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

    Neuronpedia

    INDEX
    Explanations

    proper names related to individuals

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns, particularly names of individuals and their repeated mentions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     bour
    -0.68
    inen
    -0.68
    amed
    -0.67
    ament
    -0.66
    aku
    -0.63
    estinal
    -0.63
    ishly
    -0.63
    isites
    -0.62
    chest
    -0.61
    enger
    -0.60
    POSITIVE LOGITS
    rouch
    0.86
    ursor
    0.83
    nces
    0.75
    ancel
    0.72
     ashes
    0.71
    rossover
    0.69
    urrency
    0.69
    cled
    0.69
    aught
    0.68
    anyon
    0.68
    Act Density 0.186%

    No Known Activations