# andyrdt/refusal\_direction

<https://repo.masters-masters.com/en/repos/andyrdt/refusal_direction>

[Home](https://repo.masters-masters.com/en) › [andyrdt](https://repo.masters-masters.com/en/owners/andyrdt) › [refusal\_direction](https://repo.masters-masters.com/en/repos/andyrdt/refusal_direction)

andyrdt/refusal\_direction has 480 stars on GitHub. Its RepoMasters card: Alpha tier. Code and results accompanying the paper "Refusal in Language Models Is…

| Field | Value |
| --- | --- |
| Owner | [andyrdt](https://repo.masters-masters.com/en/owners/andyrdt) |
| Stars | 480 |
| Forks | 126 |
| Language | [Python](https://repo.masters-masters.com/en/languages/python) |
| License | Apache-2.0 |
| Created | 2024 |
| Tier | [Alpha (250 to 699 ★)](https://repo.masters-masters.com/en/ranks/b_movie) |
| View on GitHub | [andyrdt/refusal\_direction](https://github.com/andyrdt/refusal_direction) |

## Repository

Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".

The tier for repos with 250 to 699 ★. Published odds per card pulled: 20%. Rarity: Uncommon.

## Nearby in Python

- [davnords/LoMa · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/davnords/LoMa)
- [xiaomi-research/onevl · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/xiaomi-research/onevl)
- [Tencent-Hunyuan/Hy-Nomas · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/Tencent-Hunyuan/Hy-Nomas)
- [clustershell/clustershell · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/clustershell/clustershell)
- [nephila/djangocms-blog · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/nephila/djangocms-blog)
- [DjangoGirls/djangogirls · ★ 480 · Python · Alpha](https://repo.masters-masters.com/en/repos/DjangoGirls/djangogirls)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.