adamkarvonen / SAEBenchLinks
☆109Updated 3 weeks ago
Alternatives and similar repositories for SAEBench
Users that are interested in SAEBench are comparing it to the libraries listed below
Sorting:
- ☆183Updated 3 weeks ago
- ☆154Updated 8 months ago
- ☆234Updated 10 months ago
- Sparse Autoencoder for Mechanistic Interpretability☆257Updated last year
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆200Updated this week
- Using sparse coding to find distributed representations used by neural networks.☆261Updated last year
- ☆103Updated 5 months ago
- ☆51Updated 8 months ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆57Updated 9 months ago
- ☆324Updated 3 weeks ago
- Steering Llama 2 with Contrastive Activation Addition☆167Updated last year
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆207Updated 7 months ago
- Modified to support crosscoder training.☆21Updated last week
- Sparse Autoencoder Training Library☆54Updated 3 months ago
- ☆124Updated last year
- Sparse probing paper full code.☆58Updated last year
- Steering vectors for transformer language models in Pytorch / Huggingface☆119Updated 5 months ago
- A library for efficient patching and automatic circuit discovery.☆73Updated 2 weeks ago
- Unified access to Large Language Model modules using NNsight☆37Updated 2 weeks ago
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆40Updated last year
- ☆81Updated 5 months ago
- Evaluate interpretability methods on localizing and disentangling concepts in LLMs.☆52Updated 10 months ago
- Attribution-based Parameter Decomposition☆28Updated last month
- ☆121Updated last year
- ☆15Updated 2 months ago
- ☆47Updated 2 weeks ago
- Mechanistic Interpretability Visualizations using React☆272Updated 7 months ago
- Improving Steering Vectors by Targeting Sparse Autoencoder Features☆24Updated 8 months ago
- ☆64Updated this week
- LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces☆96Updated last year