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

    Neuronpedia

    INDEX
    Explanations

    Muhammad and cuda

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    持ち
    1.26
    며
    1.25
    اک
    1.23
    いますが
    1.19
    つ
    1.14
    ק
    1.13
    larda
    1.12
    ləşdir
    1.11
    行う
    1.09
    اً
    1.08
    POSITIVE LOGITS
    ه
    1.39
     getAll
    1.16
     on
    1.11
     or
    1.11
    ు
    1.09
     was
    1.06
     and
    1.05
     is
    1.02
    a
    1.02
    ς
    1.02
    Act Density 0.018%

    No Known Activations