SAELens

add experimental embedding SAEs

#4

SAEs on the raw embedding vectors (not including the sqrt(d_model) scaling that happens inside the model fwd pass).
Not trained on BOS, PAD, EOS and any token that didn't occur in the training data.

Aric changed pull request status to open
ArthurConmyGDM changed pull request status to merged

Sign up or log in to comment