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

    Neuronpedia

    INDEX
    Explanations

    at some

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Объ
    2.33
     Dzięki
    2.27
     Несмотря
    2.20
     Před
    2.16
     Maeda
    2.12
     Перед
    2.07
    Перед
    2.05
     Такой
    2.03
     Органи
    2.03
     Помимо
    2.03
    POSITIVE LOGITS
    }^{*}
    2.49
    is
    2.32
    }^{
    2.16
    an
    2.04
    }^{\
    1.98
    $(".
    1.94
    un
    1.90
    ges
    1.90
    }-
    1.90
    }^{-}
    1.85
    Act Density 0.016%

    No Known Activations