Commit
•
117abd6
1
Parent(s):
3c99844
huggingartists
Browse files- README.md +3 -3
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- trainer_state.json +575 -7
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
| 45 |
dataset = load_dataset("huggingartists/eminem")
|
| 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 Eminem'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/eminem")
|
| 46 |
```
|
| 47 |
|
| 48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/3mcytn62/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 Eminem's lyrics.
|
| 53 |
|
| 54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3rxlo957) for full transparency and reproducibility.
|
| 55 |
|
| 56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3rxlo957/artifacts) is logged and versioned.
|
| 57 |
|
| 58 |
## How to use
|
| 59 |
|
evaluation.txt
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"eval_loss": 2.
|
|
|
|
| 1 |
+
{"eval_loss": 2.0775227546691895, "eval_runtime": 29.9794, "eval_samples_per_second": 22.315, "eval_steps_per_second": 2.802, "epoch": 4.0}
|
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 sha256:068ab5abee507d02e0ceddd81264e97625f1b2536ccddc34bf29b3c8b3ddf237
|
| 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:ed02c35a40feea9337942fb2d0af6e045d9ae959c5043e7a949880e6920ec465
|
| 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:c794e8f1590769642f891be807aaea565ce2ef03feb2f6701d1072683993b5ca
|
| 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 14503
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec4a65a6312bd30b99e17460a422429487e35f933075a640659a7e061f47d2ca
|
| 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 sha256:0d75ff6bd24c5b9f2bcd92e3bca8ca05f4520330d9b4bbbc3cf07c312fdc8ba2
|
| 3 |
size 623
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 2.
|
| 3 |
-
"best_model_checkpoint": "output/eminem/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -1644,11 +1644,579 @@
|
|
| 1644 |
"eval_samples_per_second": 22.329,
|
| 1645 |
"eval_steps_per_second": 2.807,
|
| 1646 |
"step": 1347
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1647 |
}
|
| 1648 |
],
|
| 1649 |
-
"max_steps":
|
| 1650 |
-
"num_train_epochs":
|
| 1651 |
-
"total_flos":
|
| 1652 |
"trial_name": null,
|
| 1653 |
"trial_params": null
|
| 1654 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.0775227546691895,
|
| 3 |
+
"best_model_checkpoint": "output/eminem/checkpoint-1808",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 1808,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1644 |
"eval_samples_per_second": 22.329,
|
| 1645 |
"eval_steps_per_second": 2.807,
|
| 1646 |
"step": 1347
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.99,
|
| 1650 |
+
"learning_rate": 0.000135403446735711,
|
| 1651 |
+
"loss": 2.1728,
|
| 1652 |
+
"step": 1350
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 3.0,
|
| 1656 |
+
"learning_rate": 0.00013482119897697306,
|
| 1657 |
+
"loss": 2.2896,
|
| 1658 |
+
"step": 1355
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 3.0,
|
| 1662 |
+
"eval_loss": 2.07898211479187,
|
| 1663 |
+
"eval_runtime": 29.7025,
|
| 1664 |
+
"eval_samples_per_second": 22.523,
|
| 1665 |
+
"eval_steps_per_second": 2.828,
|
| 1666 |
+
"step": 1356
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 3.01,
|
| 1670 |
+
"learning_rate": 0.0001341589832217866,
|
| 1671 |
+
"loss": 2.2492,
|
| 1672 |
+
"step": 1360
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 3.02,
|
| 1676 |
+
"learning_rate": 0.00013341759915466326,
|
| 1677 |
+
"loss": 2.4062,
|
| 1678 |
+
"step": 1365
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 3.03,
|
| 1682 |
+
"learning_rate": 0.00013259794206290322,
|
| 1683 |
+
"loss": 2.4505,
|
| 1684 |
+
"step": 1370
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 3.04,
|
| 1688 |
+
"learning_rate": 0.00013170100175545615,
|
| 1689 |
+
"loss": 2.3168,
|
| 1690 |
+
"step": 1375
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 3.05,
|
| 1694 |
+
"learning_rate": 0.00013072786136763862,
|
| 1695 |
+
"loss": 2.4331,
|
| 1696 |
+
"step": 1380
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 3.06,
|
| 1700 |
+
"learning_rate": 0.0001296796960531519,
|
| 1701 |
+
"loss": 2.3448,
|
| 1702 |
+
"step": 1385
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 3.08,
|
| 1706 |
+
"learning_rate": 0.00012855777156497896,
|
| 1707 |
+
"loss": 2.5707,
|
| 1708 |
+
"step": 1390
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 3.09,
|
| 1712 |
+
"learning_rate": 0.00012736344272687528,
|
| 1713 |
+
"loss": 2.3018,
|
| 1714 |
+
"step": 1395
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 3.1,
|
| 1718 |
+
"learning_rate": 0.00012609815179729852,
|
| 1719 |
+
"loss": 2.2497,
|
| 1720 |
+
"step": 1400
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 3.11,
|
| 1724 |
+
"learning_rate": 0.00012476342672775167,
|
| 1725 |
+
"loss": 2.2297,
|
| 1726 |
+
"step": 1405
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 3.12,
|
| 1730 |
+
"learning_rate": 0.00012336087931764827,
|
| 1731 |
+
"loss": 2.5308,
|
| 1732 |
+
"step": 1410
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 3.13,
|
| 1736 |
+
"learning_rate": 0.00012189220326791867,
|
| 1737 |
+
"loss": 2.4956,
|
| 1738 |
+
"step": 1415
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 3.14,
|
| 1742 |
+
"learning_rate": 0.00012035917213571493,
|
| 1743 |
+
"loss": 2.259,
|
| 1744 |
+
"step": 1420
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 3.15,
|
| 1748 |
+
"learning_rate": 0.0001187636371926859,
|
| 1749 |
+
"loss": 2.5147,
|
| 1750 |
+
"step": 1425
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 3.16,
|
| 1754 |
+
"learning_rate": 0.00011710752518939737,
|
| 1755 |
+
"loss": 2.5795,
|
| 1756 |
+
"step": 1430
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 3.17,
|
| 1760 |
+
"learning_rate": 0.00011539283602861218,
|
| 1761 |
+
"loss": 2.4607,
|
| 1762 |
+
"step": 1435
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 3.19,
|
| 1766 |
+
"learning_rate": 0.00011362164035022977,
|
| 1767 |
+
"loss": 2.4427,
|
| 1768 |
+
"step": 1440
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 3.2,
|
| 1772 |
+
"learning_rate": 0.00011179607703080082,
|
| 1773 |
+
"loss": 2.3976,
|
| 1774 |
+
"step": 1445
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 3.21,
|
| 1778 |
+
"learning_rate": 0.00010991835060064675,
|
| 1779 |
+
"loss": 2.3127,
|
| 1780 |
+
"step": 1450
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 3.22,
|
| 1784 |
+
"learning_rate": 0.00010799072858169134,
|
| 1785 |
+
"loss": 2.1402,
|
| 1786 |
+
"step": 1455
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 3.23,
|
| 1790 |
+
"learning_rate": 0.0001060155387492263,
|
| 1791 |
+
"loss": 2.319,
|
| 1792 |
+
"step": 1460
|
| 1793 |
+
},
|
| 1794 |
+
{
|
| 1795 |
+
"epoch": 3.24,
|
| 1796 |
+
"learning_rate": 0.00010399516632091525,
|
| 1797 |
+
"loss": 2.6207,
|
| 1798 |
+
"step": 1465
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 3.25,
|
| 1802 |
+
"learning_rate": 0.000101932051076431,
|
| 1803 |
+
"loss": 2.5146,
|
| 1804 |
+
"step": 1470
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 3.26,
|
| 1808 |
+
"learning_rate": 9.982868441120418e-05,
|
| 1809 |
+
"loss": 2.513,
|
| 1810 |
+
"step": 1475
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 3.27,
|
| 1814 |
+
"learning_rate": 9.768760632784142e-05,
|
| 1815 |
+
"loss": 2.5331,
|
| 1816 |
+
"step": 1480
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 3.29,
|
| 1820 |
+
"learning_rate": 9.551140236884636e-05,
|
| 1821 |
+
"loss": 2.283,
|
| 1822 |
+
"step": 1485
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 3.3,
|
| 1826 |
+
"learning_rate": 9.330270049434448e-05,
|
| 1827 |
+
"loss": 2.5398,
|
| 1828 |
+
"step": 1490
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 3.31,
|
| 1832 |
+
"learning_rate": 9.106416790859171e-05,
|
| 1833 |
+
"loss": 2.4427,
|
| 1834 |
+
"step": 1495
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 3.32,
|
| 1838 |
+
"learning_rate": 8.879850783908373e-05,
|
| 1839 |
+
"loss": 2.4624,
|
| 1840 |
+
"step": 1500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 3.33,
|
| 1844 |
+
"learning_rate": 8.650845627216773e-05,
|
| 1845 |
+
"loss": 2.3922,
|
| 1846 |
+
"step": 1505
|
| 1847 |
+
},
|
| 1848 |
+
{
|
| 1849 |
+
"epoch": 3.34,
|
| 1850 |
+
"learning_rate": 8.419677864910008e-05,
|
| 1851 |
+
"loss": 2.3814,
|
| 1852 |
+
"step": 1510
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 3.35,
|
| 1856 |
+
"learning_rate": 8.186626652652247e-05,
|
| 1857 |
+
"loss": 2.4956,
|
| 1858 |
+
"step": 1515
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 3.36,
|
| 1862 |
+
"learning_rate": 7.951973420541575e-05,
|
| 1863 |
+
"loss": 2.4434,
|
| 1864 |
+
"step": 1520
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 3.37,
|
| 1868 |
+
"learning_rate": 7.716001533257615e-05,
|
| 1869 |
+
"loss": 2.2408,
|
| 1870 |
+
"step": 1525
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 3.38,
|
| 1874 |
+
"learning_rate": 7.478995947873418e-05,
|
| 1875 |
+
"loss": 2.333,
|
| 1876 |
+
"step": 1530
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 3.4,
|
| 1880 |
+
"learning_rate": 7.241242869744915e-05,
|
| 1881 |
+
"loss": 2.412,
|
| 1882 |
+
"step": 1535
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 3.41,
|
| 1886 |
+
"learning_rate": 7.00302940689176e-05,
|
| 1887 |
+
"loss": 2.1938,
|
| 1888 |
+
"step": 1540
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 3.42,
|
| 1892 |
+
"learning_rate": 6.764643223289375e-05,
|
| 1893 |
+
"loss": 2.4952,
|
| 1894 |
+
"step": 1545
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 3.43,
|
| 1898 |
+
"learning_rate": 6.526372191488815e-05,
|
| 1899 |
+
"loss": 2.346,
|
| 1900 |
+
"step": 1550
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 3.44,
|
| 1904 |
+
"learning_rate": 6.288504044985084e-05,
|
| 1905 |
+
"loss": 2.4523,
|
| 1906 |
+
"step": 1555
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 3.45,
|
| 1910 |
+
"learning_rate": 6.051326030753298e-05,
|
| 1911 |
+
"loss": 2.3246,
|
| 1912 |
+
"step": 1560
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 3.46,
|
| 1916 |
+
"learning_rate": 5.815124562372384e-05,
|
| 1917 |
+
"loss": 2.3573,
|
| 1918 |
+
"step": 1565
|
| 1919 |
+
},
|
| 1920 |
+
{
|
| 1921 |
+
"epoch": 3.47,
|
| 1922 |
+
"learning_rate": 5.580184874155164e-05,
|
| 1923 |
+
"loss": 2.5265,
|
| 1924 |
+
"step": 1570
|
| 1925 |
+
},
|
| 1926 |
+
{
|
| 1927 |
+
"epoch": 3.48,
|
| 1928 |
+
"learning_rate": 5.346790676702522e-05,
|
| 1929 |
+
"loss": 2.4817,
|
| 1930 |
+
"step": 1575
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 3.5,
|
| 1934 |
+
"learning_rate": 5.115223814297577e-05,
|
| 1935 |
+
"loss": 2.3745,
|
| 1936 |
+
"step": 1580
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 3.51,
|
| 1940 |
+
"learning_rate": 4.885763924553609e-05,
|
| 1941 |
+
"loss": 2.3791,
|
| 1942 |
+
"step": 1585
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 3.52,
|
| 1946 |
+
"learning_rate": 4.658688100726727e-05,
|
| 1947 |
+
"loss": 2.438,
|
| 1948 |
+
"step": 1590
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 3.53,
|
| 1952 |
+
"learning_rate": 4.434270557101133e-05,
|
| 1953 |
+
"loss": 2.2344,
|
| 1954 |
+
"step": 1595
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 3.54,
|
| 1958 |
+
"learning_rate": 4.212782297850713e-05,
|
| 1959 |
+
"loss": 2.3451,
|
| 1960 |
+
"step": 1600
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 3.55,
|
| 1964 |
+
"learning_rate": 3.9944907897778607e-05,
|
| 1965 |
+
"loss": 2.4611,
|
| 1966 |
+
"step": 1605
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 3.56,
|
| 1970 |
+
"learning_rate": 3.7796596393228573e-05,
|
| 1971 |
+
"loss": 2.4182,
|
| 1972 |
+
"step": 1610
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 3.57,
|
| 1976 |
+
"learning_rate": 3.568548274236061e-05,
|
| 1977 |
+
"loss": 2.5135,
|
| 1978 |
+
"step": 1615
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 3.58,
|
| 1982 |
+
"learning_rate": 3.361411630295785e-05,
|
| 1983 |
+
"loss": 2.5817,
|
| 1984 |
+
"step": 1620
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 3.6,
|
| 1988 |
+
"learning_rate": 3.158499843450252e-05,
|
| 1989 |
+
"loss": 2.3341,
|
| 1990 |
+
"step": 1625
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 3.61,
|
| 1994 |
+
"learning_rate": 2.9600579477565787e-05,
|
| 1995 |
+
"loss": 2.3179,
|
| 1996 |
+
"step": 1630
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 3.62,
|
| 2000 |
+
"learning_rate": 2.7663255794801203e-05,
|
| 2001 |
+
"loss": 2.29,
|
| 2002 |
+
"step": 1635
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 3.63,
|
| 2006 |
+
"learning_rate": 2.57753668771237e-05,
|
| 2007 |
+
"loss": 2.5001,
|
| 2008 |
+
"step": 1640
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 3.64,
|
| 2012 |
+
"learning_rate": 2.3939192518565938e-05,
|
| 2013 |
+
"loss": 2.5561,
|
| 2014 |
+
"step": 1645
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 3.65,
|
| 2018 |
+
"learning_rate": 2.215695006322396e-05,
|
| 2019 |
+
"loss": 2.4063,
|
| 2020 |
+
"step": 1650
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 3.66,
|
| 2024 |
+
"learning_rate": 2.0430791727616646e-05,
|
| 2025 |
+
"loss": 2.377,
|
| 2026 |
+
"step": 1655
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 3.67,
|
| 2030 |
+
"learning_rate": 1.8762802001692604e-05,
|
| 2031 |
+
"loss": 2.6203,
|
| 2032 |
+
"step": 1660
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 3.68,
|
| 2036 |
+
"learning_rate": 1.715499513162315e-05,
|
| 2037 |
+
"loss": 2.434,
|
| 2038 |
+
"step": 1665
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 3.69,
|
| 2042 |
+
"learning_rate": 1.5609312687419068e-05,
|
| 2043 |
+
"loss": 2.2921,
|
| 2044 |
+
"step": 1670
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 3.71,
|
| 2048 |
+
"learning_rate": 1.4127621218315069e-05,
|
| 2049 |
+
"loss": 2.2474,
|
| 2050 |
+
"step": 1675
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 3.72,
|
| 2054 |
+
"learning_rate": 1.2711709998742125e-05,
|
| 2055 |
+
"loss": 2.4056,
|
| 2056 |
+
"step": 1680
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"epoch": 3.73,
|
| 2060 |
+
"learning_rate": 1.1363288867620807e-05,
|
| 2061 |
+
"loss": 2.1388,
|
| 2062 |
+
"step": 1685
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 3.74,
|
| 2066 |
+
"learning_rate": 1.0083986163578448e-05,
|
| 2067 |
+
"loss": 2.4683,
|
| 2068 |
+
"step": 1690
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 3.75,
|
| 2072 |
+
"learning_rate": 8.875346758583873e-06,
|
| 2073 |
+
"loss": 2.2937,
|
| 2074 |
+
"step": 1695
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 3.76,
|
| 2078 |
+
"learning_rate": 7.73883019237949e-06,
|
| 2079 |
+
"loss": 2.4202,
|
| 2080 |
+
"step": 1700
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 3.77,
|
| 2084 |
+
"learning_rate": 6.6758089099546965e-06,
|
| 2085 |
+
"loss": 2.5003,
|
| 2086 |
+
"step": 1705
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 3.78,
|
| 2090 |
+
"learning_rate": 5.6875666041964786e-06,
|
| 2091 |
+
"loss": 2.3068,
|
| 2092 |
+
"step": 1710
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 3.79,
|
| 2096 |
+
"learning_rate": 4.775296665718002e-06,
|
| 2097 |
+
"loss": 2.2147,
|
| 2098 |
+
"step": 1715
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 3.81,
|
| 2102 |
+
"learning_rate": 3.940100741730249e-06,
|
| 2103 |
+
"loss": 2.4371,
|
| 2104 |
+
"step": 1720
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 3.82,
|
| 2108 |
+
"learning_rate": 3.1829874057071e-06,
|
| 2109 |
+
"loss": 2.3255,
|
| 2110 |
+
"step": 1725
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 3.83,
|
| 2114 |
+
"learning_rate": 2.504870939442012e-06,
|
| 2115 |
+
"loss": 2.3511,
|
| 2116 |
+
"step": 1730
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 3.84,
|
| 2120 |
+
"learning_rate": 1.9065702289715435e-06,
|
| 2121 |
+
"loss": 2.587,
|
| 2122 |
+
"step": 1735
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 3.85,
|
| 2126 |
+
"learning_rate": 1.3888077756974934e-06,
|
| 2127 |
+
"loss": 2.4066,
|
| 2128 |
+
"step": 1740
|
| 2129 |
+
},
|
| 2130 |
+
{
|
| 2131 |
+
"epoch": 3.86,
|
| 2132 |
+
"learning_rate": 9.522088239021213e-07,
|
| 2133 |
+
"loss": 2.3034,
|
| 2134 |
+
"step": 1745
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 3.87,
|
| 2138 |
+
"learning_rate": 5.973006057099719e-07,
|
| 2139 |
+
"loss": 2.2354,
|
| 2140 |
+
"step": 1750
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 3.88,
|
| 2144 |
+
"learning_rate": 3.2451170440812575e-07,
|
| 2145 |
+
"loss": 2.4044,
|
| 2146 |
+
"step": 1755
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 3.89,
|
| 2150 |
+
"learning_rate": 1.3417153689369494e-07,
|
| 2151 |
+
"loss": 2.2793,
|
| 2152 |
+
"step": 1760
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 3.9,
|
| 2156 |
+
"learning_rate": 2.650995587359215e-08,
|
| 2157 |
+
"loss": 2.3649,
|
| 2158 |
+
"step": 1765
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 3.92,
|
| 2162 |
+
"learning_rate": 1.6569722969137058e-09,
|
| 2163 |
+
"loss": 2.5027,
|
| 2164 |
+
"step": 1770
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 3.93,
|
| 2168 |
+
"learning_rate": 5.964259835509249e-08,
|
| 2169 |
+
"loss": 2.3313,
|
| 2170 |
+
"step": 1775
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 3.94,
|
| 2174 |
+
"learning_rate": 2.0039681123962964e-07,
|
| 2175 |
+
"loss": 2.3823,
|
| 2176 |
+
"step": 1780
|
| 2177 |
+
},
|
| 2178 |
+
{
|
| 2179 |
+
"epoch": 3.95,
|
| 2180 |
+
"learning_rate": 4.2374963770060374e-07,
|
| 2181 |
+
"loss": 2.3593,
|
| 2182 |
+
"step": 1785
|
| 2183 |
+
},
|
| 2184 |
+
{
|
| 2185 |
+
"epoch": 3.96,
|
| 2186 |
+
"learning_rate": 7.294313593049221e-07,
|
| 2187 |
+
"loss": 2.4781,
|
| 2188 |
+
"step": 1790
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 3.97,
|
| 2192 |
+
"learning_rate": 1.1170728381450536e-06,
|
| 2193 |
+
"loss": 2.4511,
|
| 2194 |
+
"step": 1795
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 3.98,
|
| 2198 |
+
"learning_rate": 1.5862059626056908e-06,
|
| 2199 |
+
"loss": 2.0358,
|
| 2200 |
+
"step": 1800
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 3.99,
|
| 2204 |
+
"learning_rate": 2.1362642126510735e-06,
|
| 2205 |
+
"loss": 2.4982,
|
| 2206 |
+
"step": 1805
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 4.0,
|
| 2210 |
+
"eval_loss": 2.0775227546691895,
|
| 2211 |
+
"eval_runtime": 29.8853,
|
| 2212 |
+
"eval_samples_per_second": 22.386,
|
| 2213 |
+
"eval_steps_per_second": 2.811,
|
| 2214 |
+
"step": 1808
|
| 2215 |
}
|
| 2216 |
],
|
| 2217 |
+
"max_steps": 1808,
|
| 2218 |
+
"num_train_epochs": 4,
|
| 2219 |
+
"total_flos": 1886005886976000.0,
|
| 2220 |
"trial_name": null,
|
| 2221 |
"trial_params": null
|
| 2222 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2671
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2740716c0b718fe72809d1fd0e35d4a082f6a5fd5928fa62e0b59f56579d0256
|
| 3 |
size 2671
|