Arthur Conmy

> Making it so `model.run_with_saes` resets the SAE states to the same state as before the function was called would be a big improvement @nix-apollo I suggested this here https://github.com/neelnanda-io/TransformerLens/pull/536#discussion_r1569800199...

@neelnanda-io I moved discussion to the Open Source slack about this issue as the interface is better. TLDR support both permanently attached SAEs with add_sae and transiently attached SAEs with...

Unclear what the solution should be. There are plausibly three different parameter counts that are helpful: 1. Parameters in training 2. Parameters ignoring embeddings 3. Parameters used now (e.g folding...

I think https://github.com/EleutherAI/lm-evaluation-harness would be a good place to start here. Anyone doing this should be aware that this is going to be refactored: https://www.youtube.com/watch?v=6qDOUeQTp0I so should probably chat to...

I don't think the refactor is done. I guess HF Evals are different fromt the way lm-eval-harness downloads datasets, maybe we should do this? All three datasets mentioned are in...

 The code snippet in the upload guide has a typo. It should say `api.upload_large_folder` not `api.upload_folder`