the open
    research
    Contribute
    Back to the reading lists

    Safety, alignment & interpretability· 2024

    Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

    Evan Hubinger (first author) · 2024

    Read the original

    arXiv 2401.05566OpenAlex W4390833079

    On the reading lists

    Why it matters

    Demonstrated experimentally that deceptive backdoor behaviors can survive conventional safety training.

    Condensed from the expert lists' notes.

    The record

    First released
    2024 (from arXiv 2401.05566)
    Citations
    625 citations, as counted by Papers With Code (CC BY-SA 4.0)
    Named by
    both expert lists
    Score
    0.350 · computed rank 481 · consensus 0.505, must-read 0.5, citations percentile 0.426 (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.