Training in progress, step 372, 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 144805440
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4db2aaaf0e71f4163eb1948613e8b578254a0de5a98794daab0fd666ff0e9335
|
| 3 |
size 144805440
|
last-checkpoint/optimizer.pt
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 sha256:97e66fb0be2050837da15f447248f80a1c314ca1bc2a805d9029af38817476ae
|
| 3 |
+
size 74292308
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:956eaecea1513e07a98b3b792863d7c1c440c5da69923f2e87c397dfc0da01f4
|
| 3 |
size 14244
|
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:d6484c651d3d8bf75888a89e2001dae4da70b271bc6cafa91a994f199bf79e17
|
| 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,
|
|
@@ -1743,6 +1743,874 @@
|
|
| 1743 |
"learning_rate": 5.079976035714976e-05,
|
| 1744 |
"loss": 0.7198,
|
| 1745 |
"step": 248
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1746 |
}
|
| 1747 |
],
|
| 1748 |
"logging_steps": 1,
|
|
@@ -1762,7 +2630,7 @@
|
|
| 1762 |
"attributes": {}
|
| 1763 |
}
|
| 1764 |
},
|
| 1765 |
-
"total_flos":
|
| 1766 |
"train_batch_size": 4,
|
| 1767 |
"trial_name": null,
|
| 1768 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7503782148260212,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 372,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1743 |
"learning_rate": 5.079976035714976e-05,
|
| 1744 |
"loss": 0.7198,
|
| 1745 |
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5022692889561271,
|
| 1749 |
+
"grad_norm": 0.6312721371650696,
|
| 1750 |
+
"learning_rate": 5.047986931116205e-05,
|
| 1751 |
+
"loss": 0.7787,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5042864346949067,
|
| 1756 |
+
"grad_norm": 0.6478500962257385,
|
| 1757 |
+
"learning_rate": 5.015995861989287e-05,
|
| 1758 |
+
"loss": 0.6478,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5063035804336863,
|
| 1763 |
+
"grad_norm": 0.370857834815979,
|
| 1764 |
+
"learning_rate": 4.984004138010715e-05,
|
| 1765 |
+
"loss": 0.4671,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5083207261724659,
|
| 1770 |
+
"grad_norm": 0.432366281747818,
|
| 1771 |
+
"learning_rate": 4.952013068883795e-05,
|
| 1772 |
+
"loss": 0.6026,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5103378719112456,
|
| 1777 |
+
"grad_norm": 0.4328523576259613,
|
| 1778 |
+
"learning_rate": 4.920023964285025e-05,
|
| 1779 |
+
"loss": 0.6082,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.5123550176500252,
|
| 1784 |
+
"grad_norm": 0.40599900484085083,
|
| 1785 |
+
"learning_rate": 4.8880381338104777e-05,
|
| 1786 |
+
"loss": 0.4708,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.5143721633888049,
|
| 1791 |
+
"grad_norm": 0.43550002574920654,
|
| 1792 |
+
"learning_rate": 4.8560568869221805e-05,
|
| 1793 |
+
"loss": 0.5816,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.5163893091275845,
|
| 1798 |
+
"grad_norm": 0.40016648173332214,
|
| 1799 |
+
"learning_rate": 4.8240815328945226e-05,
|
| 1800 |
+
"loss": 0.5691,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.518406454866364,
|
| 1805 |
+
"grad_norm": 0.437029629945755,
|
| 1806 |
+
"learning_rate": 4.7921133807606424e-05,
|
| 1807 |
+
"loss": 0.5555,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.5204236006051437,
|
| 1812 |
+
"grad_norm": 0.39981645345687866,
|
| 1813 |
+
"learning_rate": 4.760153739258843e-05,
|
| 1814 |
+
"loss": 0.5786,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.5224407463439233,
|
| 1819 |
+
"grad_norm": 0.4225994646549225,
|
| 1820 |
+
"learning_rate": 4.728203916779009e-05,
|
| 1821 |
+
"loss": 0.5744,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.524457892082703,
|
| 1826 |
+
"grad_norm": 0.4146437346935272,
|
| 1827 |
+
"learning_rate": 4.69626522130905e-05,
|
| 1828 |
+
"loss": 0.5433,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.5264750378214826,
|
| 1833 |
+
"grad_norm": 0.41293078660964966,
|
| 1834 |
+
"learning_rate": 4.6643389603813486e-05,
|
| 1835 |
+
"loss": 0.5461,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.5284921835602622,
|
| 1840 |
+
"grad_norm": 0.4155355989933014,
|
| 1841 |
+
"learning_rate": 4.632426441019227e-05,
|
| 1842 |
+
"loss": 0.5554,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.5305093292990418,
|
| 1847 |
+
"grad_norm": 0.4077325761318207,
|
| 1848 |
+
"learning_rate": 4.600528969683448e-05,
|
| 1849 |
+
"loss": 0.4872,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.5325264750378215,
|
| 1854 |
+
"grad_norm": 0.4461796283721924,
|
| 1855 |
+
"learning_rate": 4.568647852218725e-05,
|
| 1856 |
+
"loss": 0.5266,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.5345436207766011,
|
| 1861 |
+
"grad_norm": 0.40012630820274353,
|
| 1862 |
+
"learning_rate": 4.5367843938002694e-05,
|
| 1863 |
+
"loss": 0.5459,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.5365607665153808,
|
| 1868 |
+
"grad_norm": 0.45490312576293945,
|
| 1869 |
+
"learning_rate": 4.504939898880339e-05,
|
| 1870 |
+
"loss": 0.6255,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.5385779122541604,
|
| 1875 |
+
"grad_norm": 0.41126129031181335,
|
| 1876 |
+
"learning_rate": 4.473115671134859e-05,
|
| 1877 |
+
"loss": 0.5786,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.5405950579929399,
|
| 1882 |
+
"grad_norm": 0.3975431025028229,
|
| 1883 |
+
"learning_rate": 4.441313013410039e-05,
|
| 1884 |
+
"loss": 0.5707,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.5426122037317196,
|
| 1889 |
+
"grad_norm": 0.40477657318115234,
|
| 1890 |
+
"learning_rate": 4.409533227669033e-05,
|
| 1891 |
+
"loss": 0.5079,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.5446293494704992,
|
| 1896 |
+
"grad_norm": 0.4053346514701843,
|
| 1897 |
+
"learning_rate": 4.377777614938647e-05,
|
| 1898 |
+
"loss": 0.6502,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.5466464952092789,
|
| 1903 |
+
"grad_norm": 0.4417206346988678,
|
| 1904 |
+
"learning_rate": 4.3460474752560724e-05,
|
| 1905 |
+
"loss": 0.6215,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.5486636409480585,
|
| 1910 |
+
"grad_norm": 0.4087637960910797,
|
| 1911 |
+
"learning_rate": 4.314344107615665e-05,
|
| 1912 |
+
"loss": 0.5612,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.5506807866868382,
|
| 1917 |
+
"grad_norm": 0.3786230683326721,
|
| 1918 |
+
"learning_rate": 4.282668809915758e-05,
|
| 1919 |
+
"loss": 0.5299,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.5526979324256177,
|
| 1924 |
+
"grad_norm": 0.3989951014518738,
|
| 1925 |
+
"learning_rate": 4.251022878905543e-05,
|
| 1926 |
+
"loss": 0.5832,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.5547150781643974,
|
| 1931 |
+
"grad_norm": 0.45247653126716614,
|
| 1932 |
+
"learning_rate": 4.219407610131971e-05,
|
| 1933 |
+
"loss": 0.5932,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.556732223903177,
|
| 1938 |
+
"grad_norm": 0.40874481201171875,
|
| 1939 |
+
"learning_rate": 4.187824297886715e-05,
|
| 1940 |
+
"loss": 0.6128,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.5587493696419567,
|
| 1945 |
+
"grad_norm": 0.42016932368278503,
|
| 1946 |
+
"learning_rate": 4.156274235153189e-05,
|
| 1947 |
+
"loss": 0.536,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.5607665153807363,
|
| 1952 |
+
"grad_norm": 0.4106806814670563,
|
| 1953 |
+
"learning_rate": 4.1247587135536126e-05,
|
| 1954 |
+
"loss": 0.587,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.5627836611195158,
|
| 1959 |
+
"grad_norm": 0.4463154077529907,
|
| 1960 |
+
"learning_rate": 4.0932790232961235e-05,
|
| 1961 |
+
"loss": 0.5121,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.5648008068582955,
|
| 1966 |
+
"grad_norm": 0.45139235258102417,
|
| 1967 |
+
"learning_rate": 4.0618364531219775e-05,
|
| 1968 |
+
"loss": 0.5219,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.5668179525970751,
|
| 1973 |
+
"grad_norm": 0.45914503931999207,
|
| 1974 |
+
"learning_rate": 4.030432290252771e-05,
|
| 1975 |
+
"loss": 0.6114,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.5688350983358548,
|
| 1980 |
+
"grad_norm": 0.43565309047698975,
|
| 1981 |
+
"learning_rate": 3.999067820337757e-05,
|
| 1982 |
+
"loss": 0.577,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.5708522440746344,
|
| 1987 |
+
"grad_norm": 0.4499742090702057,
|
| 1988 |
+
"learning_rate": 3.967744327401197e-05,
|
| 1989 |
+
"loss": 0.537,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.5728693898134141,
|
| 1994 |
+
"grad_norm": 0.44511550664901733,
|
| 1995 |
+
"learning_rate": 3.936463093789813e-05,
|
| 1996 |
+
"loss": 0.6069,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.5748865355521936,
|
| 2001 |
+
"grad_norm": 0.4442926347255707,
|
| 2002 |
+
"learning_rate": 3.9052254001202746e-05,
|
| 2003 |
+
"loss": 0.601,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.5769036812909732,
|
| 2008 |
+
"grad_norm": 0.4780077636241913,
|
| 2009 |
+
"learning_rate": 3.8740325252267785e-05,
|
| 2010 |
+
"loss": 0.5617,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.5789208270297529,
|
| 2015 |
+
"grad_norm": 0.4674513638019562,
|
| 2016 |
+
"learning_rate": 3.842885746108693e-05,
|
| 2017 |
+
"loss": 0.6245,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.5809379727685325,
|
| 2022 |
+
"grad_norm": 0.49114686250686646,
|
| 2023 |
+
"learning_rate": 3.811786337878284e-05,
|
| 2024 |
+
"loss": 0.6625,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.5829551185073122,
|
| 2029 |
+
"grad_norm": 0.5057556629180908,
|
| 2030 |
+
"learning_rate": 3.780735573708508e-05,
|
| 2031 |
+
"loss": 0.6384,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.5849722642460918,
|
| 2036 |
+
"grad_norm": 0.4727246165275574,
|
| 2037 |
+
"learning_rate": 3.7497347247808846e-05,
|
| 2038 |
+
"loss": 0.7091,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.5869894099848714,
|
| 2043 |
+
"grad_norm": 0.5270060300827026,
|
| 2044 |
+
"learning_rate": 3.718785060233471e-05,
|
| 2045 |
+
"loss": 0.673,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.589006555723651,
|
| 2050 |
+
"grad_norm": 0.5089999437332153,
|
| 2051 |
+
"learning_rate": 3.687887847108894e-05,
|
| 2052 |
+
"loss": 0.6378,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.5910237014624307,
|
| 2057 |
+
"grad_norm": 0.5260120630264282,
|
| 2058 |
+
"learning_rate": 3.657044350302479e-05,
|
| 2059 |
+
"loss": 0.6278,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.5930408472012103,
|
| 2064 |
+
"grad_norm": 0.5153591632843018,
|
| 2065 |
+
"learning_rate": 3.6262558325104695e-05,
|
| 2066 |
+
"loss": 0.6463,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.59505799293999,
|
| 2071 |
+
"grad_norm": 0.5067986845970154,
|
| 2072 |
+
"learning_rate": 3.595523554178336e-05,
|
| 2073 |
+
"loss": 0.6946,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.5970751386787695,
|
| 2078 |
+
"grad_norm": 0.5289713740348816,
|
| 2079 |
+
"learning_rate": 3.564848773449172e-05,
|
| 2080 |
+
"loss": 0.6358,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.5990922844175491,
|
| 2085 |
+
"grad_norm": 0.5743799805641174,
|
| 2086 |
+
"learning_rate": 3.5342327461121805e-05,
|
| 2087 |
+
"loss": 0.6664,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.6011094301563288,
|
| 2092 |
+
"grad_norm": 0.6473139524459839,
|
| 2093 |
+
"learning_rate": 3.503676725551278e-05,
|
| 2094 |
+
"loss": 0.6851,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.6031265758951084,
|
| 2099 |
+
"grad_norm": 0.6861950755119324,
|
| 2100 |
+
"learning_rate": 3.473181962693773e-05,
|
| 2101 |
+
"loss": 0.6171,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.6051437216338881,
|
| 2106 |
+
"grad_norm": 0.6407880783081055,
|
| 2107 |
+
"learning_rate": 3.442749705959152e-05,
|
| 2108 |
+
"loss": 0.5958,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.6071608673726677,
|
| 2113 |
+
"grad_norm": 0.3541804552078247,
|
| 2114 |
+
"learning_rate": 3.412381201207979e-05,
|
| 2115 |
+
"loss": 0.5327,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.6091780131114473,
|
| 2120 |
+
"grad_norm": 0.3857956528663635,
|
| 2121 |
+
"learning_rate": 3.3820776916908857e-05,
|
| 2122 |
+
"loss": 0.5675,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.6111951588502269,
|
| 2127 |
+
"grad_norm": 0.41454771161079407,
|
| 2128 |
+
"learning_rate": 3.351840417997679e-05,
|
| 2129 |
+
"loss": 0.5419,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.6132123045890066,
|
| 2134 |
+
"grad_norm": 0.41526558995246887,
|
| 2135 |
+
"learning_rate": 3.321670618006543e-05,
|
| 2136 |
+
"loss": 0.5429,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.6152294503277862,
|
| 2141 |
+
"grad_norm": 0.4046734571456909,
|
| 2142 |
+
"learning_rate": 3.291569526833372e-05,
|
| 2143 |
+
"loss": 0.5813,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.6172465960665658,
|
| 2148 |
+
"grad_norm": 0.43739229440689087,
|
| 2149 |
+
"learning_rate": 3.2615383767812056e-05,
|
| 2150 |
+
"loss": 0.6181,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.6192637418053455,
|
| 2155 |
+
"grad_norm": 0.42888522148132324,
|
| 2156 |
+
"learning_rate": 3.231578397289772e-05,
|
| 2157 |
+
"loss": 0.492,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.621280887544125,
|
| 2162 |
+
"grad_norm": 0.4459590017795563,
|
| 2163 |
+
"learning_rate": 3.2016908148851624e-05,
|
| 2164 |
+
"loss": 0.5323,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.6232980332829047,
|
| 2169 |
+
"grad_norm": 0.440335750579834,
|
| 2170 |
+
"learning_rate": 3.1718768531296196e-05,
|
| 2171 |
+
"loss": 0.5439,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.6253151790216843,
|
| 2176 |
+
"grad_norm": 0.422626793384552,
|
| 2177 |
+
"learning_rate": 3.142137732571437e-05,
|
| 2178 |
+
"loss": 0.538,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.627332324760464,
|
| 2183 |
+
"grad_norm": 0.3488753139972687,
|
| 2184 |
+
"learning_rate": 3.112474670695008e-05,
|
| 2185 |
+
"loss": 0.4664,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.6293494704992436,
|
| 2190 |
+
"grad_norm": 0.4092547297477722,
|
| 2191 |
+
"learning_rate": 3.0828888818709656e-05,
|
| 2192 |
+
"loss": 0.4762,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.6313666162380231,
|
| 2197 |
+
"grad_norm": 0.3810945749282837,
|
| 2198 |
+
"learning_rate": 3.053381577306481e-05,
|
| 2199 |
+
"loss": 0.4807,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.6333837619768028,
|
| 2204 |
+
"grad_norm": 0.3901941180229187,
|
| 2205 |
+
"learning_rate": 3.0239539649956665e-05,
|
| 2206 |
+
"loss": 0.505,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.6354009077155824,
|
| 2211 |
+
"grad_norm": 0.3677188456058502,
|
| 2212 |
+
"learning_rate": 2.9946072496701334e-05,
|
| 2213 |
+
"loss": 0.5334,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.6374180534543621,
|
| 2218 |
+
"grad_norm": 0.3920693099498749,
|
| 2219 |
+
"learning_rate": 2.9653426327496647e-05,
|
| 2220 |
+
"loss": 0.5632,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.6394351991931417,
|
| 2225 |
+
"grad_norm": 0.3857486844062805,
|
| 2226 |
+
"learning_rate": 2.9361613122930304e-05,
|
| 2227 |
+
"loss": 0.5151,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.6414523449319214,
|
| 2232 |
+
"grad_norm": 0.4166855812072754,
|
| 2233 |
+
"learning_rate": 2.9070644829489434e-05,
|
| 2234 |
+
"loss": 0.534,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.6434694906707009,
|
| 2239 |
+
"grad_norm": 0.40017688274383545,
|
| 2240 |
+
"learning_rate": 2.8780533359071504e-05,
|
| 2241 |
+
"loss": 0.4967,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.6454866364094806,
|
| 2246 |
+
"grad_norm": 0.45257654786109924,
|
| 2247 |
+
"learning_rate": 2.8491290588496668e-05,
|
| 2248 |
+
"loss": 0.5979,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.6475037821482602,
|
| 2253 |
+
"grad_norm": 0.4371342360973358,
|
| 2254 |
+
"learning_rate": 2.820292835902148e-05,
|
| 2255 |
+
"loss": 0.5411,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.6495209278870399,
|
| 2260 |
+
"grad_norm": 0.4240793287754059,
|
| 2261 |
+
"learning_rate": 2.7915458475854283e-05,
|
| 2262 |
+
"loss": 0.5436,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.6515380736258195,
|
| 2267 |
+
"grad_norm": 0.41990602016448975,
|
| 2268 |
+
"learning_rate": 2.762889270767175e-05,
|
| 2269 |
+
"loss": 0.523,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.653555219364599,
|
| 2274 |
+
"grad_norm": 0.45127952098846436,
|
| 2275 |
+
"learning_rate": 2.7343242786137168e-05,
|
| 2276 |
+
"loss": 0.5283,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.6555723651033787,
|
| 2281 |
+
"grad_norm": 0.4676419496536255,
|
| 2282 |
+
"learning_rate": 2.7058520405420123e-05,
|
| 2283 |
+
"loss": 0.5815,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.6575895108421583,
|
| 2288 |
+
"grad_norm": 0.42602765560150146,
|
| 2289 |
+
"learning_rate": 2.677473722171786e-05,
|
| 2290 |
+
"loss": 0.5368,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.659606656580938,
|
| 2295 |
+
"grad_norm": 0.43028366565704346,
|
| 2296 |
+
"learning_rate": 2.649190485277792e-05,
|
| 2297 |
+
"loss": 0.5435,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.6616238023197176,
|
| 2302 |
+
"grad_norm": 0.4180251955986023,
|
| 2303 |
+
"learning_rate": 2.621003487742264e-05,
|
| 2304 |
+
"loss": 0.529,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.6636409480584973,
|
| 2309 |
+
"grad_norm": 0.4283643364906311,
|
| 2310 |
+
"learning_rate": 2.5929138835075152e-05,
|
| 2311 |
+
"loss": 0.6257,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.6656580937972768,
|
| 2316 |
+
"grad_norm": 0.4437110424041748,
|
| 2317 |
+
"learning_rate": 2.564922822528686e-05,
|
| 2318 |
+
"loss": 0.5602,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.6676752395360565,
|
| 2323 |
+
"grad_norm": 0.464622437953949,
|
| 2324 |
+
"learning_rate": 2.5370314507266756e-05,
|
| 2325 |
+
"loss": 0.5834,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.6696923852748361,
|
| 2330 |
+
"grad_norm": 0.4433290958404541,
|
| 2331 |
+
"learning_rate": 2.5092409099412227e-05,
|
| 2332 |
+
"loss": 0.5993,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.6717095310136157,
|
| 2337 |
+
"grad_norm": 0.4574563503265381,
|
| 2338 |
+
"learning_rate": 2.4815523378841726e-05,
|
| 2339 |
+
"loss": 0.6211,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.6737266767523954,
|
| 2344 |
+
"grad_norm": 0.43739160895347595,
|
| 2345 |
+
"learning_rate": 2.4539668680928784e-05,
|
| 2346 |
+
"loss": 0.4989,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.675743822491175,
|
| 2351 |
+
"grad_norm": 0.4508601129055023,
|
| 2352 |
+
"learning_rate": 2.4264856298838213e-05,
|
| 2353 |
+
"loss": 0.6101,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.6777609682299546,
|
| 2358 |
+
"grad_norm": 0.4525294005870819,
|
| 2359 |
+
"learning_rate": 2.399109748306355e-05,
|
| 2360 |
+
"loss": 0.5605,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.6797781139687342,
|
| 2365 |
+
"grad_norm": 0.444223552942276,
|
| 2366 |
+
"learning_rate": 2.371840344096665e-05,
|
| 2367 |
+
"loss": 0.5952,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.6817952597075139,
|
| 2372 |
+
"grad_norm": 0.4897475242614746,
|
| 2373 |
+
"learning_rate": 2.3446785336318754e-05,
|
| 2374 |
+
"loss": 0.5829,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.6838124054462935,
|
| 2379 |
+
"grad_norm": 0.48455387353897095,
|
| 2380 |
+
"learning_rate": 2.317625428884348e-05,
|
| 2381 |
+
"loss": 0.6363,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.6858295511850732,
|
| 2386 |
+
"grad_norm": 0.5362589955329895,
|
| 2387 |
+
"learning_rate": 2.290682137376169e-05,
|
| 2388 |
+
"loss": 0.6597,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.6878466969238527,
|
| 2393 |
+
"grad_norm": 0.518014669418335,
|
| 2394 |
+
"learning_rate": 2.263849762133788e-05,
|
| 2395 |
+
"loss": 0.5839,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.6898638426626323,
|
| 2400 |
+
"grad_norm": 0.504257082939148,
|
| 2401 |
+
"learning_rate": 2.237129401642887e-05,
|
| 2402 |
+
"loss": 0.6091,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.691880988401412,
|
| 2407 |
+
"grad_norm": 0.5075199007987976,
|
| 2408 |
+
"learning_rate": 2.2105221498033862e-05,
|
| 2409 |
+
"loss": 0.5976,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.6938981341401916,
|
| 2414 |
+
"grad_norm": 0.5348207354545593,
|
| 2415 |
+
"learning_rate": 2.1840290958846816e-05,
|
| 2416 |
+
"loss": 0.6331,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.6959152798789713,
|
| 2421 |
+
"grad_norm": 0.6212106347084045,
|
| 2422 |
+
"learning_rate": 2.157651324481033e-05,
|
| 2423 |
+
"loss": 0.7058,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.6979324256177509,
|
| 2428 |
+
"grad_norm": 0.5681262612342834,
|
| 2429 |
+
"learning_rate": 2.131389915467173e-05,
|
| 2430 |
+
"loss": 0.6257,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.6999495713565305,
|
| 2435 |
+
"grad_norm": 0.5732640624046326,
|
| 2436 |
+
"learning_rate": 2.1052459439541005e-05,
|
| 2437 |
+
"loss": 0.6812,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7019667170953101,
|
| 2442 |
+
"grad_norm": 0.5906322002410889,
|
| 2443 |
+
"learning_rate": 2.0792204802450515e-05,
|
| 2444 |
+
"loss": 0.6659,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.7039838628340898,
|
| 2449 |
+
"grad_norm": 0.6746000051498413,
|
| 2450 |
+
"learning_rate": 2.0533145897917057e-05,
|
| 2451 |
+
"loss": 0.6781,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.7060010085728694,
|
| 2456 |
+
"grad_norm": 0.7686178684234619,
|
| 2457 |
+
"learning_rate": 2.0275293331505436e-05,
|
| 2458 |
+
"loss": 0.7789,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.708018154311649,
|
| 2463 |
+
"grad_norm": 0.3885047733783722,
|
| 2464 |
+
"learning_rate": 2.0018657659394496e-05,
|
| 2465 |
+
"loss": 0.5026,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.7100353000504287,
|
| 2470 |
+
"grad_norm": 0.4007488489151001,
|
| 2471 |
+
"learning_rate": 1.976324938794482e-05,
|
| 2472 |
+
"loss": 0.5077,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.7120524457892082,
|
| 2477 |
+
"grad_norm": 0.4343924820423126,
|
| 2478 |
+
"learning_rate": 1.9509078973268645e-05,
|
| 2479 |
+
"loss": 0.5361,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.7140695915279879,
|
| 2484 |
+
"grad_norm": 0.42959120869636536,
|
| 2485 |
+
"learning_rate": 1.9256156820801895e-05,
|
| 2486 |
+
"loss": 0.551,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.7160867372667675,
|
| 2491 |
+
"grad_norm": 0.4766654372215271,
|
| 2492 |
+
"learning_rate": 1.9004493284877995e-05,
|
| 2493 |
+
"loss": 0.5484,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.7181038830055472,
|
| 2498 |
+
"grad_norm": 0.41945040225982666,
|
| 2499 |
+
"learning_rate": 1.875409866830422e-05,
|
| 2500 |
+
"loss": 0.5211,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.7201210287443268,
|
| 2505 |
+
"grad_norm": 0.4797287881374359,
|
| 2506 |
+
"learning_rate": 1.850498322193972e-05,
|
| 2507 |
+
"loss": 0.531,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.7221381744831064,
|
| 2512 |
+
"grad_norm": 0.4228487014770508,
|
| 2513 |
+
"learning_rate": 1.825715714427594e-05,
|
| 2514 |
+
"loss": 0.5317,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.724155320221886,
|
| 2519 |
+
"grad_norm": 0.4410184919834137,
|
| 2520 |
+
"learning_rate": 1.8010630581019095e-05,
|
| 2521 |
+
"loss": 0.5857,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.7261724659606656,
|
| 2526 |
+
"grad_norm": 0.3858675956726074,
|
| 2527 |
+
"learning_rate": 1.7765413624674866e-05,
|
| 2528 |
+
"loss": 0.4873,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.7281896116994453,
|
| 2533 |
+
"grad_norm": 0.42531728744506836,
|
| 2534 |
+
"learning_rate": 1.752151631413511e-05,
|
| 2535 |
+
"loss": 0.4964,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.7302067574382249,
|
| 2540 |
+
"grad_norm": 0.4145807921886444,
|
| 2541 |
+
"learning_rate": 1.7278948634266968e-05,
|
| 2542 |
+
"loss": 0.5002,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.7322239031770046,
|
| 2547 |
+
"grad_norm": 0.41224363446235657,
|
| 2548 |
+
"learning_rate": 1.703772051550412e-05,
|
| 2549 |
+
"loss": 0.5269,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.7342410489157841,
|
| 2554 |
+
"grad_norm": 0.3786209225654602,
|
| 2555 |
+
"learning_rate": 1.679784183344014e-05,
|
| 2556 |
+
"loss": 0.4877,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.7362581946545638,
|
| 2561 |
+
"grad_norm": 0.4014703035354614,
|
| 2562 |
+
"learning_rate": 1.6559322408424287e-05,
|
| 2563 |
+
"loss": 0.4801,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.7382753403933434,
|
| 2568 |
+
"grad_norm": 0.4308634400367737,
|
| 2569 |
+
"learning_rate": 1.6322172005159435e-05,
|
| 2570 |
+
"loss": 0.5762,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.7402924861321231,
|
| 2575 |
+
"grad_norm": 0.39501944184303284,
|
| 2576 |
+
"learning_rate": 1.608640033230236e-05,
|
| 2577 |
+
"loss": 0.5248,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.7423096318709027,
|
| 2582 |
+
"grad_norm": 0.44487130641937256,
|
| 2583 |
+
"learning_rate": 1.5852017042066214e-05,
|
| 2584 |
+
"loss": 0.5026,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.7443267776096822,
|
| 2589 |
+
"grad_norm": 0.41640734672546387,
|
| 2590 |
+
"learning_rate": 1.5619031729825402e-05,
|
| 2591 |
+
"loss": 0.5583,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.7463439233484619,
|
| 2596 |
+
"grad_norm": 0.42025187611579895,
|
| 2597 |
+
"learning_rate": 1.538745393372281e-05,
|
| 2598 |
+
"loss": 0.5471,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.7483610690872415,
|
| 2603 |
+
"grad_norm": 0.42271459102630615,
|
| 2604 |
+
"learning_rate": 1.5157293134279244e-05,
|
| 2605 |
+
"loss": 0.5239,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.7503782148260212,
|
| 2610 |
+
"grad_norm": 0.47474098205566406,
|
| 2611 |
+
"learning_rate": 1.492855875400534e-05,
|
| 2612 |
+
"loss": 0.5699,
|
| 2613 |
+
"step": 372
|
| 2614 |
}
|
| 2615 |
],
|
| 2616 |
"logging_steps": 1,
|
|
|
|
| 2630 |
"attributes": {}
|
| 2631 |
}
|
| 2632 |
},
|
| 2633 |
+
"total_flos": 6.386815140070687e+17,
|
| 2634 |
"train_batch_size": 4,
|
| 2635 |
"trial_name": null,
|
| 2636 |
"trial_params": null
|