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

    Neuronpedia

    INDEX
    Explanations

    even in contrast or exception

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    k
    0.69
    é
    0.68
    f
    0.67
    ু
    0.67
    ون
    0.66
    kannya
    0.66
    いた
    0.64
    та
    0.62
    تی
    0.61
    го
    0.61
    POSITIVE LOGITS
    2
    0.71
    ized
    0.61
    .
    0.58
     когато
    0.55
    6
    0.54
     представить
    0.53
    7
    0.53
     चाहिए
    0.52
    мм
    0.52
    imo
    0.52
    Act Density 0.093%

    No Known Activations