EleutherAI / delphiLinks
Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models know themselves through automated interpretability.
☆218Updated last week
Alternatives and similar repositories for delphi
Users that are interested in delphi are comparing it to the libraries listed below
Sorting:
- ☆190Updated last week
- ☆179Updated 11 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆222Updated 10 months ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆59Updated 11 months ago
- ☆247Updated last year
- ☆131Updated this week
- A library for efficient patching and automatic circuit discovery.☆78Updated 3 months ago
- Sparse Autoencoder Training Library☆55Updated 5 months ago
- Steering Llama 2 with Contrastive Activation Addition☆188Updated last year
- Steering vectors for transformer language models in Pytorch / Huggingface☆125Updated 8 months ago
- ☆54Updated 11 months ago
- Using sparse coding to find distributed representations used by neural networks.☆279Updated last year
- ☆126Updated last year
- ☆128Updated last year
- ☆354Updated 2 months ago
- Code for reproducing our paper "Not All Language Model Features Are Linear"☆82Updated 10 months ago
- Stanford NLP Python library for benchmarking the utility of LLM interpretability methods☆136Updated 3 months ago
- ☆92Updated last year
- Function Vectors in Large Language Models (ICLR 2024)☆181Updated 6 months ago
- Sparse Autoencoder for Mechanistic Interpretability☆272Updated last year
- Mechanistic Interpretability Visualizations using React☆293Updated 10 months ago
- For OpenMOSS Mechanistic Interpretability Team's Sparse Autoencoder (SAE) research.☆157Updated this week
- Sparsify transformers with SAEs and transcoders☆640Updated last week
- ☆107Updated 8 months ago
- A toolkit for describing model features and intervening on those features to steer behavior.☆207Updated 11 months ago
- Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".☆290Updated 4 months ago
- ☆73Updated last week
- Modified to support crosscoder training.☆23Updated 2 weeks ago
- ☆234Updated last year
- Attribution-based Parameter Decomposition☆31Updated 4 months ago