← Search

Bart Bussmann

2 accepted papers

2025

Learning Multi-Level Features with Matryoshka Sparse Autoencoders

ICML 2025poster

Sparse autoencoders (SAEs) have emerged as a powerful tool for interpreting neural networks by extracting the concepts represented in their activations. However, choosing the size of the SAE dictionary (i.e. number of learned concepts) creates a tension: as dictionary size increases to capture more…

2025

Sparse Autoencoders Do Not Find Canonical Units of Analysis

ICLR 2025poster

A common goal of mechanistic interpretability is to decompose the activations of neural networks into features: interpretable properties of the input computed by the model. Sparse autoencoders (SAEs) are a popular method for finding these features in LLMs, and it has been postulated that they can be…

Cited by 1SourcePDFScholar