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

    Neuronpedia

    INDEX
    Explanations

    references to discrimination based on various criteria such as race, gender, sexual orientation, and physical characteristics

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to discrimination and bias based on various personal characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Reviewer
    -0.76
     Purg
    -0.66
    metal
    -0.65
    jet
    -0.64
    bage
    -0.63
     Sunder
    -0.62
    ////////////////////////////////
    -0.61
    bj
    -0.61
    uckland
    -0.61
    invoke
    -0.61
    POSITIVE LOGITS
     ethnicity
    1.12
     nationality
    1.11
     gender
    1.02
     geography
    0.92
     colour
    0.91
     severity
    0.88
     likeness
    0.87
     resemblance
    0.85
     proximity
    0.85
     color
    0.85
    Act Density 0.296%

    No Known Activations