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

    Neuronpedia

    INDEX
    Explanations

    words related to alphabets or alphabetical order

    oai_token-act-pair · gpt-3.5-turbo

    references to the alphabet and its various characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    erm
    -0.73
    db
    -0.70
     Anderson
    -0.69
    cca
    -0.69
    arte
    -0.68
    mb
    -0.68
    erer
    -0.67
    dm
    -0.67
    DB
    -0.65
     Protesters
    -0.63
    POSITIVE LOGITS
     alphabet
    1.33
    ically
    1.05
    ical
    1.00
     soup
    0.99
    abet
    0.94
    alogy
    0.89
    matical
    0.83
     Soup
    0.79
    icals
    0.79
    seed
    0.79
    Act Density 0.014%

    No Known Activations