Update README.md
Browse files
README.md
CHANGED
@@ -18,7 +18,7 @@ palmer is a series of ~1b parameters language models fine-tuned to be used as ba
|
|
18 |
|palmer-001| 0.2807| 0.5524| 0.7106| 0.5896|
|
19 |
|tinyllama-2.5t|0.3191|0.5896| 0.7307| 0.5872|
|
20 |
|palmer-002|0.3242|**0.5956**|**0.7345**|0.5888|
|
21 |
-
|palmer-002-ultra|**0.3319**|
|
22 |
|
23 |
### training
|
24 |
Training took ~3.5 P100 gpu hours. It was trained on 15,000 gpt-4 shuffled samples. palmer was fine-tuned using lower learning rates ensuring it keeps as much general knowledge as possible.
|
|
|
18 |
|palmer-001| 0.2807| 0.5524| 0.7106| 0.5896|
|
19 |
|tinyllama-2.5t|0.3191|0.5896| 0.7307| 0.5872|
|
20 |
|palmer-002|0.3242|**0.5956**|**0.7345**|0.5888|
|
21 |
+
|palmer-002-ultra|**0.3319**| 0.5877 |0.7252|**0.6038**|
|
22 |
|
23 |
### training
|
24 |
Training took ~3.5 P100 gpu hours. It was trained on 15,000 gpt-4 shuffled samples. palmer was fine-tuned using lower learning rates ensuring it keeps as much general knowledge as possible.
|