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

    Neuronpedia

    INDEX
    Explanations

    sum calculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    т
    1.79
    ब्दिक
    1.74
    is
    1.70
    的大
    1.67
    鹖
    1.61
     Deshalb
    1.59
     glimpse
    1.58
    ig
    1.57
     soorten
    1.57
     سبتمبر
    1.56
    POSITIVE LOGITS
    ми
    2.05
    dır
    1.97
    てもら
    1.89
    د
    1.82
    ний
    1.80
     изменений
    1.73
    Ｋ
    1.73
     Sü
    1.71
    то
    1.70
    ции
    1.69
    Act Density 0.042%

    No Known Activations