SAEfarer

Development installation

Create a virtual environment and and install saefarer in editable mode with the optional development dependencies:

python -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"

You then need to install the JavaScript dependencies and run the development server.

npm install
npm run dev

Open example.ipynb in JupyterLab, VS Code, or your favorite editor to start developing. Changes made in js/ will be reflected in the notebook.

References

This project uses code from the following repositories:

• OpenAI's sparse_autoencoder
• EleutherAI's sae
• SAELens
• 1L-Sparse-Autoencoder
• TransformerLens
• sae_vis
• SAEDashboard