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

    Neuronpedia

    INDEX
    Explanations

    comma followed by conjunction or adverb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    1.36
    的是
    0.93
    :
    0.90
    ،
    0.88
    (?)
    0.87
    ,--
    0.87
    .--
    0.85
    /
    0.84
    ?
    0.83
    Due
    0.81
    POSITIVE LOGITS
    并且
    1.80
    並且
    1.72
     тобто
    1.67
     takže
    1.65
    而且
    1.64
    而
    1.63
     чтобы
    1.62
     což
    1.61
    そして
    1.59
     그리고
    1.58
    Act Density 0.063%

    No Known Activations