David Chanin
I will contribute to the open-source SAELens library, and support Decode Research with their other projects.
SaferAI
Support for SaferAI’s technical and governance research and education programs to enable responsible and safe AI.
Tom McGrath
Find the best settings for SAE training we can, then scale across models
Glen M. Taggart
By rapid iteration on possible alternative architectures & training techniques
Johnny Lin
Lovis Heindrich