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

    Neuronpedia

    INDEX
    Explanations

    lists of items marked by 'etc' or similar abbreviations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the term "etc." as a shorthand for listing items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Lords
    -0.80
    inka
    -0.78
    wives
    -0.69
     Advocate
    -0.66
     Kazakhstan
    -0.62
     shadow
    -0.62
     Dwarf
    -0.61
     Gor
    -0.61
     Lord
    -0.61
     Gos
    -0.60
    POSITIVE LOGITS
    eter
    1.23
    etc
    1.11
    .,
    1.04
    ignt
    0.96
    .?
    0.94
    ãģĨ
    0.93
    uits
    0.91
    eteria
    0.90
     etc
    0.89
    .).
    0.88
    Act Density 0.016%

    No Known Activations