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

    Neuronpedia

    INDEX
    Explanations

    phrases separated by punctuations like quotes and periods

    oai_token-act-pair · gpt-3.5-turbo

    conversational phrases and statements expressing opinions or sentiments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tremend
    -0.86
     citiz
    -0.73
     eleph
    -0.71
     oun
    -0.70
     unnecess
    -0.70
     metic
    -0.69
     occas
    -0.67
     hemor
    -0.67
     exha
    -0.63
    eatures
    -0.63
    POSITIVE LOGITS
    ↵
    0.69
    035
    0.68
    Indeed
    0.68
    PHOTOS
    0.66
    Attempts
    0.66
    Chel
    0.66
    Correct
    0.64
    DragonMagazine
    0.63
    ????????
    0.63
    Beta
    0.63
    Act Density 0.286%

    No Known Activations