Patrick Leask

PhD Candidate

Durham University Computer Science

He/him

Within this scope...

subprojects

24 Apr 2025
Latent mechanistic interpretability

timeline

Paper & Poster
24 Apr 2025
Sparse Autoencoders Do Not Find Canonical Units of Analysis
Conference poster
15 Dec 2024
BatchTopK Sparse Autoencoders
Conference poster
15 Dec 2024
Stitching Sparse Autoencoders of Different Sizes
Workshop
6 May 2024
AI Forensics IRL Meeting
Preprint
1 Nov 2023
CoinRun: Solving Goal Misgeneralisation