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

    Neuronpedia

    INDEX
    Explanations

    reinforce harmful societal biases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     voluptate
    0.75
     consom
    0.72
    耗
    0.69
    aturamento
    0.69
     Tark
    0.68
     ইনিংস
    0.68
    ভিশ
    0.67
    Timelapse
    0.67
     hadronic
    0.67
    ផ្គ
    0.67
    POSITIVE LOGITS
     discrimination
    3.24
     discriminatory
    2.98
     Discrimination
    2.86
     prejudice
    2.81
     bias
    2.64
     discriminate
    2.63
     discrimin
    2.59
     discrim
    2.58
     racism
    2.50
     discriminated
    2.49
    Act Density 0.609%

    No Known Activations