|
--- |
|
license: apache-2.0 |
|
tags: |
|
- mistral |
|
- conversational |
|
- exl2 |
|
- rpcal |
|
language: |
|
- en |
|
--- |
|
> [!WARNING] |
|
> **EXPERIMENT** <br> |
|
> This is my experimental quantization of [Mistral-Nemo-Instruct-2407-OAS](https://huggingface.co/Kooten/Mistral-Nemo-Instruct-2407-norefuse-OAS/tree/main) by [Kooten](https://huggingface.co/Kooten) <br> |
|
> Quantized using the [cleaned PIPPA](https://huggingface.co/datasets/royallab/PIPPA-cleaned) roleplay dataset. |
|
|
|
## Quantized variant |
|
|
|
- [8.0 bits per weight](https://huggingface.co/bluuwhale/Mistral-Nemo-Instruct-OAS-exl2-rpcal/tree/8.0bpw) |
|
- [5.0 bits per weight](https://huggingface.co/bluuwhale/Mistral-Nemo-Instruct-OAS-exl2-rpcal/tree/5.0bpw) |
|
|
|
## Original card |
|
|
|
Orthogonal activation steering to remove refusals (Unscensor) |
|
|
|
Original model: [mistralai/Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407) |
