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

    Neuronpedia

    INDEX
    Explanations

    circle geometry

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Qxd
    2.66
    ियल
    2.63
     ljud
    2.56
     siehe
    2.53
    ส์
    2.52
    ,]
    2.51
    ijiet
    2.42
    Jogo
    2.42
    вига
    2.42
    त
    2.41
    POSITIVE LOGITS
    ت
    3.41
     setan
    3.25
    ר
    3.06
    л
    2.98
     fatto
    2.93
    蚓
    2.88
     त्याचे
    2.82
    ကျင်
    2.80
     нулю
    2.77
     étrangers
    2.76
    Act Density 0.018%

    No Known Activations