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

    Neuronpedia

    INDEX
    Explanations

    the most

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    m
    0.66
    henyl
    0.58
    '.$
    0.57
    ים
    0.56
    müş
    0.56
    u
    0.55
    什么是
    0.55
     regard
    0.54
    <0x80>
    0.54
    bmp
    0.54
    POSITIVE LOGITS
     starters
    0.66
    ه
    0.65
    НУ
    0.65
     сроки
    0.63
     posterity
    0.62
    GaussianBlur
    0.62
    期間
    0.61
    erun
    0.61
     sake
    0.61
     purposes
    0.60
    Act Density 0.029%

    No Known Activations