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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns, particularly the number 4

    oai_token-act-pair · gpt-3.5-turbo

    instances of the number four in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vier
    -0.72
     Whitman
    -0.71
    esville
    -0.70
    venue
    -0.66
    idence
    -0.65
    UTION
    -0.64
     Vaugh
    -0.62
     woo
    -0.61
    becca
    -0.60
    aries
    -0.59
    POSITIVE LOGITS
    teenth
    1.42
    teen
    1.36
    eva
    1.10
    some
    0.96
    hyde
    0.91
    Kids
    0.91
    Chan
    0.86
    square
    0.85
    cyl
    0.82
    amaz
    0.82
    Act Density 0.121%

    No Known Activations