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

    Neuronpedia

    INDEX
    Explanations

    Express middleware setup

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    na
    0.85
    r
    0.82
    ।
    0.78
    to
    0.76
    3
    0.71
    ta
    0.71
    re
    0.69
    。
    0.68
     بواسطة
    0.66
    ny
    0.65
    POSITIVE LOGITS
    ع
    0.84
    त्तों
    0.72
     use
    0.70
    ệt
    0.68
    ق
    0.67
    হ্ন
    0.66
     частина
    0.66
    স
    0.65
    ব
    0.65
    ট
    0.64
    Act Density 0.009%

    No Known Activations