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

    Neuronpedia

    INDEX
    Explanations

    sentencing disparities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ot
    2.69
    ak
    2.39
    ar
    1.97
    ात
    1.92
    ros
    1.88
    uk
    1.87
    akn
    1.81
    icht
    1.80
    arán
    1.69
    na
    1.67
    POSITIVE LOGITS
    то
    2.03
    tion
    1.88
    я
    1.83
    бна
    1.74
    ль
    1.70
    ция
    1.68
    یت
    1.64
    특별시
    1.62
    ры
    1.59
     הרב
    1.59
    Act Density 0.031%

    No Known Activations