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

    Neuronpedia

    INDEX
    Explanations

    hack, github, hub

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ции
    1.30
    ರ
    1.25
     donné
    1.23
    тная
    1.19
    ר
    1.19
    δά
    1.17
    ры
    1.16
     оригі
    1.16
    р
    1.13
     Лі
    1.12
    POSITIVE LOGITS
    n
    1.31
    m
    1.30
    t
    1.28
    s
    1.27
     iż
    1.27
    en
    1.22
    ")->
    1.19
    ers
    1.16
    the
    1.16
    stories
    1.12
    Act Density 0.137%

    No Known Activations