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

    Neuronpedia

    INDEX
    Explanations

    extended periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    1
    1.03
    ก
    0.79
    你
    0.79
    ков
    0.77
    的
    0.77
    но
    0.75
    ل
    0.73
    РИ
    0.71
    ig
    0.71
    ів
    0.70
    POSITIVE LOGITS
    s
    0.79
    ?’
    0.79
     بندی
    0.73
     آباد
    0.73
    uksen
    0.71
    extended
    0.70
     هستند
    0.69
     آیا
    0.69
     extended
    0.68
    0.67
    Act Density 0.002%

    No Known Activations