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

    Neuronpedia

    INDEX
    Explanations

    preferred route

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Level
    0.78
     Beware
    0.68
     level
    0.66
     gums
    0.65
    či
    0.64
    (\
    0.63
    elbe
    0.63
    er
    0.63
     consequence
    0.63
     Depart
    0.62
    POSITIVE LOGITS
    アメリカ
    0.75
    USART
    0.74
    Conver
    0.70
    都
    0.70
     কুম
    0.69
     japonais
    0.68
    ௌ
    0.68
     découvrir
    0.67
    IT
    0.67
    ມັນ
    0.66
    Act Density 0.001%

    No Known Activations