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

    Neuronpedia

    INDEX
    Explanations

    exclude files and folders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    1.30
    r
    0.91
    ד
    0.87
    ↵↵
    0.86
     A
    0.85
     (
    0.78
    <h2>
    0.73
    dan
    0.73
    ่
    0.70
    l
    0.69
    POSITIVE LOGITS
    ри
    1.19
    of
    1.17
    ofthe
    0.92
     peu
    0.84
    are
    0.84
     desolate
    0.82
    ol
    0.80
    ли
    0.80
    ត្រូវការ
    0.80
     rapporto
    0.80
    Act Density 0.019%

    No Known Activations