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

    Neuronpedia

    INDEX
    Explanations

    URL click, children, Japanese, Python

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    да
    0.91
    ان
    0.86
    در
    0.77
    ير
    0.74
    ل
    0.71
    ق
    0.71
    மன்
    0.70
    دو
    0.68
     izgrad
    0.68
    ارش
    0.67
    POSITIVE LOGITS
     Gains
    0.79
     Tomb
    0.76
     Drain
    0.75
    ленная
    0.74
     Нужно
    0.73
     kết
    0.73
     только
    0.73
     Mỹ
    0.72
     BCH
    0.72
     Catawiki
    0.72
    Act Density 0.002%

    No Known Activations