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

    Neuronpedia

    INDEX
    Explanations

    characteristics related to human identity or personal attributes like race, ethnicity, religion, sexual orientation, nationality, and physical features

    oai_token-act-pair · gpt-3.5-turbo

    terms related to identity and discrimination based on various characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    EMS
    -0.76
    imar
    -0.76
     Oper
    -0.75
    ERG
    -0.69
    INTER
    -0.67
    vae
    -0.66
    aug
    -0.66
    ald
    -0.66
    Dispatch
    -0.65
     Rak
    -0.65
    POSITIVE LOGITS
     ancestry
    0.88
     affiliation
    0.86
     backgrounds
    0.85
     ethnicity
    0.81
     coloring
    0.77
     discrimination
    0.77
     prejudice
    0.76
     affili
    0.76
     preference
    0.76
     stripe
    0.74
    Act Density 0.096%

    No Known Activations