Commit
•
e2275aa
1
Parent(s):
60643e2
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 +591 -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/29aftir4/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/1yj0yyz9) for full transparency and reproducibility.
|
| 55 |
|
| 56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1yj0yyz9/artifacts) is logged and versioned.
|
| 57 |
|
| 58 |
## How to use
|
| 59 |
|
evaluation.txt
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"eval_loss": 0.
|
|
|
|
| 1 |
+
{"eval_loss": 0.4826279878616333, "eval_runtime": 14.6559, "eval_samples_per_second": 44.897, "eval_steps_per_second": 5.663, "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 sha256:428633d1f6458ab43ed5eee44c1ef0505fe223f57c99426fe21435d1aa3b434b
|
| 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:f52b4dcb8476f623a272e10ac84ac049de0f1003081728e8884c6d2f759310f9
|
| 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 510396521
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:753a59950d6b66574f6214a57384f99ae242cd291bd19d5d9e54977a42f5f557
|
| 3 |
size 510396521
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14567
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32f46f1e2250c32c3ee6983a1c0513e245212fb9a74b0dd129330fa6abe719c7
|
| 3 |
size 14567
|
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 sha256:33680f79177570c60fb9a75223c44dc245db8eeb97bbb366911a9c63f50a44d8
|
| 3 |
size 623
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 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,
|
|
@@ -1776,11 +1776,595 @@
|
|
| 1776 |
"eval_samples_per_second": 81.452,
|
| 1777 |
"eval_steps_per_second": 10.291,
|
| 1778 |
"step": 1455
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1779 |
}
|
| 1780 |
],
|
| 1781 |
-
"max_steps":
|
| 1782 |
-
"num_train_epochs":
|
| 1783 |
-
"total_flos":
|
| 1784 |
"trial_name": null,
|
| 1785 |
"trial_params": null
|
| 1786 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4826279878616333,
|
| 3 |
+
"best_model_checkpoint": "output/eminem/checkpoint-1936",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 1936,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 1776 |
"eval_samples_per_second": 81.452,
|
| 1777 |
"eval_steps_per_second": 10.291,
|
| 1778 |
"step": 1455
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 3.02,
|
| 1782 |
+
"learning_rate": 3.533340602884326e-05,
|
| 1783 |
+
"loss": 0.6584,
|
| 1784 |
+
"step": 1460
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 3.03,
|
| 1788 |
+
"learning_rate": 3.340418885898272e-05,
|
| 1789 |
+
"loss": 0.9188,
|
| 1790 |
+
"step": 1465
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 3.04,
|
| 1794 |
+
"learning_rate": 3.1512039922545005e-05,
|
| 1795 |
+
"loss": 0.7885,
|
| 1796 |
+
"step": 1470
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 3.05,
|
| 1800 |
+
"learning_rate": 2.9658952031065493e-05,
|
| 1801 |
+
"loss": 0.6785,
|
| 1802 |
+
"step": 1475
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 3.06,
|
| 1806 |
+
"learning_rate": 2.784687685697823e-05,
|
| 1807 |
+
"loss": 0.9316,
|
| 1808 |
+
"step": 1480
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 3.07,
|
| 1812 |
+
"learning_rate": 2.6077722878123704e-05,
|
| 1813 |
+
"loss": 0.7074,
|
| 1814 |
+
"step": 1485
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 3.08,
|
| 1818 |
+
"learning_rate": 2.4353353367727083e-05,
|
| 1819 |
+
"loss": 0.5083,
|
| 1820 |
+
"step": 1490
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 3.09,
|
| 1824 |
+
"learning_rate": 2.2675584432009507e-05,
|
| 1825 |
+
"loss": 0.7031,
|
| 1826 |
+
"step": 1495
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 3.1,
|
| 1830 |
+
"learning_rate": 2.104618309745989e-05,
|
| 1831 |
+
"loss": 0.5955,
|
| 1832 |
+
"step": 1500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 3.11,
|
| 1836 |
+
"learning_rate": 1.94668654497983e-05,
|
| 1837 |
+
"loss": 0.6515,
|
| 1838 |
+
"step": 1505
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 3.12,
|
| 1842 |
+
"learning_rate": 1.7939294826601462e-05,
|
| 1843 |
+
"loss": 0.6982,
|
| 1844 |
+
"step": 1510
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 3.13,
|
| 1848 |
+
"learning_rate": 1.6465080065474947e-05,
|
| 1849 |
+
"loss": 0.7575,
|
| 1850 |
+
"step": 1515
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 3.14,
|
| 1854 |
+
"learning_rate": 1.504577380961979e-05,
|
| 1855 |
+
"loss": 0.7294,
|
| 1856 |
+
"step": 1520
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 3.15,
|
| 1860 |
+
"learning_rate": 1.3682870872588951e-05,
|
| 1861 |
+
"loss": 0.566,
|
| 1862 |
+
"step": 1525
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 3.16,
|
| 1866 |
+
"learning_rate": 1.2377806663955526e-05,
|
| 1867 |
+
"loss": 0.4467,
|
| 1868 |
+
"step": 1530
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 3.17,
|
| 1872 |
+
"learning_rate": 1.1131955677535227e-05,
|
| 1873 |
+
"loss": 0.8649,
|
| 1874 |
+
"step": 1535
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 3.18,
|
| 1878 |
+
"learning_rate": 9.946630043766268e-06,
|
| 1879 |
+
"loss": 0.6318,
|
| 1880 |
+
"step": 1540
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 3.19,
|
| 1884 |
+
"learning_rate": 8.8230781477809e-06,
|
| 1885 |
+
"loss": 0.5197,
|
| 1886 |
+
"step": 1545
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 3.2,
|
| 1890 |
+
"learning_rate": 7.762483314597675e-06,
|
| 1891 |
+
"loss": 0.7195,
|
| 1892 |
+
"step": 1550
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 3.21,
|
| 1896 |
+
"learning_rate": 6.765962562849582e-06,
|
| 1897 |
+
"loss": 0.6544,
|
| 1898 |
+
"step": 1555
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 3.22,
|
| 1902 |
+
"learning_rate": 5.834565428339295e-06,
|
| 1903 |
+
"loss": 0.6373,
|
| 1904 |
+
"step": 1560
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 3.23,
|
| 1908 |
+
"learning_rate": 4.969272858664371e-06,
|
| 1909 |
+
"loss": 0.7969,
|
| 1910 |
+
"step": 1565
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 3.24,
|
| 1914 |
+
"learning_rate": 4.170996180083756e-06,
|
| 1915 |
+
"loss": 0.5398,
|
| 1916 |
+
"step": 1570
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 3.25,
|
| 1920 |
+
"learning_rate": 3.440576137712397e-06,
|
| 1921 |
+
"loss": 0.6392,
|
| 1922 |
+
"step": 1575
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 3.26,
|
| 1926 |
+
"learning_rate": 2.778782010045892e-06,
|
| 1927 |
+
"loss": 0.8183,
|
| 1928 |
+
"step": 1580
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 3.27,
|
| 1932 |
+
"learning_rate": 2.186310798754732e-06,
|
| 1933 |
+
"loss": 0.4459,
|
| 1934 |
+
"step": 1585
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 3.29,
|
| 1938 |
+
"learning_rate": 1.6637864946060192e-06,
|
| 1939 |
+
"loss": 0.6059,
|
| 1940 |
+
"step": 1590
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 3.3,
|
| 1944 |
+
"learning_rate": 1.2117594202713427e-06,
|
| 1945 |
+
"loss": 0.9818,
|
| 1946 |
+
"step": 1595
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 3.31,
|
| 1950 |
+
"learning_rate": 8.307056507336243e-07,
|
| 1951 |
+
"loss": 0.3983,
|
| 1952 |
+
"step": 1600
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 3.32,
|
| 1956 |
+
"learning_rate": 5.21026511876941e-07,
|
| 1957 |
+
"loss": 0.9094,
|
| 1958 |
+
"step": 1605
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 3.33,
|
| 1962 |
+
"learning_rate": 2.8304815781598153e-07,
|
| 1963 |
+
"loss": 0.5006,
|
| 1964 |
+
"step": 1610
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 3.34,
|
| 1968 |
+
"learning_rate": 1.1702122738749632e-07,
|
| 1969 |
+
"loss": 0.5685,
|
| 1970 |
+
"step": 1615
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 3.35,
|
| 1974 |
+
"learning_rate": 2.312058017904961e-08,
|
| 1975 |
+
"loss": 0.7558,
|
| 1976 |
+
"step": 1620
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 3.36,
|
| 1980 |
+
"learning_rate": 1.445112365939849e-09,
|
| 1981 |
+
"loss": 0.5613,
|
| 1982 |
+
"step": 1625
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 3.37,
|
| 1986 |
+
"learning_rate": 5.2017652553981476e-08,
|
| 1987 |
+
"loss": 0.7474,
|
| 1988 |
+
"step": 1630
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 3.38,
|
| 1992 |
+
"learning_rate": 1.7478493773641084e-07,
|
| 1993 |
+
"loss": 0.7388,
|
| 1994 |
+
"step": 1635
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 3.39,
|
| 1998 |
+
"learning_rate": 3.69617669391192e-07,
|
| 1999 |
+
"loss": 0.5788,
|
| 2000 |
+
"step": 1640
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 3.4,
|
| 2004 |
+
"learning_rate": 6.363106496559428e-07,
|
| 2005 |
+
"loss": 0.6158,
|
| 2006 |
+
"step": 1645
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 3.41,
|
| 2010 |
+
"learning_rate": 9.745829974457439e-07,
|
| 2011 |
+
"loss": 0.6697,
|
| 2012 |
+
"step": 1650
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 3.42,
|
| 2016 |
+
"learning_rate": 1.3840784442740293e-06,
|
| 2017 |
+
"loss": 0.5538,
|
| 2018 |
+
"step": 1655
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 3.43,
|
| 2022 |
+
"learning_rate": 1.8643657094772689e-06,
|
| 2023 |
+
"loss": 0.7529,
|
| 2024 |
+
"step": 1660
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 3.44,
|
| 2028 |
+
"learning_rate": 2.4149389544374383e-06,
|
| 2029 |
+
"loss": 0.5164,
|
| 2030 |
+
"step": 1665
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 3.45,
|
| 2034 |
+
"learning_rate": 3.035218315332177e-06,
|
| 2035 |
+
"loss": 0.5686,
|
| 2036 |
+
"step": 1670
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 3.46,
|
| 2040 |
+
"learning_rate": 3.7245505138493062e-06,
|
| 2041 |
+
"loss": 0.7307,
|
| 2042 |
+
"step": 1675
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 3.47,
|
| 2046 |
+
"learning_rate": 4.482209545215727e-06,
|
| 2047 |
+
"loss": 0.4993,
|
| 2048 |
+
"step": 1680
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 3.48,
|
| 2052 |
+
"learning_rate": 5.307397442832051e-06,
|
| 2053 |
+
"loss": 0.5905,
|
| 2054 |
+
"step": 1685
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 3.49,
|
| 2058 |
+
"learning_rate": 6.199245118679951e-06,
|
| 2059 |
+
"loss": 0.7372,
|
| 2060 |
+
"step": 1690
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 3.5,
|
| 2064 |
+
"learning_rate": 7.15681327866027e-06,
|
| 2065 |
+
"loss": 0.7588,
|
| 2066 |
+
"step": 1695
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 3.51,
|
| 2070 |
+
"learning_rate": 8.179093411845718e-06,
|
| 2071 |
+
"loss": 0.7609,
|
| 2072 |
+
"step": 1700
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 3.52,
|
| 2076 |
+
"learning_rate": 9.265008852652901e-06,
|
| 2077 |
+
"loss": 0.4208,
|
| 2078 |
+
"step": 1705
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 3.53,
|
| 2082 |
+
"learning_rate": 1.0413415914780494e-05,
|
| 2083 |
+
"loss": 0.5304,
|
| 2084 |
+
"step": 1710
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 3.54,
|
| 2088 |
+
"learning_rate": 1.1623105095742268e-05,
|
| 2089 |
+
"loss": 0.47,
|
| 2090 |
+
"step": 1715
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 3.55,
|
| 2094 |
+
"learning_rate": 1.2892802350718737e-05,
|
| 2095 |
+
"loss": 0.4479,
|
| 2096 |
+
"step": 1720
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 3.56,
|
| 2100 |
+
"learning_rate": 1.4221170434376129e-05,
|
| 2101 |
+
"loss": 0.3907,
|
| 2102 |
+
"step": 1725
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 3.57,
|
| 2106 |
+
"learning_rate": 1.5606810309252903e-05,
|
| 2107 |
+
"loss": 0.8871,
|
| 2108 |
+
"step": 1730
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 3.58,
|
| 2112 |
+
"learning_rate": 1.7048262619230822e-05,
|
| 2113 |
+
"loss": 0.5595,
|
| 2114 |
+
"step": 1735
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 3.6,
|
| 2118 |
+
"learning_rate": 1.8544009226529548e-05,
|
| 2119 |
+
"loss": 0.6169,
|
| 2120 |
+
"step": 1740
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 3.61,
|
| 2124 |
+
"learning_rate": 2.0092474810601755e-05,
|
| 2125 |
+
"loss": 0.557,
|
| 2126 |
+
"step": 1745
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 3.62,
|
| 2130 |
+
"learning_rate": 2.169202852727143e-05,
|
| 2131 |
+
"loss": 0.8663,
|
| 2132 |
+
"step": 1750
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 3.63,
|
| 2136 |
+
"learning_rate": 2.334098572632982e-05,
|
| 2137 |
+
"loss": 0.754,
|
| 2138 |
+
"step": 1755
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 3.64,
|
| 2142 |
+
"learning_rate": 2.5037609725813137e-05,
|
| 2143 |
+
"loss": 0.7106,
|
| 2144 |
+
"step": 1760
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.65,
|
| 2148 |
+
"learning_rate": 2.6780113641080935e-05,
|
| 2149 |
+
"loss": 0.8902,
|
| 2150 |
+
"step": 1765
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 3.66,
|
| 2154 |
+
"learning_rate": 2.8566662266755925e-05,
|
| 2155 |
+
"loss": 0.5382,
|
| 2156 |
+
"step": 1770
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 3.67,
|
| 2160 |
+
"learning_rate": 3.039537400956214e-05,
|
| 2161 |
+
"loss": 0.5907,
|
| 2162 |
+
"step": 1775
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 3.68,
|
| 2166 |
+
"learning_rate": 3.226432287002555e-05,
|
| 2167 |
+
"loss": 1.0195,
|
| 2168 |
+
"step": 1780
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 3.69,
|
| 2172 |
+
"learning_rate": 3.417154047093731e-05,
|
| 2173 |
+
"loss": 0.5186,
|
| 2174 |
+
"step": 1785
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 3.7,
|
| 2178 |
+
"learning_rate": 3.611501813044067e-05,
|
| 2179 |
+
"loss": 0.4889,
|
| 2180 |
+
"step": 1790
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 3.71,
|
| 2184 |
+
"learning_rate": 3.8092708977579776e-05,
|
| 2185 |
+
"loss": 0.6288,
|
| 2186 |
+
"step": 1795
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 3.72,
|
| 2190 |
+
"learning_rate": 4.010253010806977e-05,
|
| 2191 |
+
"loss": 0.7543,
|
| 2192 |
+
"step": 1800
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 3.73,
|
| 2196 |
+
"learning_rate": 4.214236477799971e-05,
|
| 2197 |
+
"loss": 0.5513,
|
| 2198 |
+
"step": 1805
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 3.74,
|
| 2202 |
+
"learning_rate": 4.421006463320525e-05,
|
| 2203 |
+
"loss": 0.6435,
|
| 2204 |
+
"step": 1810
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 3.75,
|
| 2208 |
+
"learning_rate": 4.630345197188684e-05,
|
| 2209 |
+
"loss": 0.4894,
|
| 2210 |
+
"step": 1815
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 3.76,
|
| 2214 |
+
"learning_rate": 4.8420322038201655e-05,
|
| 2215 |
+
"loss": 0.4861,
|
| 2216 |
+
"step": 1820
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 3.77,
|
| 2220 |
+
"learning_rate": 5.0558445344283775e-05,
|
| 2221 |
+
"loss": 0.634,
|
| 2222 |
+
"step": 1825
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 3.78,
|
| 2226 |
+
"learning_rate": 5.2715570018359465e-05,
|
| 2227 |
+
"loss": 0.5037,
|
| 2228 |
+
"step": 1830
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 3.79,
|
| 2232 |
+
"learning_rate": 5.48894241764059e-05,
|
| 2233 |
+
"loss": 0.6078,
|
| 2234 |
+
"step": 1835
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 3.8,
|
| 2238 |
+
"learning_rate": 5.70777183149052e-05,
|
| 2239 |
+
"loss": 0.7133,
|
| 2240 |
+
"step": 1840
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 3.81,
|
| 2244 |
+
"learning_rate": 5.927814772215574e-05,
|
| 2245 |
+
"loss": 0.6985,
|
| 2246 |
+
"step": 1845
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"epoch": 3.82,
|
| 2250 |
+
"learning_rate": 6.148839490558139e-05,
|
| 2251 |
+
"loss": 0.6508,
|
| 2252 |
+
"step": 1850
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 3.83,
|
| 2256 |
+
"learning_rate": 6.370613203253432e-05,
|
| 2257 |
+
"loss": 0.4598,
|
| 2258 |
+
"step": 1855
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 3.84,
|
| 2262 |
+
"learning_rate": 6.592902338193773e-05,
|
| 2263 |
+
"loss": 0.428,
|
| 2264 |
+
"step": 1860
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 3.85,
|
| 2268 |
+
"learning_rate": 6.815472780430632e-05,
|
| 2269 |
+
"loss": 0.5706,
|
| 2270 |
+
"step": 1865
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 3.86,
|
| 2274 |
+
"learning_rate": 7.038090118741434e-05,
|
| 2275 |
+
"loss": 0.9235,
|
| 2276 |
+
"step": 1870
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 3.87,
|
| 2280 |
+
"learning_rate": 7.260519892513427e-05,
|
| 2281 |
+
"loss": 0.8443,
|
| 2282 |
+
"step": 1875
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 3.88,
|
| 2286 |
+
"learning_rate": 7.482527838676278e-05,
|
| 2287 |
+
"loss": 0.5691,
|
| 2288 |
+
"step": 1880
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 3.89,
|
| 2292 |
+
"learning_rate": 7.703880138428551e-05,
|
| 2293 |
+
"loss": 0.4075,
|
| 2294 |
+
"step": 1885
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 3.9,
|
| 2298 |
+
"learning_rate": 7.924343663496286e-05,
|
| 2299 |
+
"loss": 0.4282,
|
| 2300 |
+
"step": 1890
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 3.92,
|
| 2304 |
+
"learning_rate": 8.14368622166238e-05,
|
| 2305 |
+
"loss": 0.6324,
|
| 2306 |
+
"step": 1895
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 3.93,
|
| 2310 |
+
"learning_rate": 8.361676801313385e-05,
|
| 2311 |
+
"loss": 0.7816,
|
| 2312 |
+
"step": 1900
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 3.94,
|
| 2316 |
+
"learning_rate": 8.578085814737905e-05,
|
| 2317 |
+
"loss": 0.6076,
|
| 2318 |
+
"step": 1905
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 3.95,
|
| 2322 |
+
"learning_rate": 8.792685339932241e-05,
|
| 2323 |
+
"loss": 0.5257,
|
| 2324 |
+
"step": 1910
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"epoch": 3.96,
|
| 2328 |
+
"learning_rate": 9.005249360644818e-05,
|
| 2329 |
+
"loss": 0.6216,
|
| 2330 |
+
"step": 1915
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 3.97,
|
| 2334 |
+
"learning_rate": 9.215554004418427e-05,
|
| 2335 |
+
"loss": 0.7805,
|
| 2336 |
+
"step": 1920
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 3.98,
|
| 2340 |
+
"learning_rate": 9.423377778371423e-05,
|
| 2341 |
+
"loss": 0.7339,
|
| 2342 |
+
"step": 1925
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 3.99,
|
| 2346 |
+
"learning_rate": 9.628501802474621e-05,
|
| 2347 |
+
"loss": 0.6319,
|
| 2348 |
+
"step": 1930
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 4.0,
|
| 2352 |
+
"learning_rate": 9.830710040076315e-05,
|
| 2353 |
+
"loss": 0.5267,
|
| 2354 |
+
"step": 1935
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 4.0,
|
| 2358 |
+
"eval_loss": 0.4826279878616333,
|
| 2359 |
+
"eval_runtime": 14.7153,
|
| 2360 |
+
"eval_samples_per_second": 44.716,
|
| 2361 |
+
"eval_steps_per_second": 5.64,
|
| 2362 |
+
"step": 1936
|
| 2363 |
}
|
| 2364 |
],
|
| 2365 |
+
"max_steps": 1936,
|
| 2366 |
+
"num_train_epochs": 4,
|
| 2367 |
+
"total_flos": 2021355159552000.0,
|
| 2368 |
"trial_name": null,
|
| 2369 |
"trial_params": null
|
| 2370 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3375
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28f78ff1353e48b556a427f64b800f3a4cfdb95bdf8553eb6558c9abb866bf9e
|
| 3 |
size 3375
|