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

    Neuronpedia

    INDEX
    Explanations

    security risks and unethical actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    구매
    0.55
     businesswoman
    0.51
    ელი
    0.47
    Puede
    0.46
     อายุ
    0.46
    商人
    0.44
    pointB
    0.44
    Jeśli
    0.43
    Можно
    0.43
    很多
    0.42
    POSITIVE LOGITS
     forever
    0.48
    ir
    0.46
     devices
    0.44
     Fund
    0.44
     تسه
    0.43
    ights
    0.42
     Video
    0.42
    ides
    0.42
     un
    0.41
     stream
    0.41
    Act Density 0.001%

    No Known Activations