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

    Neuronpedia

    INDEX
    Explanations

    code and exclamation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     않았
    0.46
    "
    0.44
    ก
    0.40
    ः
    0.38
    ẖ
    0.38
    "。
    0.37
     bagaimana
    0.37
    ۸
    0.37
     ہوئی۔
    0.36
    dokument
    0.36
    POSITIVE LOGITS
    se
    0.56
    pe
    0.44
    ا
    0.41
    re
    0.39
    ya
    0.38
    pre
    0.38
    pt
    0.38
    sh
    0.36
    ye
    0.36
    ge
    0.36
    Act Density 1.357%

    No Known Activations