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

    Neuronpedia

    INDEX
    Explanations

    Frances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    та
    2.03
    ak
    2.02
    ip
    1.75
    да
    1.75
    ణి
    1.71
    verts
    1.70
    ли
    1.66
    ные
    1.66
    ى
    1.66
     turnt
    1.63
    POSITIVE LOGITS
     دوسرے
    2.20
    ።
    2.16
    Durante
    2.05
    ل
    2.00
    doesn
    1.97
    cript
    1.96
    don
    1.91
    ierto
    1.88
    Дру
    1.87
    ség
    1.86
    Act Density 0.002%

    No Known Activations