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

    Neuronpedia

    INDEX
    Explanations

    Time-mixing and Channel-mixing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    را
    0.52
    s
    0.47
    غ
    0.46
    ס
    0.45
    斯
    0.45
     comunque
    0.44
    of
    0.42
    ים
    0.42
    ਾ
    0.42
     را
    0.41
    POSITIVE LOGITS
     SimSun
    0.56
    amiseks
    0.51
     familiarize
    0.49
    是中国
    0.48
    മല്ല
    0.48
     China
    0.48
     ImportError
    0.48
    дентификаторы
    0.48
    ⺀
    0.47
     encrypt
    0.47
    Act Density 0.000%

    No Known Activations