neelnanda-io / 1L-Sparse-AutoencoderLinks
☆126Updated last year
Alternatives and similar repositories for 1L-Sparse-Autoencoder
Users that are interested in 1L-Sparse-Autoencoder are comparing it to the libraries listed below
Sorting:
- ☆237Updated 10 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆211Updated 8 months ago
- Mechanistic Interpretability Visualizations using React☆280Updated 8 months ago
- ☆327Updated this week
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆206Updated this week
- ☆111Updated last month
- Sparse Autoencoder for Mechanistic Interpretability☆258Updated last year
- Using sparse coding to find distributed representations used by neural networks.☆262Updated last year
- Sparse Autoencoder Training Library☆54Updated 3 months ago
- ☆52Updated 9 months ago
- ☆184Updated last month
- Notebooks accompanying Anthropic's "Toy Models of Superposition" paper☆129Updated 2 years ago
- ☆103Updated 6 months ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆58Updated 9 months ago
- ☆162Updated 9 months ago
- Steering vectors for transformer language models in Pytorch / Huggingface☆122Updated 6 months ago
- ☆122Updated last year
- A library for efficient patching and automatic circuit discovery.☆76Updated last month
- ☆276Updated last year
- Steering Llama 2 with Contrastive Activation Addition☆174Updated last year
- Modified to support crosscoder training.☆21Updated 3 weeks ago
- ☆28Updated last year
- Erasing concepts from neural representations with provable guarantees☆232Updated 6 months ago
- Sparsify transformers with SAEs and transcoders☆608Updated this week
- Attribution-based Parameter Decomposition☆28Updated 2 months ago
- Sparse probing paper full code.☆59Updated last year
- ☆32Updated last year
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆630Updated this week
- ☆66Updated last week
- Code for my NeurIPS 2024 ATTRIB paper titled "Attribution Patching Outperforms Automated Circuit Discovery"☆40Updated last year