34 lines
1.2 KiB
Markdown
34 lines
1.2 KiB
Markdown
|
---
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
tags:
|
||
|
|
- uncensored
|
||
|
|
- abliterated
|
||
|
|
base_model:
|
||
|
|
- allura-forge/Llama-3.3-8B-Instruct
|
||
|
|
---
|
||
|
|
This is an abliterated version of [Llama-3.3-8B-Instruct](https://huggingface.co/allura-forge/Llama-3.3-8B-Instruct), made using [Heretic v1.1.0](https://github.com/p-e-w/heretic)
|
||
|
|
|
||
|
|
This is the model with [128k context](https://huggingface.co/shb777/Llama-3.3-8B-Instruct).
|
||
|
|
|
||
|
|
The quantizations were created using an imatrix merged from [combined\_en\_medium](https://huggingface.co/datasets/eaddario/imatrix-calibration/blob/main/combined_en_medium.parquet) and [harmful.txt](https://github.com/Sumandora/remove-refusals-with-transformers) to leverage the abliterated nature of the model.
|
||
|
|
|
||
|
|
## Performance
|
||
|
|
|
||
|
|
| Metric | This model | Original model |
|
||
|
|
| :----- | :--------: | :---------------------------: |
|
||
|
|
| **Refusals** | 5/100 | 95/100 |
|
||
|
|
| KL divergence | 0.0919 | 0 |
|
||
|
|
|
||
|
|
|
||
|
|
## BibTeX entry and citation info
|
||
|
|
```bibtex
|
||
|
|
@misc{heretic,
|
||
|
|
author = {Weidmann, Philipp Emanuel},
|
||
|
|
title = {Heretic: Fully automatic censorship removal for language models},
|
||
|
|
year = {2025},
|
||
|
|
publisher = {GitHub},
|
||
|
|
journal = {GitHub repository},
|
||
|
|
howpublished = {\url{https://github.com/p-e-w/heretic}}
|
||
|
|
}
|
||
|
|
```
|