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

    Neuronpedia

    INDEX
    Explanations

    social construct, in pieces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    atter
    0.42
     COUNCIL
    0.41
     kV
    0.40
    如下图
    0.40
     باتیں
    0.39
    Tovar
    0.39
    LineWidth
    0.38
    ufthansa
    0.38
     বলেন
    0.38
     المجلس
    0.38
    POSITIVE LOGITS
    ன்
    0.52
    ни
    0.49
     Règles
    0.48
    m
    0.46
    l
    0.46
    ベ
    0.46
     инструк
    0.45
     ಇ
    0.45
    ması
    0.45
    રસ
    0.45
    Act Density 0.025%

    No Known Activations