Update README.md
Browse files
README.md
CHANGED
|
@@ -52,6 +52,8 @@ Trained [Qwen2.5-14B-Instruct](https://huggingface.co/Qwen/Qwen2.5-14B-Instruct)
|
|
| 52 |
|
| 53 |
* Merged all <b>DPO checkpoints</b> and <b>SLERP</b> variations with <b>MODEL_STOCK</b> to analyze geometric properties and get the most *performant* aspects of all runs/merges. *Model Stock* was chosen due to the similarity between the merged models.
|
| 54 |
|
|
|
|
|
|
|
| 55 |
## Recipe
|
| 56 |
|
| 57 |
```yaml
|
|
|
|
| 52 |
|
| 53 |
* Merged all <b>DPO checkpoints</b> and <b>SLERP</b> variations with <b>MODEL_STOCK</b> to analyze geometric properties and get the most *performant* aspects of all runs/merges. *Model Stock* was chosen due to the similarity between the merged models.
|
| 54 |
|
| 55 |
+
* This was chosen due to the fact that evaluation for *ORPO* is unclear, so it's hard to know which runs are the best.
|
| 56 |
+
|
| 57 |
## Recipe
|
| 58 |
|
| 59 |
```yaml
|