Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    code and programming

    np_max-act · gemini-2.0-flash

    The neuron activates on parts of “How do I…?”–style questions, especially the “How,” “do,” “I,” and the final “?” that mark a how-to query.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     moderators
    -0.06
     coefficients
    -0.06
     equipments
    -0.06
    areas
    -0.06
    (string
    -0.06
    ()][
    -0.06
    _membership
    -0.06
     cinema
    -0.06
     communities
    -0.06
    perfil
    -0.06
    POSITIVE LOGITS
    pub
    0.07
    few
    0.07
     ليس
    0.07
    '));
↵
    0.07
     filmy
    0.06
    ","",
    0.06
    ản
    0.06
     -->↵↵↵
    0.06
     adına
    0.06
     Hà
    0.06
    Act Density 0.076%

    No Known Activations