Update README.md
Browse files
README.md
CHANGED
|
@@ -17,4 +17,6 @@ TODO: make good model card
|
|
| 17 |
|
| 18 |
Trained from Apertus-8B Base using liger kernel with a two step Anchored SFT > LD-DPO recipe.
|
| 19 |
|
| 20 |
-
Alpaca template, no system.
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
Trained from Apertus-8B Base using liger kernel with a two step Anchored SFT > LD-DPO recipe.
|
| 19 |
|
| 20 |
+
Alpaca template, no system.
|
| 21 |
+
|
| 22 |
+
.7 temp, top_p .95, no rep pen or dry
|