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

    Neuronpedia

    INDEX
    Explanations

    function return descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    1.05
    >
    0.75
    _
    0.70
    ),
    0.64
    ，
    0.64
    ,\
    0.63
    =\{
    0.62
    ispo
    0.62
     finales
    0.61
    ,'
    0.61
    POSITIVE LOGITS
    ia
    0.70
    igence
    0.67
    specific
    0.67
    as
    0.66
    다면
    0.66
    n
    0.64
    다
    0.60
    signal
    0.59
    sense
    0.59
    sust
    0.58
    Act Density 0.073%

    No Known Activations