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

    Neuronpedia

    INDEX
    Explanations

    code snippets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ع
    1.09
     bilangan
    0.93
    рия
    0.91
    ح
    0.91
    리
    0.87
     étudier
    0.84
    ující
    0.81
     également
    0.80
     voire
    0.80
    进而
    0.80
    POSITIVE LOGITS
    дентификаторы
    0.98
    ль
    0.98
    あります
    0.96
    леты
    0.93
     trusty
    0.91
     Snoop
    0.88
    льники
    0.87
    те
    0.85
     pesky
    0.84
    ੇ
    0.83
    Act Density 0.104%

    No Known Activations