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

    Neuronpedia

    INDEX
    Explanations

    spokespeople or representatives mentioned in texts

    oai_token-act-pair · gpt-3.5-turbo

    references to female figures and spokespersons in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    kefeller
    -0.87
    ype
    -0.81
    iago
    -0.77
    ypes
    -0.74
    oday
    -0.74
    vernment
    -0.73
    ustom
    -0.71
    acca
    -0.71
    venants
    -0.70
    resy
    -0.70
    POSITIVE LOGITS
     Anne
    1.29
     Marie
    1.23
     Louise
    1.18
     herself
    1.10
     Elizabeth
    1.08
     Mae
    1.08
     Isabel
    1.08
     Mary
    1.07
     Patricia
    1.06
     Diana
    1.05
    Act Density 0.159%

    No Known Activations