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

    Neuronpedia

    INDEX
    Explanations

    hyphenated phrases indicating a comparison or distinction

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to higher income or socioeconomic status

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Dickinson
    -0.81
    ulhu
    -0.74
     Gazette
    -0.73
     Pengu
    -0.71
     Wiz
    -0.70
     Quan
    -0.69
     Cotton
    -0.68
     Saud
    -0.67
     Indigo
    -0.67
     Drink
    -0.66
    POSITIVE LOGITS
    than
    1.70
    sounding
    1.17
    upper
    1.09
    worldly
    1.07
    educated
    1.05
    biased
    1.00
    edged
    1.00
    looking
    1.00
    ranked
    0.98
    leaning
    0.98
    Act Density 0.064%

    No Known Activations