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

    Neuronpedia

    INDEX
    Explanations

    storg, org, fore, orb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    g
    2.16
    c
    1.59
    ق
    1.44
    is
    1.42
    ز
    1.38
    h
    1.37
    k
    1.32
    ع
    1.29
    ج
    1.23
    ض
    1.22
    POSITIVE LOGITS
     as
    1.21
    <0x0D>
    1.13
    as
    1.07
    ка
    1.01
    ের
    0.96
    な
    0.89
     are
    0.89
    DD
    0.88
    YL
    0.87
     sustained
    0.86
    Act Density 0.000%

    No Known Activations