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

    Neuronpedia

    INDEX
    Explanations

    browsers and their functions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ي
    0.90
    to
    0.89
    ে
    0.86
    و
    0.85
    ك
    0.79
    i
    0.79
    ر
    0.77
    ו
    0.76
    де
    0.74
    The
    0.68
    POSITIVE LOGITS
     as
    0.91
     can
    0.86
     which
    0.79
     are
    0.75
    ۹
    0.74
     they
    0.73
    ری
    0.73
    ERS
    0.71
     می
    0.69
     of
    0.65
    Act Density 0.001%

    No Known Activations