collaborators

6 papers

cs.CV2026

Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds

Thomas Fel, Matthew Kowal, Mozes Jacobs +22

What is the geometry of a visual percept? The most widely used protocols for decomposing neural network representations into interpretable parts treat concepts as isolated directio…

cs.CL2025

From Memorization to Reasoning in the Spectrum of Loss Curvature

Jack Merullo, Srihita Vatsavaya, Lucius Bushnaq +1

We characterize how memorization is represented in transformer models and show that it can be disentangled in the weights of both language models (LMs) and vision transformers (ViT…

cs.LG2025

Stochastic Parameter Decomposition

Lucius Bushnaq, Dan Braun, Lee Sharkey

A key step in reverse engineering neural networks is to decompose them into simpler parts that can be studied in relative isolation. Linear parameter decomposition -- a framework t…

cs.LG2025

Identifying Sparsely Active Circuits Through Local Loss Landscape Decomposition

Brianna Chrisman, Lucius Bushnaq, Lee Sharkey

Much of mechanistic interpretability has focused on understanding the activation spaces of large neural networks. However, activation space-based approaches reveal little about the…

cs.LG2025

Interpretability in Parameter Space: Minimizing Mechanistic Description Length with Attribution-based Parameter Decomposition

Dan Braun, Lucius Bushnaq, Stefan Heimersheim +2

Mechanistic interpretability aims to understand the internal mechanisms learned by neural networks. Despite recent progress toward this goal, it remains unclear how best to decompo…

cs.LG2025

Open Problems in Mechanistic Interpretability

Lee Sharkey, Bilal Chughtai, Joshua Batson +26

Mechanistic interpretability aims to understand the computational mechanisms underlying neural networks' capabilities in order to accomplish concrete scientific and engineering goa…