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

    Neuronpedia

    INDEX
    Explanations

    roll

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     will
    -0.07
     would
    -0.07
    وبة
    -0.07
     quiere
    -0.07
     can
    -0.06
     is
    -0.06
     не
    -0.06
     didn
    -0.06
     прит
    -0.06
     získ
    -0.06
    POSITIVE LOGITS
    attered
    0.07
    _Rect
    0.06
    \Model
    0.06
    ्श
    0.06
     tém
    0.06
    (cd
    0.06
     Savaşı
    0.06
    abol
    0.06
    amilia
    0.06
     Tus
    0.06
    Act Density 0.142%

    No Known Activations