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

    Neuronpedia

    INDEX
    Explanations

    html/xml tags and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ン
    0.98
    a
    0.88
    ة
    0.83
    oted
    0.81
    aing
    0.80
    ffler
    0.80
    arians
    0.79
     था
    0.79
    ılarak
    0.79
    oire
    0.77
    POSITIVE LOGITS
    á
    1.42
    é
    1.30
    ></
    1.23
     in
    1.16
    and
    1.09
    ре
    1.09
    u
    1.08
    ية
    1.03
    are
    1.02
    ки
    1.02
    Act Density 0.001%

    No Known Activations