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

    Neuronpedia

    INDEX
    Explanations

    consider

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Windsor
    -0.10
    ző
    -0.08
     трудно
    -0.08
     몇
    -0.08
    습니다
    -0.08
     máte
    -0.08
     Rochester
    -0.08
     quelques
    -0.08
     imamo
    -0.08
     blazer
    -0.08
    POSITIVE LOGITS
     numbering
    0.09
    のみ
    0.08
     errone
    0.08
     considers
    0.08
     treats
    0.08
     carbonation
    0.08
     treat
    0.08
    认为
    0.08
    .case
    0.08
     CCR
    0.08
    Act Density 0.020%

    No Known Activations