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

    Neuronpedia

    INDEX
    Explanations

    code comments and paths

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    с
    0.91
    m
    0.90
    end
    0.87
    ol
    0.85
    룀
    0.85
    //
    0.85
    DOUT
    0.84
    fek
    0.84
    od
    0.83
    //!
    0.82
    POSITIVE LOGITS
     znači
    1.19
    ました
    1.13
    ский
    1.05
    ına
    1.05
     ກ
    1.03
     へ
    0.98
    ጴ
    0.96
     ก
    0.96
     為
    0.95
    事人
    0.95
    Act Density 0.000%

    No Known Activations