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

    Neuronpedia

    INDEX
    Explanations

    management, temperament, legs, cosplay, violent, safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ｎ
    0.46
    kében
    0.43
    ｉ
    0.41
    参数
    0.40
    Despatx
    0.40
    媟
    0.40
    шке
    0.40
    腟
    0.40
    ｊ
    0.40
    estomac
    0.39
    POSITIVE LOGITS
    0.49
    ↵↵
    0.43
     dahil
    0.41
     و
    0.41
     juga
    0.40
    ،
    0.40
    ؛
    0.38
     também
    0.38
     tari
    0.38
     (
    0.38
    Act Density 0.543%

    No Known Activations