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

    Neuronpedia

    INDEX
    Explanations

    dehumanize, demonize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    も
    0.88
    도
    0.83
    在
    0.81
    其他
    0.78
    もら
    0.73
     импера
    0.70
    残
    0.70
    镜
    0.69
    Т
    0.68
    使
    0.68
    POSITIVE LOGITS
    ม
    0.95
    ad
    0.82
    ू
    0.78
    n
    0.77
    il
    0.77
    w
    0.75
    ط
    0.75
     at
    0.74
    č
    0.74
    ap
    0.73
    Act Density 0.020%

    No Known Activations