SOURCE-LINKED INTELLIGENCE

Interpreting Latent Protein Language Model Features with Geometric Annotations

arXiv · AI, language, vision and robotics · article · Aug 26, 2026 · UTC

Protein language models (pLMs) encode information about protein sequences which enable downstream tasks such as structure prediction, but their internal representations are not well understood. Sparse autoencoders (SAEs) provide a promising tool to disentangle latent pLM representations into interpretable features, but existing annotation pipelines largely rely on protein-level annotations derived from database labels and LLM annotations of top activating sequences. Such annotations can overlook the localized residue-level and geometric patterns encoded by sparse features. We introduce an auto

Read original source ↗ Open in workspace

recordType

paper

region

Global