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

    Neuronpedia

    INDEX
    Explanations

    phrases enclosed in quotation marks

    oai_token-act-pair · gpt-3.5-turbo

    quoted dialogue or text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Klu
    -0.81
     Flavoring
    -0.78
    hattan
    -0.63
    omorphic
    -0.62
    abase
    -0.62
     peg
    -0.62
     auditor
    -0.61
     Unicode
    -0.61
     sshd
    -0.60
     chair
    -0.60
    POSITIVE LOGITS
    isSpecialOrderable
    0.99
    soft
    0.94
    WHERE
    0.93
    Where
    0.86
    Returns
    0.83
    Hello
    0.82
    altern
    0.82
    Good
    0.81
    Sport
    0.81
    FOR
    0.80
    Act Density 0.066%

    No Known Activations