Comment by Taywon Min (taywon-min) on An X-Ray is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation · 2025-02-24T11:30:44.077Z · LW · GW

Thanks for the great work. I think that multimodal sparse auto encoders is a promising direction. Do you think it is possible / worthwhile to train SAEs on vla models like OpenVLA? I haven't seen any related work training or interpreting action models using SAE work, and am curious of your thoughts.