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

    Neuronpedia

    INDEX
    Explanations

    references to gender or specific groups of people (e.g. girls, ladies)

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "girls" and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rehend
    -0.81
    eering
    -0.79
    OLOG
    -0.76
    BLIC
    -0.73
    Closure
    -0.71
    utherford
    -0.71
    SPONSORED
    -0.71
    PDATE
    -0.70
    rawdownloadcloneembedreportprint
    -0.68
    OLOGY
    -0.67
    POSITIVE LOGITS
    folk
    1.02
     girls
    0.95
     Scouts
    0.90
    girls
    0.88
     panties
    0.86
    hips
    0.84
    Girls
    0.84
    mith
    0.82
     Girls
    0.79
    riages
    0.79
    Act Density 0.023%

    No Known Activations