the open
    research
    Contribute
    Back to the reading lists

    Safety, alignment & interpretability· 2023

    Sparse Autoencoders Find Highly Interpretable Features in Language Models

    Robert Huben (first author) · 2023

    Read the original

    arXiv 2309.08600OpenAlex W4386839891

    On the reading lists

    Why it matters

    Established sparse autoencoders as a practical method for discovering interpretable language-model features.

    Condensed from the expert lists' notes.

    The record

    First released
    2023 (from arXiv 2309.08600)
    Citations
    1,697 citations, as counted by Papers With Code (CC BY-SA 4.0)
    Named by
    both expert lists
    Score
    0.415 · computed rank 343 · consensus 0.54, must-read 0, citations percentile 0.723 (2020s · safety), awards 0, existing lists 0. Method

    Data sources

    Catalogue data from OpenAlex (CC0), Crossref (CC0 metadata; no abstracts are shown), arXiv, and Papers With Code (CC BY-SA 4.0). Each count names its source. See the method.