Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,14 @@ language:
|
|
| 10 |
---
|
| 11 |
# Codestral-22B-v0.1-abliterated-v3 Model Card
|
| 12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
[My original Jupyter "cookbook" to replicate the methodology can be found here](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)
|
| 14 |
|
| 15 |
[My personal library o' code used](https://github.com/FailSpy/abliterator) (WIP, looking to improve and generalize)
|
|
|
|
| 10 |
---
|
| 11 |
# Codestral-22B-v0.1-abliterated-v3 Model Card
|
| 12 |
|
| 13 |
+
## Quantization
|
| 14 |
+
|
| 15 |
+
This model is using github-code as a calibration dataset, hoping to provide better performance than other models using wikitext
|
| 16 |
+
|
| 17 |
+
BTW, I added chat template to tokenizer_config.json
|
| 18 |
+
|
| 19 |
+
## Original Model Card
|
| 20 |
+
|
| 21 |
[My original Jupyter "cookbook" to replicate the methodology can be found here](https://huggingface.co/failspy/llama-3-70B-Instruct-abliterated/blob/main/ortho_cookbook.ipynb)
|
| 22 |
|
| 23 |
[My personal library o' code used](https://github.com/FailSpy/abliterator) (WIP, looking to improve and generalize)
|