the open
    research
    Contribute
    Back to the reading lists

    Safety, alignment & interpretability· 2024

    Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet

    Adly Templeton (first author) · 2024

    Read the original

    arXiv 2605.29358OpenAlex W7162776923

    On the reading lists

    Why it matters

    Scaled interpretable feature extraction to a production language model and demonstrated causal feature steering.

    Condensed from the expert lists' notes.

    The record

    First released
    2024 (from oracle lists)
    Citations
    898 citations, as counted by Semantic Scholar logoSemantic Scholar (non-commercial licence)
    Named by
    both expert lists
    Score
    0.434 · computed rank 315 · consensus 0.46, must-read 0.5, citations percentile 0.702 (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), and Semantic Scholar logoSemantic Scholar (used under its non-commercial API licence). Each count names its source. See the method.