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

    Neuronpedia

    INDEX
    Explanations

    flaky

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ことで
    1.09
    ння
    1.09
    ع
    0.95
    取り
    0.91
    of
    0.89
    ,
    0.89
    ق
    0.88
     כ
    0.86
    示す
    0.86
    ंनी
    0.86
    POSITIVE LOGITS
    in
    1.51
    et
    1.42
    ak
    1.39
    riss
    1.26
    ل
    1.25
    inni
    1.24
    r
    1.21
    inę
    1.18
    nál
    1.16
    ou
    1.15
    Act Density 0.000%

    No Known Activations