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

    Neuronpedia

    INDEX
    Explanations

    incorrectly spelled words marked with the abbreviation "sic."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sic" used to indicate a quote or a citation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    VIEW
    -0.79
    eus
    -0.78
    Recomm
    -0.73
    ships
    -0.72
    çĦ
    -0.71
    places
    -0.70
    MQ
    -0.68
    woods
    -0.68
    yards
    -0.67
    Winged
    -0.67
    POSITIVE LOGITS
    sic
    0.94
    querade
    0.72
     offend
    0.71
    oman
    0.70
    lass
    0.64
    afe
    0.64
    atin
    0.64
     hij
    0.64
    cia
    0.62
    hin
    0.62
    Act Density 0.008%

    No Known Activations