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

    Neuronpedia

    INDEX
    Explanations

    over

    np_max-act · gemini-2.0-flash

    phrases indicating someone assuming control or a role/position, especially “take over as” style leadership or responsibility transitions.

    oai_token-act-pair · gpt-5Triggered by @yooniel31
    New Auto-Interp
    Negative Logits
     bulls
    -0.07
    ेल
    -0.06
     enables
    -0.06
     вул
    -0.06
    Add
    -0.06
     Stars
    -0.06
     Cry
    -0.06
     tenga
    -0.05
    cth
    -0.05
    inz
    -0.05
    POSITIVE LOGITS
     overse
    0.07
    ادا
    0.07
    _round
    0.07
     autom
    0.07
     gradu
    0.07
     Autom
    0.07
    _read
    0.07
    /from
    0.06
    errar
    0.06
    792
    0.06
    Act Density 0.008%

    No Known Activations