Steering Llama 2 with Contrastive Activation Addition
☆254May 23, 2024Updated 2 years ago
Alternatives and similar repositories for CAA
Users that are interested in CAA are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization☆51Jul 28, 2024Updated 2 years ago
- Steering vectors for transformer language models in Pytorch / Huggingface☆163Feb 21, 2025Updated last year
- LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces☆107Sep 21, 2023Updated 3 years ago
- [ICLR 2025] General-purpose activation steering library☆191Updated this week
- Materials for "Multi-property Steering of Large Language Models with Dynamic Activation Composition"☆14Nov 22, 2024Updated last year
- Proton VPN Special Offer - Get 70% off • AdSpecial partner offer. Trusted by over 100 million users worldwide. Tested, Approved and Recommended by Experts.
- ☆19Sep 1, 2025Updated last year
- Improving Steering Vectors by Targeting Sparse Autoencoder Features☆30Nov 20, 2024Updated last year
- Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".☆458Jun 13, 2025Updated last year
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆218Mar 12, 2026Updated 7 months ago
- ☆225Oct 14, 2025Updated last year
- Algebraic value editing in pretrained language models☆72Nov 1, 2023Updated 2 years ago
- A resource repository for representation engineering in large language models☆156Nov 14, 2024Updated last year
- ☆30Aug 2, 2024Updated 2 years ago
- Training Sparse Autoencoders on Language Models☆1,555Updated this week
- End-to-end encrypted email - Proton Mail • AdSpecial offer: 40% Off Yearly / 80% Off First Month. All Proton services are open source and independently audited for security.
- ☆298Oct 1, 2024Updated 2 years ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆276Feb 27, 2026Updated 7 months ago
- A curated list of resources for activation engineering☆140Oct 2, 2025Updated last year
- An exploration of LLM steering☆29Jun 15, 2024Updated 2 years ago
- Representation Engineering: A Top-Down Approach to AI Transparency☆1,048Aug 14, 2024Updated 2 years ago
- Code release for the paper "Style Vectors for Steering Generative Large Language Models", accepted to the Findings of the EACL 2024.☆37Sep 26, 2024Updated 2 years ago
- [NAACL'25 Oral] Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering☆86Jun 20, 2026Updated 3 months ago
- Official codebase for "Analyzing the Generalization and Reliability of Steering Vectors"☆23Dec 14, 2024Updated last year
- [ACL 2025] Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms☆43Jun 4, 2025Updated last year
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Evaluate interpretability methods on localizing and disentangling concepts in LLMs.☆58Oct 30, 2025Updated 11 months ago
- Sparse Autoencoder for Mechanistic Interpretability☆307Jul 20, 2024Updated 2 years ago
- Inference-Time Intervention: Eliciting Truthful Answers from a Language Model☆587Jan 28, 2025Updated last year
- This repository contains the code used for the experiments in the paper "Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity…☆32Oct 27, 2025Updated 11 months ago
- Official code for Steering Large Language Models using Conceptors, presented at the NeurIPS 2024 MINT Workshop.☆16Mar 13, 2025Updated last year
- Stanford NLP Python library for understanding and improving PyTorch models via interventions☆905Mar 6, 2026Updated 7 months ago
- A library for mechanistic interpretability of GPT-style language models☆3,963Updated this week
- Sparsify transformers with SAEs and transcoders☆741Updated this week
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆1,123Updated this week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- Official repo for EMNLP'24 paper "SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning"☆31Oct 1, 2024Updated 2 years ago
- A collection of different ways to implement accessing and modifying internal model activations for LLMs☆24Oct 18, 2024Updated last year
- Tools for understanding how transformer predictions are built layer-by-layer☆618Aug 7, 2025Updated last year
- ☆118Aug 8, 2024Updated 2 years ago
- ☆29Oct 6, 2024Updated 2 years ago
- Function Vectors in Large Language Models (ICLR 2024)☆200Apr 30, 2026Updated 5 months ago
- Official repo for NeurIPS'24 paper "WAGLE: Strategic Weight Attribution for Effective and Modular Unlearning in Large Language Models"☆19Dec 16, 2024Updated last year