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

    Neuronpedia

    INDEX
    Explanations

    Mike followed by last names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ่า
    1.49
    is
    1.48
    s
    1.44
    િ
    1.39
    are
    1.34
    ים
    1.30
    land
    1.30
    as
    1.29
    िक
    1.26
    ні
    1.26
    POSITIVE LOGITS
    我
    1.16
    ,
    1.12
    ক
    1.11
    我们
    1.09
     νέα
    1.08
    你
    1.06
    ிறது
    1.05
     éventuellement
    1.04
    அ
    1.04
    该
    1.00
    Act Density 0.003%

    No Known Activations