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

    Neuronpedia

    INDEX
    Explanations

    protected characteristics and discrimination

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ERROR
    0.39
    姐妹
    0.37
     DWORD
    0.37
    房間
    0.37
    教程
    0.36
    燡
    0.36
     quarts
    0.35
    suptitle
    0.35
     eddies
    0.34
    磬
    0.34
    POSITIVE LOGITS
     religion
    0.54
     sexism
    0.53
     nationality
    0.50
    Religion
    0.49
     Religion
    0.47
    Gender
    0.47
    race
    0.46
    性别
    0.46
    Race
    0.46
    gender
    0.45
    Act Density 0.015%

    No Known Activations