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

    Neuronpedia

    INDEX
    Explanations

    based on

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    重點
    0.81
    downcase
    0.81
    重点
    0.79
    Basic
    0.79
    зии
    0.79
     มั้ย
    0.78
    itez
    0.77
    ቂያ
    0.77
    ﹕
    0.77
    戏剧
    0.76
    POSITIVE LOGITS
     off
    1.99
    off
    1.51
     Off
    1.32
     OFF
    1.28
    Off
    1.25
     solely
    1.20
     offs
    1.12
     purely
    1.11
     premise
    1.09
     around
    1.09
    Act Density 0.077%

    No Known Activations