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

    Neuronpedia

    INDEX
    Explanations

    descriptions of physical actions or commands given to others

    oai_token-act-pair · gpt-3.5-turbo

    actions related to threats and violence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pmwiki
    -0.83
     egreg
    -0.78
    [/
    -0.73
     anecd
    -0.72
     notably
    -0.70
     rhet
    -0.69
     incent
    -0.68
    iosyncr
    -0.68
    ensibly
    -0.68
     quasi
    -0.67
    POSITIVE LOGITS
     someday
    0.86
     hers
    0.82
     supper
    0.80
     him
    0.80
     daddy
    0.76
     Daddy
    0.76
     pills
    0.73
     his
    0.71
     Allah
    0.71
     downstairs
    0.69
    Act Density 0.979%

    No Known Activations