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

    Neuronpedia

    INDEX
    Explanations

    references to articles or sections in a document

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of "Article," indicating a focus on sections or references made in a document

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥīãĥ©ãĤ´ãĥ³
    -0.79
    awar
    -0.77
    ascus
    -0.77
    hiba
    -0.76
    boa
    -0.76
    etsk
    -0.76
    ergic
    -0.75
    adows
    -0.75
    escal
    -0.75
    cffffcc
    -0.74
    POSITIVE LOGITS
    ICLE
    0.83
     Continued
    0.81
     Articles
    0.79
    meal
    0.74
    hook
    0.71
    ual
    0.69
    witz
    0.68
     Mobil
    0.67
     Consent
    0.66
    Article
    0.64
    Act Density 0.018%

    No Known Activations