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

    Neuronpedia

    INDEX
    Explanations

    indefinite articles (a or an) preceding specific nouns

    oai_token-act-pair · gpt-3.5-turbo

    articles and certain pronouns in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Accounts
    -0.75
    books
    -0.70
    edit
    -0.68
    Reports
    -0.68
    Edit
    -0.66
    Events
    -0.64
    achu
    -0.63
     Cases
    -0.63
    reports
    -0.63
    agree
    -0.62
    POSITIVE LOGITS
     bunch
    0.88
     handful
    0.84
     lot
    0.81
    rouse
    0.80
    cknowled
    0.79
     stationary
    0.76
     dozen
    0.75
     piece
    0.74
    venge
    0.73
     precursor
    0.73
    Act Density 0.221%

    No Known Activations