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

    Neuronpedia

    INDEX
    Explanations

    estable / establishment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    1.20
    ri
    1.08
    t
    1.07
    are
    1.01
    ات
    1.01
    es
    0.98
    us
    0.97
    ت
    0.96
    ture
    0.96
    ა
    0.96
    POSITIVE LOGITS
     I
    0.97
    中の
    0.93
    กับ
    0.88
     ﺍ
    0.88
    =",
    0.86
     ถึง
    0.86
     extraordinaire
    0.84
     Vì
    0.83
    と思いますが
    0.83
     çıkar
    0.82
    Act Density 0.002%

    No Known Activations