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

    Neuronpedia

    INDEX
    Explanations

    Brussels sprouts and Tiananmen Square

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.65
    ama
    0.63
    ";
    0.62
    ing
    0.60
    ".
    0.57
    <0xA8>
    0.56
    ag
    0.56
    </
    0.55
    achi
    0.55
    ",
    0.55
    POSITIVE LOGITS
    드
    0.81
    Ссылки
    0.75
    Са
    0.74
     Smoking
    0.71
    Се
    0.68
    Де
    0.68
    Ре
    0.67
     Sails
    0.67
    З
    0.67
     эффект
    0.66
    Act Density 0.001%

    No Known Activations