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

    Neuronpedia

    INDEX
    Explanations

    **language identification or processing**

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    d
    1.18
    m
    1.18
    b
    1.17
    s
    1.12
    g
    1.11
    k
    1.02
    h
    1.02
    i
    0.99
    t
    0.98
    e
    0.98
    POSITIVE LOGITS
     República
    0.86
     মোতায়
    0.82
     você
    0.82
     bạn
    0.81
     คุณ
    0.79
    ạt
    0.79
    клады
    0.78
    າດ
    0.77
    𝘳
    0.77
    THING
    0.75
    Act Density 0.001%

    No Known Activations