andyrdt / refusal_directionLinks
Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".
☆290Updated 4 months ago
Alternatives and similar repositories for refusal_direction
Users that are interested in refusal_direction are comparing it to the libraries listed below
Sorting:
- Improving Alignment and Robustness with Circuit Breakers☆238Updated last year
- Steering Llama 2 with Contrastive Activation Addition☆188Updated last year
- Steering vectors for transformer language models in Pytorch / Huggingface☆126Updated 8 months ago
- LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces☆98Updated 2 years ago
- ☆181Updated 11 months ago
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆219Updated this week
- For OpenMOSS Mechanistic Interpretability Team's Sparse Autoencoder (SAE) research.☆157Updated this week
- [ICLR 2025] General-purpose activation steering library☆111Updated last month
- A toolkit for describing model features and intervening on those features to steer behavior.☆209Updated 11 months ago
- Röttger et al. (NAACL 2024): "XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models"☆116Updated 8 months ago
- Using sparse coding to find distributed representations used by neural networks.☆280Updated last year
- A resource repository for representation engineering in large language models☆138Updated 11 months ago
- ☆234Updated last year
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆136Updated 4 months ago
- ☆191Updated last week
- Persona Vectors: Monitoring and Controlling Character Traits in Language Models☆266Updated 2 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆222Updated 10 months ago
- This repository contains the code and data for the paper "SelfIE: Self-Interpretation of Large Language Model Embeddings" by Haozhe Chen,…☆52Updated 10 months ago
- WMDP is a LLM proxy benchmark for hazardous knowledge in bio, cyber, and chemical security. We also release code for RMU, an unlearning m…☆146Updated 4 months ago
- ☆131Updated last week
- Code for In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering☆190Updated 8 months ago
- Sparse Autoencoder for Mechanistic Interpretability☆276Updated last year
- A simple unified framework for evaluating LLMs☆251Updated 6 months ago
- ☆206Updated 11 months ago
- ☆195Updated 6 months ago
- ☆530Updated last year
- Code release for "Debating with More Persuasive LLMs Leads to More Truthful Answers"☆117Updated last year
- A curated list of LLM Interpretability related material - Tutorial, Library, Survey, Paper, Blog, etc..☆273Updated 7 months ago
- We jailbreak GPT-3.5 Turbo’s safety guardrails by fine-tuning it on only 10 adversarially designed examples, at a cost of less than $0.20…☆326Updated last year
- ☆102Updated 3 months ago