Update README.md
Browse files
README.md
CHANGED
|
@@ -8,6 +8,11 @@ license: apache-2.0
|
|
| 8 |
|
| 9 |
This model is a pruned Mistral model re-aligned using the Zephyr Recipe
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
## W&B workspace
|
|
|
|
| 8 |
|
| 9 |
This model is a pruned Mistral model re-aligned using the Zephyr Recipe
|
| 10 |
|
| 11 |
+
## details
|
| 12 |
+
|
| 13 |
+
- This model has 2 stages training: SFT and DPO
|
| 14 |
+
- The initial model consist on selecting some layers of the mistral model to make a smaller model
|
| 15 |
+
- the code can be found here: github.com/tcapelle/shear
|
| 16 |
|
| 17 |
|
| 18 |
## W&B workspace
|