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

    Neuronpedia

    INDEX
    Explanations

    app message notifications Toast SnackBar

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    0.85
    In
    0.82
    .
    0.80
    us
    0.79
    The
    0.75
    hus
    0.73
     in
    0.73
    och
    0.71
    cular
    0.70
    ł
    0.69
    POSITIVE LOGITS
     be
    1.04
    يف
    1.00
    מי
    0.92
    يت
    0.92
    يد
    0.91
    サ
    0.90
    ин
    0.88
    スト
    0.87
    ブ
    0.86
    নে
    0.86
    Act Density 0.004%

    No Known Activations