Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,14 @@ language:
|
|
| 17 |
This is the series of 'Base + Language + Instruct', chat vector and various methods in mergekit.
|
| 18 |
Thanks again! @beomi
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
The original paper author claims density should be around 0.2~0.3, but in reality high number gets some better result. You should try other params for better result than this!
|
| 21 |
|
| 22 |
```python
|
|
|
|
| 17 |
This is the series of 'Base + Language + Instruct', chat vector and various methods in mergekit.
|
| 18 |
Thanks again! @beomi
|
| 19 |
|
| 20 |
+
| Model | Merge Method | Score(but what?) |
|
| 21 |
+
|---|---|---|
|
| 22 |
+
| [beomi/Llama-3-Open-Ko-8B-Instruct-preview](https://huggingface.co/beomi/Llama-3-Open-Ko-8B-Instruct-preview) | chat vector | - |
|
| 23 |
+
| [kuotient/Llama-3-Ko-8B-ties](https://huggingface.co/kuotient/Llama-3-Ko-8B-ties) | Ties | - |
|
| 24 |
+
| [kuotient/Llama-3-Ko-8B-dare-ties](https://huggingface.co/kuotient/Llama-3-Ko-8B-dare-ties) | Dare-ties | - |
|
| 25 |
+
| [kuotient/Llama-3-Ko-8B-TA](https://huggingface.co/kuotient/Llama-3-Ko-8B-TA) | Task Arithmetic(maybe...? not sure about this) | - |
|
| 26 |
+
| WIP | Model stock(I don't read this paper yet but still) | - |
|
| 27 |
+
|
| 28 |
The original paper author claims density should be around 0.2~0.3, but in reality high number gets some better result. You should try other params for better result than this!
|
| 29 |
|
| 30 |
```python
|