Training in progress, step 337, 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 67126760
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ff032c2b84a94327628cf41dd74556da8b5604505cb11b3a7fbcd44644cbb23
|
| 3 |
size 67126760
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 134325882
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ab626055373aac8d72bf80d61010fbfe74b3561c5f47454fa675333369173bd
|
| 3 |
size 134325882
|
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 sha256:3ddb0dc9aaf9cc0dc4a6b0020adae3a2c77f9db62a9fc24dadaa9a18d3a18470
|
| 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:de435905b02e6e14253beeb6db9bdb52440d05130914ca4da5e77f7ebf47b817
|
| 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,
|
|
@@ -1680,6 +1680,692 @@
|
|
| 1680 |
"learning_rate": 3.657954557919183e-05,
|
| 1681 |
"loss": 1.6068,
|
| 1682 |
"step": 239
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1683 |
}
|
| 1684 |
],
|
| 1685 |
"logging_steps": 1,
|
|
@@ -1694,12 +2380,12 @@
|
|
| 1694 |
"should_evaluate": false,
|
| 1695 |
"should_log": false,
|
| 1696 |
"should_save": true,
|
| 1697 |
-
"should_training_stop":
|
| 1698 |
},
|
| 1699 |
"attributes": {}
|
| 1700 |
}
|
| 1701 |
},
|
| 1702 |
-
"total_flos":
|
| 1703 |
"train_batch_size": 4,
|
| 1704 |
"trial_name": null,
|
| 1705 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.997779422649889,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 337,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 1680 |
"learning_rate": 3.657954557919183e-05,
|
| 1681 |
"loss": 1.6068,
|
| 1682 |
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7105847520355293,
|
| 1686 |
+
"grad_norm": 0.6972988843917847,
|
| 1687 |
+
"learning_rate": 3.5942280889623026e-05,
|
| 1688 |
+
"loss": 0.6496,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7135455218356773,
|
| 1693 |
+
"grad_norm": 0.620529055595398,
|
| 1694 |
+
"learning_rate": 3.5307486283103966e-05,
|
| 1695 |
+
"loss": 0.531,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7165062916358254,
|
| 1700 |
+
"grad_norm": 0.71197110414505,
|
| 1701 |
+
"learning_rate": 3.467527329945026e-05,
|
| 1702 |
+
"loss": 0.6825,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7194670614359734,
|
| 1707 |
+
"grad_norm": 0.6315649151802063,
|
| 1708 |
+
"learning_rate": 3.404575302486039e-05,
|
| 1709 |
+
"loss": 0.4259,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7224278312361214,
|
| 1714 |
+
"grad_norm": 0.6250630617141724,
|
| 1715 |
+
"learning_rate": 3.3419036072396616e-05,
|
| 1716 |
+
"loss": 0.2993,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7253886010362695,
|
| 1721 |
+
"grad_norm": 0.6370623111724854,
|
| 1722 |
+
"learning_rate": 3.27952325625493e-05,
|
| 1723 |
+
"loss": 0.3333,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7283493708364175,
|
| 1728 |
+
"grad_norm": 0.494165301322937,
|
| 1729 |
+
"learning_rate": 3.2174452103887456e-05,
|
| 1730 |
+
"loss": 0.1763,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7313101406365655,
|
| 1735 |
+
"grad_norm": 0.4754185676574707,
|
| 1736 |
+
"learning_rate": 3.1556803773799614e-05,
|
| 1737 |
+
"loss": 0.116,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7342709104367136,
|
| 1742 |
+
"grad_norm": 0.5947216749191284,
|
| 1743 |
+
"learning_rate": 3.094239609932764e-05,
|
| 1744 |
+
"loss": 0.1401,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7372316802368616,
|
| 1749 |
+
"grad_norm": 0.7304718494415283,
|
| 1750 |
+
"learning_rate": 3.0331337038097597e-05,
|
| 1751 |
+
"loss": 0.1343,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7401924500370096,
|
| 1756 |
+
"grad_norm": 1.4958913326263428,
|
| 1757 |
+
"learning_rate": 2.9723733959350307e-05,
|
| 1758 |
+
"loss": 0.2833,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7431532198371577,
|
| 1763 |
+
"grad_norm": 1.9902845621109009,
|
| 1764 |
+
"learning_rate": 2.911969362507574e-05,
|
| 1765 |
+
"loss": 2.3228,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7461139896373057,
|
| 1770 |
+
"grad_norm": 1.4703550338745117,
|
| 1771 |
+
"learning_rate": 2.8519322171253602e-05,
|
| 1772 |
+
"loss": 1.8494,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7490747594374537,
|
| 1777 |
+
"grad_norm": 1.096100926399231,
|
| 1778 |
+
"learning_rate": 2.7922725089204426e-05,
|
| 1779 |
+
"loss": 1.3966,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7520355292376018,
|
| 1784 |
+
"grad_norm": 1.069754719734192,
|
| 1785 |
+
"learning_rate": 2.733000720705341e-05,
|
| 1786 |
+
"loss": 1.512,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7549962990377498,
|
| 1791 |
+
"grad_norm": 1.1707820892333984,
|
| 1792 |
+
"learning_rate": 2.674127267131131e-05,
|
| 1793 |
+
"loss": 1.7061,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7579570688378978,
|
| 1798 |
+
"grad_norm": 1.0361454486846924,
|
| 1799 |
+
"learning_rate": 2.6156624928574707e-05,
|
| 1800 |
+
"loss": 1.2677,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7609178386380459,
|
| 1805 |
+
"grad_norm": 1.0314826965332031,
|
| 1806 |
+
"learning_rate": 2.5576166707349385e-05,
|
| 1807 |
+
"loss": 1.7485,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7638786084381939,
|
| 1812 |
+
"grad_norm": 1.0188485383987427,
|
| 1813 |
+
"learning_rate": 2.500000000000001e-05,
|
| 1814 |
+
"loss": 1.7111,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7668393782383419,
|
| 1819 |
+
"grad_norm": 1.0937459468841553,
|
| 1820 |
+
"learning_rate": 2.4428226044828896e-05,
|
| 1821 |
+
"loss": 1.9624,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7698001480384901,
|
| 1826 |
+
"grad_norm": 1.023545742034912,
|
| 1827 |
+
"learning_rate": 2.3860945308287552e-05,
|
| 1828 |
+
"loss": 1.3442,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7727609178386381,
|
| 1833 |
+
"grad_norm": 1.0937973260879517,
|
| 1834 |
+
"learning_rate": 2.3298257467323604e-05,
|
| 1835 |
+
"loss": 1.9485,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7757216876387861,
|
| 1840 |
+
"grad_norm": 1.153558373451233,
|
| 1841 |
+
"learning_rate": 2.2740261391866637e-05,
|
| 1842 |
+
"loss": 1.8108,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7786824574389342,
|
| 1847 |
+
"grad_norm": 1.0607295036315918,
|
| 1848 |
+
"learning_rate": 2.2187055127455653e-05,
|
| 1849 |
+
"loss": 1.7313,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7816432272390822,
|
| 1854 |
+
"grad_norm": 1.0273983478546143,
|
| 1855 |
+
"learning_rate": 2.16387358780116e-05,
|
| 1856 |
+
"loss": 1.4125,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7846039970392302,
|
| 1861 |
+
"grad_norm": 1.010677456855774,
|
| 1862 |
+
"learning_rate": 2.1095399988757574e-05,
|
| 1863 |
+
"loss": 1.6405,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7875647668393783,
|
| 1868 |
+
"grad_norm": 1.0559284687042236,
|
| 1869 |
+
"learning_rate": 2.0557142929290023e-05,
|
| 1870 |
+
"loss": 1.9992,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7905255366395263,
|
| 1875 |
+
"grad_norm": 1.1667604446411133,
|
| 1876 |
+
"learning_rate": 2.002405927680374e-05,
|
| 1877 |
+
"loss": 1.6065,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7934863064396743,
|
| 1882 |
+
"grad_norm": 0.968020498752594,
|
| 1883 |
+
"learning_rate": 1.9496242699473783e-05,
|
| 1884 |
+
"loss": 1.6283,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7964470762398224,
|
| 1889 |
+
"grad_norm": 1.1873533725738525,
|
| 1890 |
+
"learning_rate": 1.897378593999693e-05,
|
| 1891 |
+
"loss": 1.622,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7994078460399704,
|
| 1896 |
+
"grad_norm": 1.0366755723953247,
|
| 1897 |
+
"learning_rate": 1.8456780799295886e-05,
|
| 1898 |
+
"loss": 1.7705,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8023686158401184,
|
| 1903 |
+
"grad_norm": 0.9406031966209412,
|
| 1904 |
+
"learning_rate": 1.794531812038901e-05,
|
| 1905 |
+
"loss": 1.3649,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8053293856402665,
|
| 1910 |
+
"grad_norm": 1.1950721740722656,
|
| 1911 |
+
"learning_rate": 1.743948777242814e-05,
|
| 1912 |
+
"loss": 2.0907,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8082901554404145,
|
| 1917 |
+
"grad_norm": 0.8741153478622437,
|
| 1918 |
+
"learning_rate": 1.6939378634907815e-05,
|
| 1919 |
+
"loss": 1.3801,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8112509252405625,
|
| 1924 |
+
"grad_norm": 0.9927524924278259,
|
| 1925 |
+
"learning_rate": 1.6445078582048155e-05,
|
| 1926 |
+
"loss": 1.6127,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8142116950407106,
|
| 1931 |
+
"grad_norm": 1.0479118824005127,
|
| 1932 |
+
"learning_rate": 1.5956674467354537e-05,
|
| 1933 |
+
"loss": 2.0404,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8171724648408586,
|
| 1938 |
+
"grad_norm": 0.9169769883155823,
|
| 1939 |
+
"learning_rate": 1.5474252108356474e-05,
|
| 1940 |
+
"loss": 1.5797,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8201332346410066,
|
| 1945 |
+
"grad_norm": 0.9381272196769714,
|
| 1946 |
+
"learning_rate": 1.4997896271528739e-05,
|
| 1947 |
+
"loss": 1.0802,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8230940044411547,
|
| 1952 |
+
"grad_norm": 0.9167252779006958,
|
| 1953 |
+
"learning_rate": 1.452769065739688e-05,
|
| 1954 |
+
"loss": 1.3707,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8260547742413027,
|
| 1959 |
+
"grad_norm": 1.007392168045044,
|
| 1960 |
+
"learning_rate": 1.4063717885830374e-05,
|
| 1961 |
+
"loss": 1.79,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8290155440414507,
|
| 1966 |
+
"grad_norm": 1.0070596933364868,
|
| 1967 |
+
"learning_rate": 1.3606059481525296e-05,
|
| 1968 |
+
"loss": 1.7477,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8319763138415989,
|
| 1973 |
+
"grad_norm": 1.103658676147461,
|
| 1974 |
+
"learning_rate": 1.315479585967978e-05,
|
| 1975 |
+
"loss": 1.8807,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8349370836417469,
|
| 1980 |
+
"grad_norm": 0.742683470249176,
|
| 1981 |
+
"learning_rate": 1.2710006311864104e-05,
|
| 1982 |
+
"loss": 0.99,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8378978534418949,
|
| 1987 |
+
"grad_norm": 0.8142567276954651,
|
| 1988 |
+
"learning_rate": 1.2271768992088489e-05,
|
| 1989 |
+
"loss": 1.1157,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.840858623242043,
|
| 1994 |
+
"grad_norm": 0.8256039023399353,
|
| 1995 |
+
"learning_rate": 1.184016090307059e-05,
|
| 1996 |
+
"loss": 1.2341,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.843819393042191,
|
| 2001 |
+
"grad_norm": 1.0862884521484375,
|
| 2002 |
+
"learning_rate": 1.1415257882705311e-05,
|
| 2003 |
+
"loss": 1.7456,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.846780162842339,
|
| 2008 |
+
"grad_norm": 1.1565598249435425,
|
| 2009 |
+
"learning_rate": 1.09971345907394e-05,
|
| 2010 |
+
"loss": 1.6508,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8497409326424871,
|
| 2015 |
+
"grad_norm": 1.1448659896850586,
|
| 2016 |
+
"learning_rate": 1.0585864495652897e-05,
|
| 2017 |
+
"loss": 1.9964,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8527017024426351,
|
| 2022 |
+
"grad_norm": 0.968397855758667,
|
| 2023 |
+
"learning_rate": 1.0181519861750078e-05,
|
| 2024 |
+
"loss": 1.6825,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8556624722427831,
|
| 2029 |
+
"grad_norm": 0.9276019930839539,
|
| 2030 |
+
"learning_rate": 9.784171736461762e-06,
|
| 2031 |
+
"loss": 1.2859,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8586232420429312,
|
| 2036 |
+
"grad_norm": 0.7998013496398926,
|
| 2037 |
+
"learning_rate": 9.393889937861694e-06,
|
| 2038 |
+
"loss": 0.6289,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8615840118430792,
|
| 2043 |
+
"grad_norm": 0.6690629124641418,
|
| 2044 |
+
"learning_rate": 9.010743042398684e-06,
|
| 2045 |
+
"loss": 0.581,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8645447816432272,
|
| 2050 |
+
"grad_norm": 0.6341821551322937,
|
| 2051 |
+
"learning_rate": 8.634798372847148e-06,
|
| 2052 |
+
"loss": 0.3543,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8675055514433753,
|
| 2057 |
+
"grad_norm": 0.5753948092460632,
|
| 2058 |
+
"learning_rate": 8.266121986477699e-06,
|
| 2059 |
+
"loss": 0.3731,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8704663212435233,
|
| 2064 |
+
"grad_norm": 0.7309054136276245,
|
| 2065 |
+
"learning_rate": 7.904778663450324e-06,
|
| 2066 |
+
"loss": 0.6859,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8734270910436713,
|
| 2071 |
+
"grad_norm": 0.5517750382423401,
|
| 2072 |
+
"learning_rate": 7.550831895431798e-06,
|
| 2073 |
+
"loss": 0.2878,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8763878608438194,
|
| 2078 |
+
"grad_norm": 0.5744172930717468,
|
| 2079 |
+
"learning_rate": 7.204343874439578e-06,
|
| 2080 |
+
"loss": 0.202,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8793486306439674,
|
| 2085 |
+
"grad_norm": 0.4893665909767151,
|
| 2086 |
+
"learning_rate": 6.865375481914016e-06,
|
| 2087 |
+
"loss": 0.1466,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8823094004441154,
|
| 2092 |
+
"grad_norm": 0.503653347492218,
|
| 2093 |
+
"learning_rate": 6.533986278020876e-06,
|
| 2094 |
+
"loss": 0.1117,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8852701702442635,
|
| 2099 |
+
"grad_norm": 0.754020631313324,
|
| 2100 |
+
"learning_rate": 6.210234491186079e-06,
|
| 2101 |
+
"loss": 0.1614,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8882309400444115,
|
| 2106 |
+
"grad_norm": 1.9576098918914795,
|
| 2107 |
+
"learning_rate": 5.894177007864271e-06,
|
| 2108 |
+
"loss": 0.5595,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8911917098445595,
|
| 2113 |
+
"grad_norm": 0.9034455418586731,
|
| 2114 |
+
"learning_rate": 5.585869362543416e-06,
|
| 2115 |
+
"loss": 1.3168,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8941524796447077,
|
| 2120 |
+
"grad_norm": 1.1294760704040527,
|
| 2121 |
+
"learning_rate": 5.285365727986707e-06,
|
| 2122 |
+
"loss": 1.9353,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8971132494448557,
|
| 2127 |
+
"grad_norm": 1.308358073234558,
|
| 2128 |
+
"learning_rate": 4.9927189057139665e-06,
|
| 2129 |
+
"loss": 2.1867,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9000740192450037,
|
| 2134 |
+
"grad_norm": 1.1667540073394775,
|
| 2135 |
+
"learning_rate": 4.707980316723837e-06,
|
| 2136 |
+
"loss": 2.2503,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9030347890451518,
|
| 2141 |
+
"grad_norm": 1.0320072174072266,
|
| 2142 |
+
"learning_rate": 4.4311999924586065e-06,
|
| 2143 |
+
"loss": 1.5944,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9059955588452998,
|
| 2148 |
+
"grad_norm": 1.141965627670288,
|
| 2149 |
+
"learning_rate": 4.16242656601315e-06,
|
| 2150 |
+
"loss": 1.8177,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9089563286454478,
|
| 2155 |
+
"grad_norm": 0.9115905165672302,
|
| 2156 |
+
"learning_rate": 3.901707263589671e-06,
|
| 2157 |
+
"loss": 1.436,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9119170984455959,
|
| 2162 |
+
"grad_norm": 0.9493201375007629,
|
| 2163 |
+
"learning_rate": 3.6490878961994878e-06,
|
| 2164 |
+
"loss": 1.9393,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9148778682457439,
|
| 2169 |
+
"grad_norm": 1.2201560735702515,
|
| 2170 |
+
"learning_rate": 3.4046128516136755e-06,
|
| 2171 |
+
"loss": 2.0329,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9178386380458919,
|
| 2176 |
+
"grad_norm": 1.1415530443191528,
|
| 2177 |
+
"learning_rate": 3.1683250865636114e-06,
|
| 2178 |
+
"loss": 1.7484,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.92079940784604,
|
| 2183 |
+
"grad_norm": 1.055849313735962,
|
| 2184 |
+
"learning_rate": 2.9402661191930804e-06,
|
| 2185 |
+
"loss": 1.7875,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.923760177646188,
|
| 2190 |
+
"grad_norm": 0.9362370371818542,
|
| 2191 |
+
"learning_rate": 2.7204760217631074e-06,
|
| 2192 |
+
"loss": 1.2803,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.926720947446336,
|
| 2197 |
+
"grad_norm": 1.1764779090881348,
|
| 2198 |
+
"learning_rate": 2.5089934136108664e-06,
|
| 2199 |
+
"loss": 1.6859,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9296817172464841,
|
| 2204 |
+
"grad_norm": 1.2218987941741943,
|
| 2205 |
+
"learning_rate": 2.30585545436387e-06,
|
| 2206 |
+
"loss": 2.025,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9326424870466321,
|
| 2211 |
+
"grad_norm": 1.0182489156723022,
|
| 2212 |
+
"learning_rate": 2.1110978374106192e-06,
|
| 2213 |
+
"loss": 1.3178,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9356032568467801,
|
| 2218 |
+
"grad_norm": 1.1746158599853516,
|
| 2219 |
+
"learning_rate": 1.9247547836289793e-06,
|
| 2220 |
+
"loss": 2.0641,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9385640266469282,
|
| 2225 |
+
"grad_norm": 1.0268217325210571,
|
| 2226 |
+
"learning_rate": 1.7468590353731495e-06,
|
| 2227 |
+
"loss": 1.5677,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9415247964470762,
|
| 2232 |
+
"grad_norm": 1.1525732278823853,
|
| 2233 |
+
"learning_rate": 1.5774418507205679e-06,
|
| 2234 |
+
"loss": 1.8057,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9444855662472242,
|
| 2239 |
+
"grad_norm": 0.9357332587242126,
|
| 2240 |
+
"learning_rate": 1.4165329979794973e-06,
|
| 2241 |
+
"loss": 1.3016,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9474463360473723,
|
| 2246 |
+
"grad_norm": 0.9553401470184326,
|
| 2247 |
+
"learning_rate": 1.2641607504584928e-06,
|
| 2248 |
+
"loss": 1.5646,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9504071058475203,
|
| 2253 |
+
"grad_norm": 0.9898755550384521,
|
| 2254 |
+
"learning_rate": 1.1203518814984214e-06,
|
| 2255 |
+
"loss": 1.5453,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9533678756476683,
|
| 2260 |
+
"grad_norm": 1.019761085510254,
|
| 2261 |
+
"learning_rate": 9.851316597681958e-07,
|
| 2262 |
+
"loss": 1.5678,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9563286454478165,
|
| 2267 |
+
"grad_norm": 1.0154638290405273,
|
| 2268 |
+
"learning_rate": 8.585238448247435e-07,
|
| 2269 |
+
"loss": 1.6164,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9592894152479645,
|
| 2274 |
+
"grad_norm": 1.1107819080352783,
|
| 2275 |
+
"learning_rate": 7.405506829382735e-07,
|
| 2276 |
+
"loss": 1.8083,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9622501850481125,
|
| 2281 |
+
"grad_norm": 1.2360882759094238,
|
| 2282 |
+
"learning_rate": 6.312329031833319e-07,
|
| 2283 |
+
"loss": 2.0795,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9652109548482606,
|
| 2288 |
+
"grad_norm": 0.8985037207603455,
|
| 2289 |
+
"learning_rate": 5.305897137965199e-07,
|
| 2290 |
+
"loss": 1.1981,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9681717246484086,
|
| 2295 |
+
"grad_norm": 1.0836095809936523,
|
| 2296 |
+
"learning_rate": 4.386387988014273e-07,
|
| 2297 |
+
"loss": 2.1089,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9711324944485566,
|
| 2302 |
+
"grad_norm": 1.0523475408554077,
|
| 2303 |
+
"learning_rate": 3.553963149013295e-07,
|
| 2304 |
+
"loss": 1.8185,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9740932642487047,
|
| 2309 |
+
"grad_norm": 1.2510104179382324,
|
| 2310 |
+
"learning_rate": 2.808768886403301e-07,
|
| 2311 |
+
"loss": 2.1252,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9770540340488527,
|
| 2316 |
+
"grad_norm": 1.1012005805969238,
|
| 2317 |
+
"learning_rate": 2.1509361383330596e-07,
|
| 2318 |
+
"loss": 1.8081,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9800148038490007,
|
| 2323 |
+
"grad_norm": 0.6544106602668762,
|
| 2324 |
+
"learning_rate": 1.580580492652084e-07,
|
| 2325 |
+
"loss": 0.4715,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9829755736491488,
|
| 2330 |
+
"grad_norm": 0.6254527568817139,
|
| 2331 |
+
"learning_rate": 1.0978021666005478e-07,
|
| 2332 |
+
"loss": 0.4152,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9859363434492968,
|
| 2337 |
+
"grad_norm": 0.6432924866676331,
|
| 2338 |
+
"learning_rate": 7.02685989200258e-08,
|
| 2339 |
+
"loss": 0.3737,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.9888971132494448,
|
| 2344 |
+
"grad_norm": 0.595024585723877,
|
| 2345 |
+
"learning_rate": 3.953013863490784e-08,
|
| 2346 |
+
"loss": 0.2546,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.9918578830495929,
|
| 2351 |
+
"grad_norm": 0.4974139332771301,
|
| 2352 |
+
"learning_rate": 1.7570236862241017e-08,
|
| 2353 |
+
"loss": 0.2191,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9948186528497409,
|
| 2358 |
+
"grad_norm": 0.5824573636054993,
|
| 2359 |
+
"learning_rate": 4.392752178278281e-09,
|
| 2360 |
+
"loss": 0.1276,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.997779422649889,
|
| 2365 |
+
"grad_norm": 0.7297351360321045,
|
| 2366 |
+
"learning_rate": 0.0,
|
| 2367 |
+
"loss": 0.1176,
|
| 2368 |
+
"step": 337
|
| 2369 |
}
|
| 2370 |
],
|
| 2371 |
"logging_steps": 1,
|
|
|
|
| 2380 |
"should_evaluate": false,
|
| 2381 |
"should_log": false,
|
| 2382 |
"should_save": true,
|
| 2383 |
+
"should_training_stop": true
|
| 2384 |
},
|
| 2385 |
"attributes": {}
|
| 2386 |
}
|
| 2387 |
},
|
| 2388 |
+
"total_flos": 3.0172541179920384e+16,
|
| 2389 |
"train_batch_size": 4,
|
| 2390 |
"trial_name": null,
|
| 2391 |
"trial_params": null
|