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

    Neuronpedia

    INDEX
    Explanations

    tokens that are encoded in a specific format resembling identifiers

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the number four

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vier
    -0.76
    esville
    -0.76
    venue
    -0.72
     Whitman
    -0.72
     Cheong
    -0.70
    clair
    -0.64
    becca
    -0.62
    iday
    -0.61
     Winchester
    -0.61
     Vaugh
    -0.61
    POSITIVE LOGITS
    teenth
    1.22
    teen
    1.16
    eva
    1.00
    hyde
    0.86
    Chan
    0.83
    some
    0.82
    66666666
    0.79
    amaz
    0.79
    cyl
    0.79
    th
    0.78
    Act Density 0.083%

    No Known Activations