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

    Neuronpedia

    INDEX
    Explanations

    division by zero error

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ल
    0.62
    จะต้อง
    0.62
    ラの
    0.59
    واي
    0.58
    رین
    0.58
    amız
    0.57
    ла
    0.56
    ور
    0.56
    ל
    0.55
    ير
    0.55
    POSITIVE LOGITS
    ode
    0.50
    iff
    0.49
    figure
    0.49
    dru
    0.49
     السيارة
    0.49
    ette
    0.48
    struct
    0.47
    Society
    0.47
     prib
    0.46
    app
    0.46
    Act Density 0.000%

    No Known Activations