Building and vetting clinical decision rules.

codes and plots for "Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs"

Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling (sklearn-compatible).

This is the official code repository for the paper NeurIPS 2024 spotlight paper "Kermut: Composite kernel regression for protein variant effects".

Super-fast BART (Bayesian Additive Regression Trees) in Python