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

    Neuronpedia

    INDEX
    Explanations

    ownership, nationality, concepts, groups

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    з
    2.22
    ী
    2.14
    сний
    1.88
    ร์
    1.75
    мо
    1.74
    ў
    1.66
    いた
    1.66
    ل
    1.66
     kinks
    1.65
    و
    1.58
    POSITIVE LOGITS
    ي
    2.45
    de
    2.38
    den
    2.33
    gi
    2.31
    da
    2.28
    des
    2.20
    data
    2.11
    gel
    2.05
    y
    2.05
    gl
    2.02
    Act Density 0.024%

    No Known Activations