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

    Neuronpedia

    INDEX
    Explanations

    mentions related to former political figures

    oai_token-act-pair · gpt-3.5-turbo

    references to political figures or their former positions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    otle
    -0.80
    ramid
    -0.79
    anguage
    -0.78
    iquette
    -0.73
    utterstock
    -0.73
    aris
    -0.71
    achus
    -0.71
    oking
    -0.70
    uden
    -0.70
    ourced
    -0.70
    POSITIVE LOGITS
     Yugoslav
    1.04
     Yugoslavia
    1.03
     President
    0.95
     presidents
    0.93
     president
    0.93
     Soviet
    0.93
     Presidents
    0.92
     colleague
    0.91
     classmate
    0.88
     classmates
    0.87
    Act Density 0.040%

    No Known Activations