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

    Neuronpedia

    INDEX
    Explanations

    numerical values at the beginning of specific patterns

    oai_token-act-pair · gpt-3.5-turbo

    the number four in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    venue
    -0.71
     Cheong
    -0.69
    vier
    -0.69
     Whitman
    -0.67
    iday
    -0.65
    esville
    -0.64
     Winchester
    -0.64
    becca
    -0.62
     woo
    -0.61
    arial
    -0.61
    POSITIVE LOGITS
    teenth
    1.28
    teen
    1.22
    eva
    1.06
    hyde
    0.94
    some
    0.87
    Kids
    0.85
    amaz
    0.85
    cyl
    0.83
    Chan
    0.82
    66666666
    0.81
    Act Density 0.100%

    No Known Activations