Commit
•
931674e
1
Parent(s):
37418f4
Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,12 @@ library_name: transformers
|
|
| 14 |
---
|
| 15 |
|
| 16 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
# gemma-2-27b-it-abliterated
|
| 19 |
Check out the <a href="https://huggingface.co/byroneverson/gemma-2-27b-it-abliterated/blob/main/abliterate-gemma-2-27b-it.ipynb">jupyter notebook</a> for details of how this model was abliterated from glm-4-9b-chat.
|
|
|
|
| 14 |
---
|
| 15 |
|
| 16 |
|
| 17 |
+
NOTE: This is a current WIP (work in progress).
|
| 18 |
+
This abliteration was 1/2 performed with llama-cpp-python (obtain direction vector) and 1/2 performed with torch (modifying .safetensors).
|
| 19 |
+
It is a rather larger model so it may take me another day or two to figure out which layer I should be using for the direction vector.
|
| 20 |
+
This is round 1, layer 20 was used for the direction vector.
|
| 21 |
+
I have not tested it yet and cannot test it until I can make a GGUF of this repo.
|
| 22 |
+
From there I will determine if I need to change the layer used, etc.
|
| 23 |
|
| 24 |
# gemma-2-27b-it-abliterated
|
| 25 |
Check out the <a href="https://huggingface.co/byroneverson/gemma-2-27b-it-abliterated/blob/main/abliterate-gemma-2-27b-it.ipynb">jupyter notebook</a> for details of how this model was abliterated from glm-4-9b-chat.
|