Alice Rigg
Bio
Alice Rigg is a mechanistic interpretability researcher based in Ottawa, Canada. She is a Machine Learning Researcher at EleutherAI and previously conducted independent research in mechanistic interpretability from 2023 to 2024. She participated in the MATS program in Summer 2023 and served as project lead for the "Towards Ambitious Mechanistic Interpretability" initiative at AI Safety Camp 2024, where her team focused on improving the quality-versus-realism tradeoff in mechanistic explanations and developing better evaluation metrics. She is a co-author of the paper "Bilinear MLPs enable weight-based mechanistic interpretability" (arXiv 2410.08417), co-authored with Michael Pearce, Thomas Dooms, Jose Oramas, and Lee Sharkey. She moderates a mechanistic interpretability Discord community and runs weekly reading groups. Her background is in mathematics, and she approaches AI safety through a technical, interpretability-focused lens.
Links
- Personal Website
- https://woog97.github.io/
- Twitter / X
- LessWrong
- -
Grants
No grants recorded.
Discussion
Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.
Details
- Last Updated
- Mar 22, 2026, 2:06 PM UTC
- Created
- Mar 20, 2026, 3:00 AM UTC