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

    Neuronpedia

    INDEX
    Explanations

    granularity, granular, granule

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    كان
    1.58
     for
    1.41
    Ре
    1.40
    Ми
    1.31
    Дан
    1.31
    洴
    1.31
    Лю
    1.30
    Тех
    1.30
    Кор
    1.30
    Ти
    1.29
    POSITIVE LOGITS
    с
    1.80
    اری
    1.36
    ین
    1.27
    อง
    1.26
    ס
    1.26
    า
    1.19
    il
    1.14
    ่
    1.12
    ня
    1.03
    ρα
    1.03
    Act Density 0.000%

    No Known Activations