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

    Neuronpedia

    INDEX
    Explanations

    the word "Ras" with varying levels of emphasis

    oai_token-act-pair · gpt-3.5-turbo

    the syllable "ras" within words

    oai_token-act-pair · gpt-4-turbo

    terms associated with the city of Madras (now Chennai)

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĨĴ
    -0.70
    riad
    -0.68
    tails
    -0.63
    tains
    -0.62
     Yankee
    -0.62
    ĵ
    -0.61
    glers
    -0.61
     warrants
    -0.61
    Roberts
    -0.60
     pointer
    -0.60
    POSITIVE LOGITS
    irez
    1.05
    hin
    1.01
    quez
    0.98
    ques
    0.97
    arin
    0.95
    sth
    0.92
    yout
    0.92
    hea
    0.91
    mond
    0.91
    agna
    0.90
    Act Density 0.010%

    No Known Activations