9.6,
|
| 1218 |
+
"learning_rate": 8.896873774751943e-05,
|
| 1219 |
+
"loss": 2.1734,
|
| 1220 |
+
"step": 950
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 9.65,
|
| 1224 |
+
"learning_rate": 9.906296962475623e-05,
|
| 1225 |
+
"loss": 2.1848,
|
| 1226 |
+
"step": 955
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 9.7,
|
| 1230 |
+
"learning_rate": 0.00010839190399658423,
|
| 1231 |
+
"loss": 2.1742,
|
| 1232 |
+
"step": 960
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 9.75,
|
| 1236 |
+
"learning_rate": 0.00011672117729665359,
|
| 1237 |
+
"loss": 2.3443,
|
| 1238 |
+
"step": 965
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 9.8,
|
| 1242 |
+
"learning_rate": 0.00012384153966663053,
|
| 1243 |
+
"loss": 1.857,
|
| 1244 |
+
"step": 970
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 9.85,
|
| 1248 |
+
"learning_rate": 0.00012957411177772762,
|
| 1249 |
+
"loss": 2.2031,
|
| 1250 |
+
"step": 975
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 9.9,
|
| 1254 |
+
"learning_rate": 0.0001337748786770289,
|
| 1255 |
+
"loss": 2.1931,
|
| 1256 |
+
"step": 980
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 9.95,
|
| 1260 |
+
"learning_rate": 0.00013633830776320065,
|
| 1261 |
+
"loss": 2.0811,
|
| 1262 |
+
"step": 985
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 10.0,
|
| 1266 |
+
"learning_rate": 0.0001372,
|
| 1267 |
+
"loss": 2.3998,
|
| 1268 |
+
"step": 990
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 10.0,
|
| 1272 |
+
"eval_loss": 2.7857184410095215,
|
| 1273 |
+
"eval_runtime": 3.2779,
|
| 1274 |
+
"eval_samples_per_second": 43.32,
|
| 1275 |
+
"eval_steps_per_second": 5.491,
|
| 1276 |
+
"step": 990
|
| 1277 |
}
|
| 1278 |
],
|
| 1279 |
+
"max_steps": 10395,
|
| 1280 |
+
"num_train_epochs": 105,
|
| 1281 |
+
"total_flos": 1032103526400000.0,
|
| 1282 |
"trial_name": null,
|
| 1283 |
"trial_params": null
|
| 1284 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2863
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87b339441b1f6019e8540c7c71ded5ff035d2f0e1f074c763f744b0da4bc0c37
|
| 3 |
size 2863
|