SOURCE-LINKED INTELLIGENCE

A Dominant Diffuse Phase in the Sparse Autoencoder Phase Diagram

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Sparse autoencoders (SAEs) are increasingly used to recover interpretable features from neural-network activations, yet systematic feature co-occurrence can cause distinct features to be absorbed or merged. The MAIS-O43 open problem proposes a controlled experiment to characterize when recovery of a true synthetic dictionary gives way to feature merging as the nesting fraction $γ$, sparsity penalty $λ$, and dictionary size $M$ vary. We implement the specified protocol and evaluate 200 independently initialized fits across ten of the 165 grid cells. We observe zero full-dictionary recoveries an

Read original source ↗ Open in workspace

recordType

paper

region

Global