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

    Neuronpedia

    INDEX
    Explanations

    biological differences and instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    un
    0.62
    de
    0.57
    uk
    0.53
    name
    0.52
    config
    0.52
    student
    0.52
    cannot
    0.52
    teacher
    0.52
    start
    0.51
    ad
    0.51
    POSITIVE LOGITS
    こそ
    0.43
     fluorescent
    0.43
    رحله
    0.42
    сную
    0.42
    ೂರ್ಯ
    0.41
     abraz
    0.41
     ngại
    0.40
     harsh
    0.40
     richt
    0.40
    상을
    0.39
    Act Density 0.002%

    No Known Activations