Training in progress, step 386, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 100198584
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d467248f80e50b7299be4a044d3c34650a4bb3d921ee97e1d2cbac5af5b7fc6
|
| 3 |
size 100198584
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 50675604
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea91a98de0b3ad51bc037e37c58325d581d5d9a687104d1c0db3b0ba2b46168e
|
| 3 |
size 50675604
|
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 sha256:43819302638e139ec073cc0bd87e7c492c70c5966a60959bb8d025a1d3e72519
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1827,6 +1827,888 @@
|
|
| 1827 |
"learning_rate": 1.6054254007398274e-05,
|
| 1828 |
"loss": 1.1638,
|
| 1829 |
"step": 260
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1830 |
}
|
| 1831 |
],
|
| 1832 |
"logging_steps": 1,
|
|
@@ -1846,7 +2728,7 @@
|
|
| 1846 |
"attributes": {}
|
| 1847 |
}
|
| 1848 |
},
|
| 1849 |
-
"total_flos":
|
| 1850 |
"train_batch_size": 4,
|
| 1851 |
"trial_name": null,
|
| 1852 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.4235939643347051,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 386,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1827 |
"learning_rate": 1.6054254007398274e-05,
|
| 1828 |
"loss": 1.1638,
|
| 1829 |
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.28641975308641976,
|
| 1833 |
+
"grad_norm": 0.10320425778627396,
|
| 1834 |
+
"learning_rate": 1.6029593094944513e-05,
|
| 1835 |
+
"loss": 1.1553,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.28751714677640605,
|
| 1840 |
+
"grad_norm": 0.14395684003829956,
|
| 1841 |
+
"learning_rate": 1.6004932182490752e-05,
|
| 1842 |
+
"loss": 1.1875,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.28861454046639234,
|
| 1847 |
+
"grad_norm": 0.11742989718914032,
|
| 1848 |
+
"learning_rate": 1.5980271270036994e-05,
|
| 1849 |
+
"loss": 1.1434,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.2897119341563786,
|
| 1854 |
+
"grad_norm": 0.12088185548782349,
|
| 1855 |
+
"learning_rate": 1.595561035758323e-05,
|
| 1856 |
+
"loss": 1.1365,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.2908093278463649,
|
| 1861 |
+
"grad_norm": 0.12207765877246857,
|
| 1862 |
+
"learning_rate": 1.5930949445129472e-05,
|
| 1863 |
+
"loss": 1.0191,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.29190672153635117,
|
| 1868 |
+
"grad_norm": 0.12891145050525665,
|
| 1869 |
+
"learning_rate": 1.590628853267571e-05,
|
| 1870 |
+
"loss": 1.1808,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.29300411522633746,
|
| 1875 |
+
"grad_norm": 0.12283164262771606,
|
| 1876 |
+
"learning_rate": 1.588162762022195e-05,
|
| 1877 |
+
"loss": 1.1244,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.29410150891632375,
|
| 1882 |
+
"grad_norm": 0.11997072398662567,
|
| 1883 |
+
"learning_rate": 1.585696670776819e-05,
|
| 1884 |
+
"loss": 1.1481,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.29519890260631,
|
| 1889 |
+
"grad_norm": 0.13282917439937592,
|
| 1890 |
+
"learning_rate": 1.5832305795314428e-05,
|
| 1891 |
+
"loss": 1.1222,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.2962962962962963,
|
| 1896 |
+
"grad_norm": 0.11063191294670105,
|
| 1897 |
+
"learning_rate": 1.5807644882860667e-05,
|
| 1898 |
+
"loss": 1.2377,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.29739368998628257,
|
| 1903 |
+
"grad_norm": 0.16821421682834625,
|
| 1904 |
+
"learning_rate": 1.5782983970406906e-05,
|
| 1905 |
+
"loss": 1.111,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.29849108367626886,
|
| 1910 |
+
"grad_norm": 0.1252758502960205,
|
| 1911 |
+
"learning_rate": 1.5758323057953148e-05,
|
| 1912 |
+
"loss": 1.1257,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.29958847736625516,
|
| 1917 |
+
"grad_norm": 0.11989396065473557,
|
| 1918 |
+
"learning_rate": 1.5733662145499384e-05,
|
| 1919 |
+
"loss": 1.2035,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.30068587105624145,
|
| 1924 |
+
"grad_norm": 0.122134268283844,
|
| 1925 |
+
"learning_rate": 1.5709001233045626e-05,
|
| 1926 |
+
"loss": 1.1897,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.3017832647462277,
|
| 1931 |
+
"grad_norm": 0.1424403041601181,
|
| 1932 |
+
"learning_rate": 1.5684340320591865e-05,
|
| 1933 |
+
"loss": 1.136,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.302880658436214,
|
| 1938 |
+
"grad_norm": 0.11523528397083282,
|
| 1939 |
+
"learning_rate": 1.5659679408138104e-05,
|
| 1940 |
+
"loss": 1.1721,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.30397805212620027,
|
| 1945 |
+
"grad_norm": 0.1188586950302124,
|
| 1946 |
+
"learning_rate": 1.5635018495684343e-05,
|
| 1947 |
+
"loss": 1.196,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.30507544581618656,
|
| 1952 |
+
"grad_norm": 0.09858889132738113,
|
| 1953 |
+
"learning_rate": 1.561035758323058e-05,
|
| 1954 |
+
"loss": 1.1256,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.30617283950617286,
|
| 1959 |
+
"grad_norm": 0.13993050158023834,
|
| 1960 |
+
"learning_rate": 1.558569667077682e-05,
|
| 1961 |
+
"loss": 1.142,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.30727023319615915,
|
| 1966 |
+
"grad_norm": 0.11930122971534729,
|
| 1967 |
+
"learning_rate": 1.556103575832306e-05,
|
| 1968 |
+
"loss": 1.1358,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.3083676268861454,
|
| 1973 |
+
"grad_norm": 0.12060839682817459,
|
| 1974 |
+
"learning_rate": 1.5536374845869298e-05,
|
| 1975 |
+
"loss": 1.2019,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.3094650205761317,
|
| 1980 |
+
"grad_norm": 0.11242841929197311,
|
| 1981 |
+
"learning_rate": 1.5511713933415537e-05,
|
| 1982 |
+
"loss": 1.0959,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.31056241426611797,
|
| 1987 |
+
"grad_norm": 0.12150558829307556,
|
| 1988 |
+
"learning_rate": 1.5487053020961776e-05,
|
| 1989 |
+
"loss": 1.1794,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.31165980795610426,
|
| 1994 |
+
"grad_norm": 0.13793016970157623,
|
| 1995 |
+
"learning_rate": 1.5462392108508015e-05,
|
| 1996 |
+
"loss": 1.0896,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.31275720164609055,
|
| 2001 |
+
"grad_norm": 0.1277860403060913,
|
| 2002 |
+
"learning_rate": 1.5437731196054254e-05,
|
| 2003 |
+
"loss": 1.1495,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.3138545953360768,
|
| 2008 |
+
"grad_norm": 0.12210772931575775,
|
| 2009 |
+
"learning_rate": 1.5413070283600496e-05,
|
| 2010 |
+
"loss": 1.108,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.3149519890260631,
|
| 2015 |
+
"grad_norm": 0.11492680013179779,
|
| 2016 |
+
"learning_rate": 1.538840937114673e-05,
|
| 2017 |
+
"loss": 1.1468,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.3160493827160494,
|
| 2022 |
+
"grad_norm": 0.10993566364049911,
|
| 2023 |
+
"learning_rate": 1.5363748458692974e-05,
|
| 2024 |
+
"loss": 1.0821,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.31714677640603567,
|
| 2029 |
+
"grad_norm": 0.11138095706701279,
|
| 2030 |
+
"learning_rate": 1.5339087546239213e-05,
|
| 2031 |
+
"loss": 1.1024,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.31824417009602196,
|
| 2036 |
+
"grad_norm": 0.12479337304830551,
|
| 2037 |
+
"learning_rate": 1.5314426633785452e-05,
|
| 2038 |
+
"loss": 1.1282,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.31934156378600825,
|
| 2043 |
+
"grad_norm": 0.1219845563173294,
|
| 2044 |
+
"learning_rate": 1.528976572133169e-05,
|
| 2045 |
+
"loss": 1.1404,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.3204389574759945,
|
| 2050 |
+
"grad_norm": 0.1182159036397934,
|
| 2051 |
+
"learning_rate": 1.526510480887793e-05,
|
| 2052 |
+
"loss": 1.173,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.3215363511659808,
|
| 2057 |
+
"grad_norm": 0.11335504055023193,
|
| 2058 |
+
"learning_rate": 1.5240443896424168e-05,
|
| 2059 |
+
"loss": 1.1762,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.3226337448559671,
|
| 2064 |
+
"grad_norm": 0.11732617020606995,
|
| 2065 |
+
"learning_rate": 1.5215782983970409e-05,
|
| 2066 |
+
"loss": 1.1762,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.32373113854595337,
|
| 2071 |
+
"grad_norm": 0.1295761615037918,
|
| 2072 |
+
"learning_rate": 1.5191122071516646e-05,
|
| 2073 |
+
"loss": 1.0506,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.32482853223593966,
|
| 2078 |
+
"grad_norm": 0.11054473370313644,
|
| 2079 |
+
"learning_rate": 1.5166461159062887e-05,
|
| 2080 |
+
"loss": 1.0934,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.32592592592592595,
|
| 2085 |
+
"grad_norm": 0.11660335958003998,
|
| 2086 |
+
"learning_rate": 1.5141800246609126e-05,
|
| 2087 |
+
"loss": 1.0956,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.3270233196159122,
|
| 2092 |
+
"grad_norm": 0.10848717391490936,
|
| 2093 |
+
"learning_rate": 1.5117139334155365e-05,
|
| 2094 |
+
"loss": 1.1147,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.3281207133058985,
|
| 2099 |
+
"grad_norm": 0.11791523545980453,
|
| 2100 |
+
"learning_rate": 1.5092478421701604e-05,
|
| 2101 |
+
"loss": 1.1918,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.3292181069958848,
|
| 2106 |
+
"grad_norm": 0.11942354589700699,
|
| 2107 |
+
"learning_rate": 1.5067817509247844e-05,
|
| 2108 |
+
"loss": 1.2093,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.33031550068587107,
|
| 2113 |
+
"grad_norm": 0.11650928854942322,
|
| 2114 |
+
"learning_rate": 1.5043156596794081e-05,
|
| 2115 |
+
"loss": 1.1171,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.33141289437585736,
|
| 2120 |
+
"grad_norm": 0.1390320360660553,
|
| 2121 |
+
"learning_rate": 1.5018495684340322e-05,
|
| 2122 |
+
"loss": 1.0934,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.3325102880658436,
|
| 2127 |
+
"grad_norm": 0.1456214338541031,
|
| 2128 |
+
"learning_rate": 1.4993834771886561e-05,
|
| 2129 |
+
"loss": 1.0721,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.3336076817558299,
|
| 2134 |
+
"grad_norm": 0.11829102784395218,
|
| 2135 |
+
"learning_rate": 1.49691738594328e-05,
|
| 2136 |
+
"loss": 1.1352,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.3347050754458162,
|
| 2141 |
+
"grad_norm": 0.11520489305257797,
|
| 2142 |
+
"learning_rate": 1.4944512946979039e-05,
|
| 2143 |
+
"loss": 1.1205,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.3358024691358025,
|
| 2148 |
+
"grad_norm": 0.1192188486456871,
|
| 2149 |
+
"learning_rate": 1.491985203452528e-05,
|
| 2150 |
+
"loss": 1.0658,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.33689986282578877,
|
| 2155 |
+
"grad_norm": 0.12938235700130463,
|
| 2156 |
+
"learning_rate": 1.4895191122071517e-05,
|
| 2157 |
+
"loss": 1.2173,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.33799725651577506,
|
| 2162 |
+
"grad_norm": 0.11106622219085693,
|
| 2163 |
+
"learning_rate": 1.4870530209617757e-05,
|
| 2164 |
+
"loss": 1.0982,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.3390946502057613,
|
| 2169 |
+
"grad_norm": 0.11311810463666916,
|
| 2170 |
+
"learning_rate": 1.4845869297163998e-05,
|
| 2171 |
+
"loss": 1.0987,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.3401920438957476,
|
| 2176 |
+
"grad_norm": 0.14787402749061584,
|
| 2177 |
+
"learning_rate": 1.4821208384710235e-05,
|
| 2178 |
+
"loss": 1.2011,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.3412894375857339,
|
| 2183 |
+
"grad_norm": 0.12094119936227798,
|
| 2184 |
+
"learning_rate": 1.4796547472256476e-05,
|
| 2185 |
+
"loss": 1.0972,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.34238683127572017,
|
| 2190 |
+
"grad_norm": 0.12174614518880844,
|
| 2191 |
+
"learning_rate": 1.4771886559802713e-05,
|
| 2192 |
+
"loss": 1.2196,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.34348422496570646,
|
| 2197 |
+
"grad_norm": 0.12063402682542801,
|
| 2198 |
+
"learning_rate": 1.4747225647348953e-05,
|
| 2199 |
+
"loss": 1.0724,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.34458161865569276,
|
| 2204 |
+
"grad_norm": 0.1335890144109726,
|
| 2205 |
+
"learning_rate": 1.4722564734895192e-05,
|
| 2206 |
+
"loss": 1.2439,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.345679012345679,
|
| 2211 |
+
"grad_norm": 0.120571069419384,
|
| 2212 |
+
"learning_rate": 1.4697903822441431e-05,
|
| 2213 |
+
"loss": 1.1374,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.3467764060356653,
|
| 2218 |
+
"grad_norm": 0.1246613934636116,
|
| 2219 |
+
"learning_rate": 1.467324290998767e-05,
|
| 2220 |
+
"loss": 1.1456,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.3478737997256516,
|
| 2225 |
+
"grad_norm": 0.1251581907272339,
|
| 2226 |
+
"learning_rate": 1.464858199753391e-05,
|
| 2227 |
+
"loss": 1.1829,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.34897119341563787,
|
| 2232 |
+
"grad_norm": 0.12001658231019974,
|
| 2233 |
+
"learning_rate": 1.4623921085080148e-05,
|
| 2234 |
+
"loss": 1.2,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.35006858710562416,
|
| 2239 |
+
"grad_norm": 0.11571818590164185,
|
| 2240 |
+
"learning_rate": 1.4599260172626389e-05,
|
| 2241 |
+
"loss": 1.0829,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.3511659807956104,
|
| 2246 |
+
"grad_norm": 0.11667678505182266,
|
| 2247 |
+
"learning_rate": 1.4574599260172627e-05,
|
| 2248 |
+
"loss": 1.1365,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.3522633744855967,
|
| 2253 |
+
"grad_norm": 0.13005587458610535,
|
| 2254 |
+
"learning_rate": 1.4549938347718866e-05,
|
| 2255 |
+
"loss": 1.0685,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.353360768175583,
|
| 2260 |
+
"grad_norm": 0.13064813613891602,
|
| 2261 |
+
"learning_rate": 1.4525277435265105e-05,
|
| 2262 |
+
"loss": 1.1665,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.3544581618655693,
|
| 2267 |
+
"grad_norm": 0.11439865827560425,
|
| 2268 |
+
"learning_rate": 1.4500616522811346e-05,
|
| 2269 |
+
"loss": 1.0968,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.35555555555555557,
|
| 2274 |
+
"grad_norm": 0.11633537709712982,
|
| 2275 |
+
"learning_rate": 1.4475955610357583e-05,
|
| 2276 |
+
"loss": 1.2706,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.35665294924554186,
|
| 2281 |
+
"grad_norm": 0.11462391167879105,
|
| 2282 |
+
"learning_rate": 1.4451294697903824e-05,
|
| 2283 |
+
"loss": 1.2981,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.3577503429355281,
|
| 2288 |
+
"grad_norm": 0.11040814965963364,
|
| 2289 |
+
"learning_rate": 1.4426633785450064e-05,
|
| 2290 |
+
"loss": 1.0935,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.3588477366255144,
|
| 2295 |
+
"grad_norm": 0.14056402444839478,
|
| 2296 |
+
"learning_rate": 1.4401972872996302e-05,
|
| 2297 |
+
"loss": 1.1784,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.3599451303155007,
|
| 2302 |
+
"grad_norm": 0.12376075237989426,
|
| 2303 |
+
"learning_rate": 1.4377311960542542e-05,
|
| 2304 |
+
"loss": 1.1781,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.361042524005487,
|
| 2309 |
+
"grad_norm": 0.13468991219997406,
|
| 2310 |
+
"learning_rate": 1.4352651048088781e-05,
|
| 2311 |
+
"loss": 1.151,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.36213991769547327,
|
| 2316 |
+
"grad_norm": 0.11974407732486725,
|
| 2317 |
+
"learning_rate": 1.432799013563502e-05,
|
| 2318 |
+
"loss": 1.0996,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.36323731138545956,
|
| 2323 |
+
"grad_norm": 0.13505803048610687,
|
| 2324 |
+
"learning_rate": 1.4303329223181259e-05,
|
| 2325 |
+
"loss": 1.0323,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.3643347050754458,
|
| 2330 |
+
"grad_norm": 0.13089002668857574,
|
| 2331 |
+
"learning_rate": 1.4278668310727498e-05,
|
| 2332 |
+
"loss": 1.0352,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.3654320987654321,
|
| 2337 |
+
"grad_norm": 0.13586758077144623,
|
| 2338 |
+
"learning_rate": 1.4254007398273737e-05,
|
| 2339 |
+
"loss": 1.1415,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.3665294924554184,
|
| 2344 |
+
"grad_norm": 0.13658414781093597,
|
| 2345 |
+
"learning_rate": 1.4229346485819977e-05,
|
| 2346 |
+
"loss": 1.2563,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.3676268861454047,
|
| 2351 |
+
"grad_norm": 0.13994932174682617,
|
| 2352 |
+
"learning_rate": 1.4204685573366214e-05,
|
| 2353 |
+
"loss": 1.1559,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.36872427983539097,
|
| 2358 |
+
"grad_norm": 0.11985601484775543,
|
| 2359 |
+
"learning_rate": 1.4180024660912455e-05,
|
| 2360 |
+
"loss": 1.1639,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.3698216735253772,
|
| 2365 |
+
"grad_norm": 0.15504884719848633,
|
| 2366 |
+
"learning_rate": 1.4155363748458694e-05,
|
| 2367 |
+
"loss": 1.0932,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.3709190672153635,
|
| 2372 |
+
"grad_norm": 0.11062606424093246,
|
| 2373 |
+
"learning_rate": 1.4130702836004933e-05,
|
| 2374 |
+
"loss": 1.1327,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.3720164609053498,
|
| 2379 |
+
"grad_norm": 0.13901466131210327,
|
| 2380 |
+
"learning_rate": 1.4106041923551172e-05,
|
| 2381 |
+
"loss": 1.1924,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.3731138545953361,
|
| 2386 |
+
"grad_norm": 0.1262989044189453,
|
| 2387 |
+
"learning_rate": 1.4081381011097412e-05,
|
| 2388 |
+
"loss": 1.1411,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.3742112482853224,
|
| 2393 |
+
"grad_norm": 0.1387164443731308,
|
| 2394 |
+
"learning_rate": 1.405672009864365e-05,
|
| 2395 |
+
"loss": 1.116,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.37530864197530867,
|
| 2400 |
+
"grad_norm": 0.13789041340351105,
|
| 2401 |
+
"learning_rate": 1.403205918618989e-05,
|
| 2402 |
+
"loss": 1.1003,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.3764060356652949,
|
| 2407 |
+
"grad_norm": 0.12197256088256836,
|
| 2408 |
+
"learning_rate": 1.400739827373613e-05,
|
| 2409 |
+
"loss": 1.1363,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.3775034293552812,
|
| 2414 |
+
"grad_norm": 0.11622656881809235,
|
| 2415 |
+
"learning_rate": 1.3982737361282368e-05,
|
| 2416 |
+
"loss": 1.2004,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.3786008230452675,
|
| 2421 |
+
"grad_norm": 0.13322965800762177,
|
| 2422 |
+
"learning_rate": 1.3958076448828609e-05,
|
| 2423 |
+
"loss": 1.0868,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.3796982167352538,
|
| 2428 |
+
"grad_norm": 0.13613948225975037,
|
| 2429 |
+
"learning_rate": 1.3933415536374848e-05,
|
| 2430 |
+
"loss": 1.0893,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.38079561042524007,
|
| 2435 |
+
"grad_norm": 0.12749332189559937,
|
| 2436 |
+
"learning_rate": 1.3908754623921086e-05,
|
| 2437 |
+
"loss": 1.1242,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.38189300411522636,
|
| 2442 |
+
"grad_norm": 0.13489745557308197,
|
| 2443 |
+
"learning_rate": 1.3884093711467325e-05,
|
| 2444 |
+
"loss": 1.0463,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.3829903978052126,
|
| 2449 |
+
"grad_norm": 0.13293969631195068,
|
| 2450 |
+
"learning_rate": 1.3859432799013564e-05,
|
| 2451 |
+
"loss": 1.0682,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.3840877914951989,
|
| 2456 |
+
"grad_norm": 0.13386596739292145,
|
| 2457 |
+
"learning_rate": 1.3834771886559803e-05,
|
| 2458 |
+
"loss": 1.1216,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.3851851851851852,
|
| 2463 |
+
"grad_norm": 0.1273600161075592,
|
| 2464 |
+
"learning_rate": 1.3810110974106044e-05,
|
| 2465 |
+
"loss": 1.1638,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.3862825788751715,
|
| 2470 |
+
"grad_norm": 0.113369882106781,
|
| 2471 |
+
"learning_rate": 1.3785450061652281e-05,
|
| 2472 |
+
"loss": 1.0776,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.38737997256515777,
|
| 2477 |
+
"grad_norm": 0.13265594840049744,
|
| 2478 |
+
"learning_rate": 1.3760789149198522e-05,
|
| 2479 |
+
"loss": 1.1413,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.388477366255144,
|
| 2484 |
+
"grad_norm": 0.10907372832298279,
|
| 2485 |
+
"learning_rate": 1.373612823674476e-05,
|
| 2486 |
+
"loss": 1.1685,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.3895747599451303,
|
| 2491 |
+
"grad_norm": 0.14246472716331482,
|
| 2492 |
+
"learning_rate": 1.3711467324291e-05,
|
| 2493 |
+
"loss": 1.1737,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.3906721536351166,
|
| 2498 |
+
"grad_norm": 0.11236773431301117,
|
| 2499 |
+
"learning_rate": 1.3686806411837238e-05,
|
| 2500 |
+
"loss": 1.2439,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.3917695473251029,
|
| 2505 |
+
"grad_norm": 0.14173471927642822,
|
| 2506 |
+
"learning_rate": 1.3662145499383479e-05,
|
| 2507 |
+
"loss": 1.0831,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.3928669410150892,
|
| 2512 |
+
"grad_norm": 0.13366690278053284,
|
| 2513 |
+
"learning_rate": 1.3637484586929716e-05,
|
| 2514 |
+
"loss": 1.1333,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.39396433470507547,
|
| 2519 |
+
"grad_norm": 0.14148059487342834,
|
| 2520 |
+
"learning_rate": 1.3612823674475957e-05,
|
| 2521 |
+
"loss": 1.1519,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.3950617283950617,
|
| 2526 |
+
"grad_norm": 0.11044170707464218,
|
| 2527 |
+
"learning_rate": 1.3588162762022197e-05,
|
| 2528 |
+
"loss": 1.255,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.396159122085048,
|
| 2533 |
+
"grad_norm": 0.11801745742559433,
|
| 2534 |
+
"learning_rate": 1.3563501849568435e-05,
|
| 2535 |
+
"loss": 1.0922,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.3972565157750343,
|
| 2540 |
+
"grad_norm": 0.11835184693336487,
|
| 2541 |
+
"learning_rate": 1.3538840937114675e-05,
|
| 2542 |
+
"loss": 1.1125,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.3983539094650206,
|
| 2547 |
+
"grad_norm": 0.1362667679786682,
|
| 2548 |
+
"learning_rate": 1.3514180024660914e-05,
|
| 2549 |
+
"loss": 1.1118,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.3994513031550069,
|
| 2554 |
+
"grad_norm": 0.1164252832531929,
|
| 2555 |
+
"learning_rate": 1.3489519112207153e-05,
|
| 2556 |
+
"loss": 1.0671,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.40054869684499317,
|
| 2561 |
+
"grad_norm": 0.12106901407241821,
|
| 2562 |
+
"learning_rate": 1.3464858199753392e-05,
|
| 2563 |
+
"loss": 1.1901,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.4016460905349794,
|
| 2568 |
+
"grad_norm": 0.1594018191099167,
|
| 2569 |
+
"learning_rate": 1.3440197287299632e-05,
|
| 2570 |
+
"loss": 1.0512,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.4027434842249657,
|
| 2575 |
+
"grad_norm": 0.11015798151493073,
|
| 2576 |
+
"learning_rate": 1.341553637484587e-05,
|
| 2577 |
+
"loss": 1.1859,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.403840877914952,
|
| 2582 |
+
"grad_norm": 0.1140451431274414,
|
| 2583 |
+
"learning_rate": 1.339087546239211e-05,
|
| 2584 |
+
"loss": 1.093,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.4049382716049383,
|
| 2589 |
+
"grad_norm": 0.1487288475036621,
|
| 2590 |
+
"learning_rate": 1.3366214549938348e-05,
|
| 2591 |
+
"loss": 1.2052,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.4060356652949246,
|
| 2596 |
+
"grad_norm": 0.12370628118515015,
|
| 2597 |
+
"learning_rate": 1.3341553637484588e-05,
|
| 2598 |
+
"loss": 1.0985,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.4071330589849108,
|
| 2603 |
+
"grad_norm": 0.11730250716209412,
|
| 2604 |
+
"learning_rate": 1.3316892725030827e-05,
|
| 2605 |
+
"loss": 1.082,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.4082304526748971,
|
| 2610 |
+
"grad_norm": 0.1203538104891777,
|
| 2611 |
+
"learning_rate": 1.3292231812577066e-05,
|
| 2612 |
+
"loss": 1.0728,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.4093278463648834,
|
| 2617 |
+
"grad_norm": 0.11578691750764847,
|
| 2618 |
+
"learning_rate": 1.3267570900123305e-05,
|
| 2619 |
+
"loss": 1.1351,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.4104252400548697,
|
| 2624 |
+
"grad_norm": 0.1101141944527626,
|
| 2625 |
+
"learning_rate": 1.3242909987669545e-05,
|
| 2626 |
+
"loss": 1.1311,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.411522633744856,
|
| 2631 |
+
"grad_norm": 0.11373890191316605,
|
| 2632 |
+
"learning_rate": 1.3218249075215783e-05,
|
| 2633 |
+
"loss": 1.0773,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.4126200274348423,
|
| 2638 |
+
"grad_norm": 0.14648625254631042,
|
| 2639 |
+
"learning_rate": 1.3193588162762023e-05,
|
| 2640 |
+
"loss": 1.2263,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.4137174211248285,
|
| 2645 |
+
"grad_norm": 0.11010008305311203,
|
| 2646 |
+
"learning_rate": 1.3168927250308264e-05,
|
| 2647 |
+
"loss": 1.1445,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.4148148148148148,
|
| 2652 |
+
"grad_norm": 0.12386184185743332,
|
| 2653 |
+
"learning_rate": 1.3144266337854501e-05,
|
| 2654 |
+
"loss": 1.0454,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.4159122085048011,
|
| 2659 |
+
"grad_norm": 0.11810237169265747,
|
| 2660 |
+
"learning_rate": 1.3119605425400742e-05,
|
| 2661 |
+
"loss": 1.1877,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.4170096021947874,
|
| 2666 |
+
"grad_norm": 0.1291743367910385,
|
| 2667 |
+
"learning_rate": 1.309494451294698e-05,
|
| 2668 |
+
"loss": 1.2161,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.4181069958847737,
|
| 2673 |
+
"grad_norm": 0.115330770611763,
|
| 2674 |
+
"learning_rate": 1.307028360049322e-05,
|
| 2675 |
+
"loss": 1.1117,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.41920438957476,
|
| 2680 |
+
"grad_norm": 0.15565675497055054,
|
| 2681 |
+
"learning_rate": 1.3045622688039458e-05,
|
| 2682 |
+
"loss": 1.1049,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.4203017832647462,
|
| 2687 |
+
"grad_norm": 0.17110465466976166,
|
| 2688 |
+
"learning_rate": 1.3020961775585699e-05,
|
| 2689 |
+
"loss": 1.1047,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.4213991769547325,
|
| 2694 |
+
"grad_norm": 0.12324660271406174,
|
| 2695 |
+
"learning_rate": 1.2996300863131936e-05,
|
| 2696 |
+
"loss": 1.091,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.4224965706447188,
|
| 2701 |
+
"grad_norm": 0.14864134788513184,
|
| 2702 |
+
"learning_rate": 1.2971639950678177e-05,
|
| 2703 |
+
"loss": 1.1008,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.4235939643347051,
|
| 2708 |
+
"grad_norm": 0.15296033024787903,
|
| 2709 |
+
"learning_rate": 1.2946979038224414e-05,
|
| 2710 |
+
"loss": 1.1242,
|
| 2711 |
+
"step": 386
|
| 2712 |
}
|
| 2713 |
],
|
| 2714 |
"logging_steps": 1,
|
|
|
|
| 2728 |
"attributes": {}
|
| 2729 |
}
|
| 2730 |
},
|
| 2731 |
+
"total_flos": 4.0076346624661094e+17,
|
| 2732 |
"train_batch_size": 4,
|
| 2733 |
"trial_name": null,
|
| 2734 |
"trial_params": null
|