search

ai-safety-foundation / sparse_autoencoder

Sparse Autoencoder for Mechanistic Interpretability
https://ai-safety-foundation.github.io/sparse_autoencoder/
MIT License
191 stars 39 forks source link

Move axis names to be used directly #130

Closed alan-cooney closed 1 year ago

alan-cooney commented 1 year ago

This makes the docs clearer