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

    Neuronpedia

    INDEX
    Explanations

    dimensions (width, height, px)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     that
    -1.44
     to
    -1.27
    :
    -1.10
    ,
    -1.10
     such
    -1.02
    НТ
    -0.98
    의
    -0.97
    zonych
    -0.96
    消息
    -0.94
     Το
    -0.94
    POSITIVE LOGITS
    前から
    1.20
    銹
    1.15
     χρήση
    1.10
    Zdrav
    1.08
    メニューは
    1.07
    時間は
    1.06
    אין
    1.03
    Второй
    1.02
    Karakter
    1.02
    roblast
    1.02
    Act Density 0.005%

    No Known Activations