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

    Neuronpedia

    INDEX
    Explanations

    versions or editions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     to
    0.90
    ி
    0.78
    ların
    0.72
     u
    0.67
    o
    0.65
     h
    0.64
     sano
    0.64
    i
    0.64
     rodeo
    0.64
    ी
    0.64
    POSITIVE LOGITS
    have
    0.79
    తో
    0.75
    en
    0.71
     versions
    0.70
    óta
    0.67
    ق
    0.67
    شد
    0.66
    ى
    0.66
    зм
    0.66
    ang
    0.65
    Act Density 0.568%

    No Known Activations