HiPACE: Hierarchical Phase-Boundary Analysis and Controlled Evaluation of Feature Absorption in Sparse Autoencoders

T1Covered by 1 source · first reported by arXiv at 25 Sept, 04:00 UTC

arXiv:2609.29551v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) decompose LLM activations into sparse dictionary atoms, so that each distinct concept gets its own feature. One recurring behavior complicates this premise: feature absorption, in which a parent concept and its…

aiscience

Sources

• T1HiPACE: Hierarchical Phase-Boundary Analysis and Controlled Evaluation of Feature Absorption in Sparse AutoencodersarXiv25 Sept, 04:00