[Linkpost] Play with SAEs on Llama 3

post by Tom McGrath, Eric Ho (eh42), Dan Balsam (dan-balsam) · 2024-09-25T22:35:44.824Z · LW · GW · 1 comments

Contents

1 comment

We (Goodfire) just put our research preview live - you can play with Llama 3 and use sparse autoencoders to read & write from its internal activations. This is a linkpost for:

• The research preview.
• Our blog post about building it.

Taking research and turning it into something you can actually use and play with has been great. It's surprising how much of a difference iterating on something when you expect it to actually be used feels; I think it's definitely pushed the quality of what you can do with SAEs up a notch.

1 comments

Comments sorted by top scores.