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

    Neuronpedia

    INDEX
    Explanations

    technical terms

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Eurostile
    0.56
     podcasts
    0.54
     bepaalde
    0.54
    uslararası
    0.54
     vissa
    0.54
     nachhalt
    0.54
     bolsas
    0.52
     revista
    0.51
     prolet
    0.51
    ymmetries
    0.51
    POSITIVE LOGITS
    这是一个
    0.60
    进行
    0.58
    simple
    0.55
    इस
    0.54
     implementation
    0.54
     glorified
    0.54
    简单的
    0.51
    using
    0.50
    Simple
    0.50
    基本的
    0.50
    Act Density 0.731%

    No Known Activations