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

    Neuronpedia

    INDEX
    Explanations

    double or triple in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    其他
    0.59
    n
    0.59
    ры
    0.53
    整个
    0.51
    k
    0.51
    j
    0.51
    ক
    0.51
    w
    0.50
    en
    0.50
    d
    0.48
    POSITIVE LOGITS
    ედერ
    0.49
     harum
    0.48
     Webinar
    0.48
     plural
    0.47
     Navigator
    0.47
     botão
    0.46
     Bunun
    0.46
     Neurology
    0.46
     Yoh
    0.46
     quotid
    0.46
    Act Density 0.000%

    No Known Activations