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

    Neuronpedia

    INDEX
    Explanations

    destroy chatbots

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     %).
    0.66
    <bos>
    0.58
     für
    0.57
     années
    0.57
    สำหรับ
    0.57
    EAR
    0.56
    {})
    0.56
     và
    0.55
     در
    0.55
     october
    0.55
    POSITIVE LOGITS
    ت
    0.68
    то
    0.65
     Plane
    0.65
     Wind
    0.60
    ת
    0.59
    ิน
    0.57
     Fly
    0.56
    ে
    0.56
     Animal
    0.55
     Investing
    0.55
    Act Density 0.000%

    No Known Activations