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

    Neuronpedia

    INDEX
    Explanations

    complications

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     are
    0.93
    นี้
    0.93
     as
    0.89
    ği
    0.87
    ید
    0.84
    ные
    0.84
     gì
    0.83
    로
    0.82
     zijn
    0.80
     gén
    0.77
    POSITIVE LOGITS
    m
    1.62
    u
    1.52
     complications
    1.28
    n
    1.20
    t
    1.18
    l
    1.15
    r
    1.12
    an
    1.11
    ar
    1.11
    k
    1.09
    Act Density 0.002%

    No Known Activations