3 papers
q-bio.NC2026
How Optimality Structures Sparse Dictionaries: A Theory for Understanding SAE Representations
William Dorrell
Sparse Autoencoders (SAEs) have found success parsing neural representations into interpretable concepts, providing a basis for understanding and control. However, what exactly SAE…
q-bio.NC2026
Convex Efficient Coding
William Dorrell, Peter E. Latham, James Whittington
Why do neurons encode information the way they do? Normative answers to this question model neural activity as the solution to an optimisation problem; for example, the celebrated…
q-bio.NC2025
Range, not Independence, Drives Modularity in Biologically Inspired Representations
Will Dorrell, Kyle Hsu, Luke Hollingsworth +6
Why do biological and artificial neurons sometimes modularise, each encoding a single meaningful variable, and sometimes entangle their representation of many variables? In this wo…