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

    Neuronpedia

    INDEX
    Explanations

    multilingual punctuation and symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    \
    0.68
    리
    0.58
    কে
    0.58
    ری
    0.57
    2
    0.57
    ست
    0.57
    _
    0.57
    로
    0.55
     arterioles
    0.55
    פ
    0.54
    POSITIVE LOGITS
    in
    0.67
    吗
    0.63
    ों
    0.57
    ים
    0.57
     แม
    0.57
    ও
    0.56
     gebruiken
    0.55
    и
    0.53
    ют
    0.52
     geeft
    0.52
    Act Density 0.022%

    No Known Activations