Training in progress, step 360, 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 83945296
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f87937b8ca27ccd4299b7c5130ae44956cd82f7f19b3390773e63ba13d89567
|
| 3 |
size 83945296
|
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:f877183493e32a2691b5c7511ce4bd58a6fd3cee41d4a62b9ca7258094d06da7
|
| 3 |
+
size 43123028
|
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:ee84c698affd10554f0eb51f34115cac8713e7377d5275afd8629075df0dbc22
|
| 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:e42796949e3f84fa5a7101ae5e25b3cdd5ef9daeacff6fbfb32969265fb052fd
|
| 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": 120,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -1703,6 +1703,854 @@
|
|
| 1703 |
"eval_samples_per_second": 14.403,
|
| 1704 |
"eval_steps_per_second": 7.237,
|
| 1705 |
"step": 240
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1706 |
}
|
| 1707 |
],
|
| 1708 |
"logging_steps": 1,
|
|
@@ -1722,7 +2570,7 @@
|
|
| 1722 |
"attributes": {}
|
| 1723 |
}
|
| 1724 |
},
|
| 1725 |
-
"total_flos": 1.
|
| 1726 |
"train_batch_size": 2,
|
| 1727 |
"trial_name": null,
|
| 1728 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7547169811320755,
|
| 5 |
"eval_steps": 120,
|
| 6 |
+
"global_step": 360,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1703 |
"eval_samples_per_second": 14.403,
|
| 1704 |
"eval_steps_per_second": 7.237,
|
| 1705 |
"step": 240
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.5052410901467506,
|
| 1709 |
+
"grad_norm": 0.4681568741798401,
|
| 1710 |
+
"learning_rate": 0.00010168171550280648,
|
| 1711 |
+
"loss": 1.4336,
|
| 1712 |
+
"step": 241
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.5073375262054507,
|
| 1716 |
+
"grad_norm": 0.4985293447971344,
|
| 1717 |
+
"learning_rate": 0.00010100905974490651,
|
| 1718 |
+
"loss": 1.2779,
|
| 1719 |
+
"step": 242
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 0.5094339622641509,
|
| 1723 |
+
"grad_norm": 0.5365903377532959,
|
| 1724 |
+
"learning_rate": 0.0001003363583222415,
|
| 1725 |
+
"loss": 1.5644,
|
| 1726 |
+
"step": 243
|
| 1727 |
+
},
|
| 1728 |
+
{
|
| 1729 |
+
"epoch": 0.5115303983228512,
|
| 1730 |
+
"grad_norm": 0.48949047923088074,
|
| 1731 |
+
"learning_rate": 9.96636416777585e-05,
|
| 1732 |
+
"loss": 1.287,
|
| 1733 |
+
"step": 244
|
| 1734 |
+
},
|
| 1735 |
+
{
|
| 1736 |
+
"epoch": 0.5136268343815513,
|
| 1737 |
+
"grad_norm": 0.5527725219726562,
|
| 1738 |
+
"learning_rate": 9.899094025509352e-05,
|
| 1739 |
+
"loss": 1.4812,
|
| 1740 |
+
"step": 245
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.5157232704402516,
|
| 1744 |
+
"grad_norm": 0.5897342562675476,
|
| 1745 |
+
"learning_rate": 9.831828449719353e-05,
|
| 1746 |
+
"loss": 1.4931,
|
| 1747 |
+
"step": 246
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.5178197064989518,
|
| 1751 |
+
"grad_norm": 0.5165984630584717,
|
| 1752 |
+
"learning_rate": 9.764570484493915e-05,
|
| 1753 |
+
"loss": 1.208,
|
| 1754 |
+
"step": 247
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.519916142557652,
|
| 1758 |
+
"grad_norm": 0.654744565486908,
|
| 1759 |
+
"learning_rate": 9.697323173576667e-05,
|
| 1760 |
+
"loss": 1.6835,
|
| 1761 |
+
"step": 248
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.5220125786163522,
|
| 1765 |
+
"grad_norm": 0.6669768691062927,
|
| 1766 |
+
"learning_rate": 9.630089560229088e-05,
|
| 1767 |
+
"loss": 1.1676,
|
| 1768 |
+
"step": 249
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 0.5241090146750524,
|
| 1772 |
+
"grad_norm": 1.1078684329986572,
|
| 1773 |
+
"learning_rate": 9.562872687092783e-05,
|
| 1774 |
+
"loss": 1.4822,
|
| 1775 |
+
"step": 250
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 0.5262054507337526,
|
| 1779 |
+
"grad_norm": 0.31687837839126587,
|
| 1780 |
+
"learning_rate": 9.495675596051777e-05,
|
| 1781 |
+
"loss": 1.1412,
|
| 1782 |
+
"step": 251
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 0.5283018867924528,
|
| 1786 |
+
"grad_norm": 0.32858502864837646,
|
| 1787 |
+
"learning_rate": 9.428501328094855e-05,
|
| 1788 |
+
"loss": 1.1039,
|
| 1789 |
+
"step": 252
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 0.5303983228511531,
|
| 1793 |
+
"grad_norm": 0.32667338848114014,
|
| 1794 |
+
"learning_rate": 9.36135292317796e-05,
|
| 1795 |
+
"loss": 1.2138,
|
| 1796 |
+
"step": 253
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.5324947589098532,
|
| 1800 |
+
"grad_norm": 0.3239830732345581,
|
| 1801 |
+
"learning_rate": 9.294233420086603e-05,
|
| 1802 |
+
"loss": 1.3107,
|
| 1803 |
+
"step": 254
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 0.5345911949685535,
|
| 1807 |
+
"grad_norm": 0.37542301416397095,
|
| 1808 |
+
"learning_rate": 9.227145856298344e-05,
|
| 1809 |
+
"loss": 1.1616,
|
| 1810 |
+
"step": 255
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.5366876310272537,
|
| 1814 |
+
"grad_norm": 0.356937050819397,
|
| 1815 |
+
"learning_rate": 9.160093267845349e-05,
|
| 1816 |
+
"loss": 1.3193,
|
| 1817 |
+
"step": 256
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 0.5387840670859538,
|
| 1821 |
+
"grad_norm": 0.31748053431510925,
|
| 1822 |
+
"learning_rate": 9.093078689176972e-05,
|
| 1823 |
+
"loss": 1.2206,
|
| 1824 |
+
"step": 257
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 0.5408805031446541,
|
| 1828 |
+
"grad_norm": 0.3468354642391205,
|
| 1829 |
+
"learning_rate": 9.026105153022454e-05,
|
| 1830 |
+
"loss": 1.3771,
|
| 1831 |
+
"step": 258
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 0.5429769392033543,
|
| 1835 |
+
"grad_norm": 0.35934552550315857,
|
| 1836 |
+
"learning_rate": 8.95917569025366e-05,
|
| 1837 |
+
"loss": 1.2956,
|
| 1838 |
+
"step": 259
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.5450733752620545,
|
| 1842 |
+
"grad_norm": 0.35253018140792847,
|
| 1843 |
+
"learning_rate": 8.892293329747922e-05,
|
| 1844 |
+
"loss": 1.2584,
|
| 1845 |
+
"step": 260
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 0.5471698113207547,
|
| 1849 |
+
"grad_norm": 0.3898825943470001,
|
| 1850 |
+
"learning_rate": 8.82546109825098e-05,
|
| 1851 |
+
"loss": 1.2658,
|
| 1852 |
+
"step": 261
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 0.549266247379455,
|
| 1856 |
+
"grad_norm": 0.44399771094322205,
|
| 1857 |
+
"learning_rate": 8.758682020239984e-05,
|
| 1858 |
+
"loss": 1.314,
|
| 1859 |
+
"step": 262
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 0.5513626834381551,
|
| 1863 |
+
"grad_norm": 0.3997746706008911,
|
| 1864 |
+
"learning_rate": 8.69195911778664e-05,
|
| 1865 |
+
"loss": 1.1614,
|
| 1866 |
+
"step": 263
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.5534591194968553,
|
| 1870 |
+
"grad_norm": 0.38755127787590027,
|
| 1871 |
+
"learning_rate": 8.625295410420451e-05,
|
| 1872 |
+
"loss": 1.209,
|
| 1873 |
+
"step": 264
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 0.5555555555555556,
|
| 1877 |
+
"grad_norm": 0.3919242024421692,
|
| 1878 |
+
"learning_rate": 8.558693914992046e-05,
|
| 1879 |
+
"loss": 1.4929,
|
| 1880 |
+
"step": 265
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.5576519916142557,
|
| 1884 |
+
"grad_norm": 0.37072527408599854,
|
| 1885 |
+
"learning_rate": 8.492157645536678e-05,
|
| 1886 |
+
"loss": 1.0455,
|
| 1887 |
+
"step": 266
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 0.559748427672956,
|
| 1891 |
+
"grad_norm": 0.39799362421035767,
|
| 1892 |
+
"learning_rate": 8.425689613137813e-05,
|
| 1893 |
+
"loss": 1.3028,
|
| 1894 |
+
"step": 267
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 0.5618448637316562,
|
| 1898 |
+
"grad_norm": 0.34042948484420776,
|
| 1899 |
+
"learning_rate": 8.359292825790859e-05,
|
| 1900 |
+
"loss": 1.2346,
|
| 1901 |
+
"step": 268
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.5639412997903563,
|
| 1905 |
+
"grad_norm": 0.41574835777282715,
|
| 1906 |
+
"learning_rate": 8.292970288267042e-05,
|
| 1907 |
+
"loss": 1.6697,
|
| 1908 |
+
"step": 269
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.5660377358490566,
|
| 1912 |
+
"grad_norm": 0.39489176869392395,
|
| 1913 |
+
"learning_rate": 8.226725001977445e-05,
|
| 1914 |
+
"loss": 1.4031,
|
| 1915 |
+
"step": 270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.5681341719077568,
|
| 1919 |
+
"grad_norm": 0.37004798650741577,
|
| 1920 |
+
"learning_rate": 8.160559964837149e-05,
|
| 1921 |
+
"loss": 1.0657,
|
| 1922 |
+
"step": 271
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.570230607966457,
|
| 1926 |
+
"grad_norm": 0.4158036410808563,
|
| 1927 |
+
"learning_rate": 8.094478171129588e-05,
|
| 1928 |
+
"loss": 1.4288,
|
| 1929 |
+
"step": 272
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 0.5723270440251572,
|
| 1933 |
+
"grad_norm": 0.39994949102401733,
|
| 1934 |
+
"learning_rate": 8.028482611371028e-05,
|
| 1935 |
+
"loss": 1.4105,
|
| 1936 |
+
"step": 273
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 0.5744234800838575,
|
| 1940 |
+
"grad_norm": 0.3869493007659912,
|
| 1941 |
+
"learning_rate": 7.96257627217524e-05,
|
| 1942 |
+
"loss": 1.452,
|
| 1943 |
+
"step": 274
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 0.5765199161425576,
|
| 1947 |
+
"grad_norm": 0.41220688819885254,
|
| 1948 |
+
"learning_rate": 7.896762136118342e-05,
|
| 1949 |
+
"loss": 1.6473,
|
| 1950 |
+
"step": 275
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.5786163522012578,
|
| 1954 |
+
"grad_norm": 0.4000371992588043,
|
| 1955 |
+
"learning_rate": 7.831043181603814e-05,
|
| 1956 |
+
"loss": 1.2974,
|
| 1957 |
+
"step": 276
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 0.5807127882599581,
|
| 1961 |
+
"grad_norm": 0.43282032012939453,
|
| 1962 |
+
"learning_rate": 7.765422382727719e-05,
|
| 1963 |
+
"loss": 1.4971,
|
| 1964 |
+
"step": 277
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.5828092243186582,
|
| 1968 |
+
"grad_norm": 0.37778139114379883,
|
| 1969 |
+
"learning_rate": 7.699902709144114e-05,
|
| 1970 |
+
"loss": 1.2826,
|
| 1971 |
+
"step": 278
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 0.5849056603773585,
|
| 1975 |
+
"grad_norm": 0.3722570538520813,
|
| 1976 |
+
"learning_rate": 7.634487125930648e-05,
|
| 1977 |
+
"loss": 1.0699,
|
| 1978 |
+
"step": 279
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.5870020964360587,
|
| 1982 |
+
"grad_norm": 0.4069075882434845,
|
| 1983 |
+
"learning_rate": 7.569178593454392e-05,
|
| 1984 |
+
"loss": 1.2477,
|
| 1985 |
+
"step": 280
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.589098532494759,
|
| 1989 |
+
"grad_norm": 0.4309166669845581,
|
| 1990 |
+
"learning_rate": 7.503980067237852e-05,
|
| 1991 |
+
"loss": 1.2098,
|
| 1992 |
+
"step": 281
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.5911949685534591,
|
| 1996 |
+
"grad_norm": 0.3983183205127716,
|
| 1997 |
+
"learning_rate": 7.438894497825235e-05,
|
| 1998 |
+
"loss": 1.2926,
|
| 1999 |
+
"step": 282
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.5932914046121593,
|
| 2003 |
+
"grad_norm": 0.4100129008293152,
|
| 2004 |
+
"learning_rate": 7.373924830648904e-05,
|
| 2005 |
+
"loss": 1.2444,
|
| 2006 |
+
"step": 283
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.5953878406708596,
|
| 2010 |
+
"grad_norm": 0.45969000458717346,
|
| 2011 |
+
"learning_rate": 7.309074005896103e-05,
|
| 2012 |
+
"loss": 1.5165,
|
| 2013 |
+
"step": 284
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.5974842767295597,
|
| 2017 |
+
"grad_norm": 0.42328140139579773,
|
| 2018 |
+
"learning_rate": 7.244344958375881e-05,
|
| 2019 |
+
"loss": 1.2259,
|
| 2020 |
+
"step": 285
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.59958071278826,
|
| 2024 |
+
"grad_norm": 0.4778403639793396,
|
| 2025 |
+
"learning_rate": 7.179740617386295e-05,
|
| 2026 |
+
"loss": 1.3914,
|
| 2027 |
+
"step": 286
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.6016771488469602,
|
| 2031 |
+
"grad_norm": 0.5297214388847351,
|
| 2032 |
+
"learning_rate": 7.115263906581829e-05,
|
| 2033 |
+
"loss": 1.6115,
|
| 2034 |
+
"step": 287
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.6037735849056604,
|
| 2038 |
+
"grad_norm": 0.5362441539764404,
|
| 2039 |
+
"learning_rate": 7.0509177438411e-05,
|
| 2040 |
+
"loss": 1.3647,
|
| 2041 |
+
"step": 288
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.6058700209643606,
|
| 2045 |
+
"grad_norm": 0.4826620817184448,
|
| 2046 |
+
"learning_rate": 6.986705041134796e-05,
|
| 2047 |
+
"loss": 1.5252,
|
| 2048 |
+
"step": 289
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.6079664570230608,
|
| 2052 |
+
"grad_norm": 0.4332706928253174,
|
| 2053 |
+
"learning_rate": 6.922628704393904e-05,
|
| 2054 |
+
"loss": 1.2502,
|
| 2055 |
+
"step": 290
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.610062893081761,
|
| 2059 |
+
"grad_norm": 0.47122621536254883,
|
| 2060 |
+
"learning_rate": 6.858691633378202e-05,
|
| 2061 |
+
"loss": 1.2901,
|
| 2062 |
+
"step": 291
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 0.6121593291404612,
|
| 2066 |
+
"grad_norm": 0.5842708945274353,
|
| 2067 |
+
"learning_rate": 6.794896721545032e-05,
|
| 2068 |
+
"loss": 1.5187,
|
| 2069 |
+
"step": 292
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 0.6142557651991615,
|
| 2073 |
+
"grad_norm": 0.4648444652557373,
|
| 2074 |
+
"learning_rate": 6.73124685591835e-05,
|
| 2075 |
+
"loss": 1.2955,
|
| 2076 |
+
"step": 293
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 0.6163522012578616,
|
| 2080 |
+
"grad_norm": 0.5161048173904419,
|
| 2081 |
+
"learning_rate": 6.667744916958085e-05,
|
| 2082 |
+
"loss": 1.4571,
|
| 2083 |
+
"step": 294
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 0.6184486373165619,
|
| 2087 |
+
"grad_norm": 0.4833495020866394,
|
| 2088 |
+
"learning_rate": 6.604393778429772e-05,
|
| 2089 |
+
"loss": 1.0478,
|
| 2090 |
+
"step": 295
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.6205450733752621,
|
| 2094 |
+
"grad_norm": 0.5789321660995483,
|
| 2095 |
+
"learning_rate": 6.541196307274517e-05,
|
| 2096 |
+
"loss": 1.3774,
|
| 2097 |
+
"step": 296
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 0.6226415094339622,
|
| 2101 |
+
"grad_norm": 0.6312127709388733,
|
| 2102 |
+
"learning_rate": 6.478155363479236e-05,
|
| 2103 |
+
"loss": 1.6907,
|
| 2104 |
+
"step": 297
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 0.6247379454926625,
|
| 2108 |
+
"grad_norm": 0.5820334553718567,
|
| 2109 |
+
"learning_rate": 6.415273799947234e-05,
|
| 2110 |
+
"loss": 1.1408,
|
| 2111 |
+
"step": 298
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 0.6268343815513627,
|
| 2115 |
+
"grad_norm": 0.6500141024589539,
|
| 2116 |
+
"learning_rate": 6.352554462369112e-05,
|
| 2117 |
+
"loss": 1.3631,
|
| 2118 |
+
"step": 299
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 0.6289308176100629,
|
| 2122 |
+
"grad_norm": 0.9458006024360657,
|
| 2123 |
+
"learning_rate": 6.290000189093959e-05,
|
| 2124 |
+
"loss": 1.3022,
|
| 2125 |
+
"step": 300
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 0.6310272536687631,
|
| 2129 |
+
"grad_norm": 0.32660362124443054,
|
| 2130 |
+
"learning_rate": 6.227613811000925e-05,
|
| 2131 |
+
"loss": 1.015,
|
| 2132 |
+
"step": 301
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 0.6331236897274634,
|
| 2136 |
+
"grad_norm": 0.2937367558479309,
|
| 2137 |
+
"learning_rate": 6.165398151371106e-05,
|
| 2138 |
+
"loss": 0.9367,
|
| 2139 |
+
"step": 302
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 0.6352201257861635,
|
| 2143 |
+
"grad_norm": 0.37501296401023865,
|
| 2144 |
+
"learning_rate": 6.103356025759759e-05,
|
| 2145 |
+
"loss": 1.2524,
|
| 2146 |
+
"step": 303
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 0.6373165618448637,
|
| 2150 |
+
"grad_norm": 0.29749903082847595,
|
| 2151 |
+
"learning_rate": 6.04149024186891e-05,
|
| 2152 |
+
"loss": 1.068,
|
| 2153 |
+
"step": 304
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 0.639412997903564,
|
| 2157 |
+
"grad_norm": 0.2890242338180542,
|
| 2158 |
+
"learning_rate": 5.9798035994202836e-05,
|
| 2159 |
+
"loss": 1.1022,
|
| 2160 |
+
"step": 305
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 0.6415094339622641,
|
| 2164 |
+
"grad_norm": 0.33415907621383667,
|
| 2165 |
+
"learning_rate": 5.918298890028591e-05,
|
| 2166 |
+
"loss": 1.4532,
|
| 2167 |
+
"step": 306
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 0.6436058700209644,
|
| 2171 |
+
"grad_norm": 0.2996525168418884,
|
| 2172 |
+
"learning_rate": 5.8569788970752114e-05,
|
| 2173 |
+
"loss": 1.0128,
|
| 2174 |
+
"step": 307
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 0.6457023060796646,
|
| 2178 |
+
"grad_norm": 0.343421071767807,
|
| 2179 |
+
"learning_rate": 5.795846395582225e-05,
|
| 2180 |
+
"loss": 1.0972,
|
| 2181 |
+
"step": 308
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 0.6477987421383647,
|
| 2185 |
+
"grad_norm": 0.32696643471717834,
|
| 2186 |
+
"learning_rate": 5.734904152086828e-05,
|
| 2187 |
+
"loss": 1.2231,
|
| 2188 |
+
"step": 309
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 0.649895178197065,
|
| 2192 |
+
"grad_norm": 0.34797459840774536,
|
| 2193 |
+
"learning_rate": 5.6741549245161285e-05,
|
| 2194 |
+
"loss": 1.2454,
|
| 2195 |
+
"step": 310
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 0.6519916142557652,
|
| 2199 |
+
"grad_norm": 0.3834897577762604,
|
| 2200 |
+
"learning_rate": 5.6136014620623525e-05,
|
| 2201 |
+
"loss": 1.288,
|
| 2202 |
+
"step": 311
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 0.6540880503144654,
|
| 2206 |
+
"grad_norm": 0.3343852162361145,
|
| 2207 |
+
"learning_rate": 5.5532465050584206e-05,
|
| 2208 |
+
"loss": 1.1281,
|
| 2209 |
+
"step": 312
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 0.6561844863731656,
|
| 2213 |
+
"grad_norm": 0.33837154507637024,
|
| 2214 |
+
"learning_rate": 5.4930927848539256e-05,
|
| 2215 |
+
"loss": 1.1146,
|
| 2216 |
+
"step": 313
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 0.6582809224318659,
|
| 2220 |
+
"grad_norm": 0.395158052444458,
|
| 2221 |
+
"learning_rate": 5.433143023691547e-05,
|
| 2222 |
+
"loss": 1.2687,
|
| 2223 |
+
"step": 314
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 0.660377358490566,
|
| 2227 |
+
"grad_norm": 0.3619198203086853,
|
| 2228 |
+
"learning_rate": 5.373399934583839e-05,
|
| 2229 |
+
"loss": 1.3387,
|
| 2230 |
+
"step": 315
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.6624737945492662,
|
| 2234 |
+
"grad_norm": 0.3589290380477905,
|
| 2235 |
+
"learning_rate": 5.3138662211904654e-05,
|
| 2236 |
+
"loss": 1.1821,
|
| 2237 |
+
"step": 316
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 0.6645702306079665,
|
| 2241 |
+
"grad_norm": 0.46267178654670715,
|
| 2242 |
+
"learning_rate": 5.25454457769583e-05,
|
| 2243 |
+
"loss": 1.3747,
|
| 2244 |
+
"step": 317
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 0.6666666666666666,
|
| 2248 |
+
"grad_norm": 0.35914307832717896,
|
| 2249 |
+
"learning_rate": 5.1954376886871746e-05,
|
| 2250 |
+
"loss": 1.0113,
|
| 2251 |
+
"step": 318
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 0.6687631027253669,
|
| 2255 |
+
"grad_norm": 0.3956226408481598,
|
| 2256 |
+
"learning_rate": 5.1365482290330645e-05,
|
| 2257 |
+
"loss": 1.2348,
|
| 2258 |
+
"step": 319
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 0.6708595387840671,
|
| 2262 |
+
"grad_norm": 0.3596416115760803,
|
| 2263 |
+
"learning_rate": 5.07787886376236e-05,
|
| 2264 |
+
"loss": 1.0949,
|
| 2265 |
+
"step": 320
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 0.6729559748427673,
|
| 2269 |
+
"grad_norm": 0.4158559739589691,
|
| 2270 |
+
"learning_rate": 5.019432247943595e-05,
|
| 2271 |
+
"loss": 1.236,
|
| 2272 |
+
"step": 321
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 0.6750524109014675,
|
| 2276 |
+
"grad_norm": 0.4549407958984375,
|
| 2277 |
+
"learning_rate": 4.961211026564837e-05,
|
| 2278 |
+
"loss": 1.3363,
|
| 2279 |
+
"step": 322
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 0.6771488469601677,
|
| 2283 |
+
"grad_norm": 0.3847133219242096,
|
| 2284 |
+
"learning_rate": 4.90321783441397e-05,
|
| 2285 |
+
"loss": 0.9304,
|
| 2286 |
+
"step": 323
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 0.6792452830188679,
|
| 2290 |
+
"grad_norm": 0.381661057472229,
|
| 2291 |
+
"learning_rate": 4.845455295959468e-05,
|
| 2292 |
+
"loss": 0.9649,
|
| 2293 |
+
"step": 324
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 0.6813417190775681,
|
| 2297 |
+
"grad_norm": 0.3821321129798889,
|
| 2298 |
+
"learning_rate": 4.787926025231634e-05,
|
| 2299 |
+
"loss": 1.128,
|
| 2300 |
+
"step": 325
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.6834381551362684,
|
| 2304 |
+
"grad_norm": 0.42627325654029846,
|
| 2305 |
+
"learning_rate": 4.730632625704288e-05,
|
| 2306 |
+
"loss": 1.2158,
|
| 2307 |
+
"step": 326
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 0.6855345911949685,
|
| 2311 |
+
"grad_norm": 0.5116820931434631,
|
| 2312 |
+
"learning_rate": 4.673577690176956e-05,
|
| 2313 |
+
"loss": 1.228,
|
| 2314 |
+
"step": 327
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 0.6876310272536688,
|
| 2318 |
+
"grad_norm": 0.37617841362953186,
|
| 2319 |
+
"learning_rate": 4.616763800657534e-05,
|
| 2320 |
+
"loss": 1.0418,
|
| 2321 |
+
"step": 328
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 0.689727463312369,
|
| 2325 |
+
"grad_norm": 0.40348488092422485,
|
| 2326 |
+
"learning_rate": 4.560193528245425e-05,
|
| 2327 |
+
"loss": 1.303,
|
| 2328 |
+
"step": 329
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 0.6918238993710691,
|
| 2332 |
+
"grad_norm": 0.39703086018562317,
|
| 2333 |
+
"learning_rate": 4.5038694330152135e-05,
|
| 2334 |
+
"loss": 1.0071,
|
| 2335 |
+
"step": 330
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 0.6939203354297694,
|
| 2339 |
+
"grad_norm": 0.4976246654987335,
|
| 2340 |
+
"learning_rate": 4.447794063900772e-05,
|
| 2341 |
+
"loss": 1.4131,
|
| 2342 |
+
"step": 331
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 0.6960167714884696,
|
| 2346 |
+
"grad_norm": 0.4676234722137451,
|
| 2347 |
+
"learning_rate": 4.391969958579948e-05,
|
| 2348 |
+
"loss": 1.4407,
|
| 2349 |
+
"step": 332
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 0.6981132075471698,
|
| 2353 |
+
"grad_norm": 0.47766080498695374,
|
| 2354 |
+
"learning_rate": 4.3363996433596954e-05,
|
| 2355 |
+
"loss": 1.196,
|
| 2356 |
+
"step": 333
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 0.70020964360587,
|
| 2360 |
+
"grad_norm": 0.45446816086769104,
|
| 2361 |
+
"learning_rate": 4.281085633061764e-05,
|
| 2362 |
+
"loss": 1.4197,
|
| 2363 |
+
"step": 334
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 0.7023060796645703,
|
| 2367 |
+
"grad_norm": 0.4232485294342041,
|
| 2368 |
+
"learning_rate": 4.2260304309088696e-05,
|
| 2369 |
+
"loss": 1.1722,
|
| 2370 |
+
"step": 335
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 0.7044025157232704,
|
| 2374 |
+
"grad_norm": 0.4267813563346863,
|
| 2375 |
+
"learning_rate": 4.171236528411436e-05,
|
| 2376 |
+
"loss": 1.1742,
|
| 2377 |
+
"step": 336
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 0.7064989517819706,
|
| 2381 |
+
"grad_norm": 0.5247170329093933,
|
| 2382 |
+
"learning_rate": 4.116706405254834e-05,
|
| 2383 |
+
"loss": 1.4083,
|
| 2384 |
+
"step": 337
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 0.7085953878406709,
|
| 2388 |
+
"grad_norm": 0.4244326949119568,
|
| 2389 |
+
"learning_rate": 4.0624425291871506e-05,
|
| 2390 |
+
"loss": 0.9588,
|
| 2391 |
+
"step": 338
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 0.710691823899371,
|
| 2395 |
+
"grad_norm": 0.47463715076446533,
|
| 2396 |
+
"learning_rate": 4.0084473559075333e-05,
|
| 2397 |
+
"loss": 1.2465,
|
| 2398 |
+
"step": 339
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 0.7127882599580713,
|
| 2402 |
+
"grad_norm": 0.5082678198814392,
|
| 2403 |
+
"learning_rate": 3.9547233289550356e-05,
|
| 2404 |
+
"loss": 1.2823,
|
| 2405 |
+
"step": 340
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 0.7148846960167715,
|
| 2409 |
+
"grad_norm": 0.4892716407775879,
|
| 2410 |
+
"learning_rate": 3.901272879598058e-05,
|
| 2411 |
+
"loss": 1.3535,
|
| 2412 |
+
"step": 341
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 0.7169811320754716,
|
| 2416 |
+
"grad_norm": 0.4783913791179657,
|
| 2417 |
+
"learning_rate": 3.848098426724306e-05,
|
| 2418 |
+
"loss": 1.2831,
|
| 2419 |
+
"step": 342
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 0.7190775681341719,
|
| 2423 |
+
"grad_norm": 0.551485002040863,
|
| 2424 |
+
"learning_rate": 3.7952023767313264e-05,
|
| 2425 |
+
"loss": 1.4206,
|
| 2426 |
+
"step": 343
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 0.7211740041928721,
|
| 2430 |
+
"grad_norm": 0.5458889007568359,
|
| 2431 |
+
"learning_rate": 3.7425871234176134e-05,
|
| 2432 |
+
"loss": 1.192,
|
| 2433 |
+
"step": 344
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 0.7232704402515723,
|
| 2437 |
+
"grad_norm": 0.5561458468437195,
|
| 2438 |
+
"learning_rate": 3.690255047874267e-05,
|
| 2439 |
+
"loss": 1.2386,
|
| 2440 |
+
"step": 345
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 0.7253668763102725,
|
| 2444 |
+
"grad_norm": 0.5652120113372803,
|
| 2445 |
+
"learning_rate": 3.6382085183772394e-05,
|
| 2446 |
+
"loss": 1.1185,
|
| 2447 |
+
"step": 346
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 0.7274633123689728,
|
| 2451 |
+
"grad_norm": 0.6339935064315796,
|
| 2452 |
+
"learning_rate": 3.586449890280172e-05,
|
| 2453 |
+
"loss": 1.406,
|
| 2454 |
+
"step": 347
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 0.7295597484276729,
|
| 2458 |
+
"grad_norm": 0.6352638006210327,
|
| 2459 |
+
"learning_rate": 3.534981505907792e-05,
|
| 2460 |
+
"loss": 1.4451,
|
| 2461 |
+
"step": 348
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 0.7316561844863732,
|
| 2465 |
+
"grad_norm": 0.7120410203933716,
|
| 2466 |
+
"learning_rate": 3.483805694449913e-05,
|
| 2467 |
+
"loss": 1.0784,
|
| 2468 |
+
"step": 349
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 0.7337526205450734,
|
| 2472 |
+
"grad_norm": 0.9480962753295898,
|
| 2473 |
+
"learning_rate": 3.432924771856029e-05,
|
| 2474 |
+
"loss": 1.1712,
|
| 2475 |
+
"step": 350
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 0.7358490566037735,
|
| 2479 |
+
"grad_norm": 0.35054638981819153,
|
| 2480 |
+
"learning_rate": 3.3823410407305015e-05,
|
| 2481 |
+
"loss": 1.3018,
|
| 2482 |
+
"step": 351
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 0.7379454926624738,
|
| 2486 |
+
"grad_norm": 0.30331677198410034,
|
| 2487 |
+
"learning_rate": 3.33205679022837e-05,
|
| 2488 |
+
"loss": 1.2488,
|
| 2489 |
+
"step": 352
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 0.740041928721174,
|
| 2493 |
+
"grad_norm": 0.30316171050071716,
|
| 2494 |
+
"learning_rate": 3.2820742959517436e-05,
|
| 2495 |
+
"loss": 0.9948,
|
| 2496 |
+
"step": 353
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 0.7421383647798742,
|
| 2500 |
+
"grad_norm": 0.3911697268486023,
|
| 2501 |
+
"learning_rate": 3.232395819846824e-05,
|
| 2502 |
+
"loss": 1.1168,
|
| 2503 |
+
"step": 354
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 0.7442348008385744,
|
| 2507 |
+
"grad_norm": 0.3382274806499481,
|
| 2508 |
+
"learning_rate": 3.1830236101015446e-05,
|
| 2509 |
+
"loss": 1.0427,
|
| 2510 |
+
"step": 355
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 0.7463312368972747,
|
| 2514 |
+
"grad_norm": 0.32363229990005493,
|
| 2515 |
+
"learning_rate": 3.1339599010438134e-05,
|
| 2516 |
+
"loss": 0.94,
|
| 2517 |
+
"step": 356
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 0.7484276729559748,
|
| 2521 |
+
"grad_norm": 0.3242010772228241,
|
| 2522 |
+
"learning_rate": 3.0852069130404284e-05,
|
| 2523 |
+
"loss": 1.2767,
|
| 2524 |
+
"step": 357
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 0.750524109014675,
|
| 2528 |
+
"grad_norm": 0.33006346225738525,
|
| 2529 |
+
"learning_rate": 3.036766852396561e-05,
|
| 2530 |
+
"loss": 0.9687,
|
| 2531 |
+
"step": 358
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 0.7526205450733753,
|
| 2535 |
+
"grad_norm": 0.35846400260925293,
|
| 2536 |
+
"learning_rate": 2.9886419112559394e-05,
|
| 2537 |
+
"loss": 1.2036,
|
| 2538 |
+
"step": 359
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 0.7547169811320755,
|
| 2542 |
+
"grad_norm": 0.35825300216674805,
|
| 2543 |
+
"learning_rate": 2.9408342675016286e-05,
|
| 2544 |
+
"loss": 1.4004,
|
| 2545 |
+
"step": 360
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 0.7547169811320755,
|
| 2549 |
+
"eval_loss": 1.2862719297409058,
|
| 2550 |
+
"eval_runtime": 13.9861,
|
| 2551 |
+
"eval_samples_per_second": 14.371,
|
| 2552 |
+
"eval_steps_per_second": 7.221,
|
| 2553 |
+
"step": 360
|
| 2554 |
}
|
| 2555 |
],
|
| 2556 |
"logging_steps": 1,
|
|
|
|
| 2570 |
"attributes": {}
|
| 2571 |
}
|
| 2572 |
},
|
| 2573 |
+
"total_flos": 1.528664851092603e+17,
|
| 2574 |
"train_batch_size": 2,
|
| 2575 |
"trial_name": null,
|
| 2576 |
"trial_params": null
|