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

    Neuronpedia

    INDEX
    Explanations

    web design

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    1.70
    of
    1.34
    k
    1.30
    to
    1.29
    s
    1.23
    c
    1.22
    by
    1.16
    ל
    1.14
    ل
    1.13
    h
    1.12
    POSITIVE LOGITS
    ٥
    1.09
    昀
    1.00
    াধিকার
    0.95
    ilerin
    0.92
    есть
    0.92
     těch
    0.89
    宒
    0.89
     ”
    0.88
    过程中
    0.88
     profiter
    0.88
    Act Density 0.002%

    No Known Activations