July 13, 2025

Call for Papers Open! Mechanistic Interpretability Workshop at NeurIPS 2025

Our call for papers is open! We invite submissions of short (max 4 pages) and long (max 9 pages) papers outlining new research, due August 22, 2025. We welcome all submissions that convincingly argue for why they further the field: i.e. which advance our ability to use the internal states of neural networks to understand them.

We are also extremely grateful to any who volunteer as reviewers, you can express interest here

Check out our website for more info: https://mechinterpworkshop.com

And please pass on the word to anyone else you know who might be interested!

Thanks,

Neel, on behalf of the organising committee