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

    Neuronpedia

    INDEX
    Explanations

    time-related information in months, years, and percentages

    oai_token-act-pair · gpt-3.5-turbo

    numerical data and statistics related to time periods and percentages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chwitz
    -0.76
    especially
    -0.71
    multipl
    -0.68
     plent
    -0.63
    lay
    -0.61
     indo
    -0.61
    alle
    -0.59
     manifold
    -0.58
    ramid
    -0.58
    trap
    -0.58
    POSITIVE LOGITS
     percent
    0.93
     cents
    0.86
    %.
    0.85
    %
    0.80
     hours
    0.76
     minutes
    0.75
     anymore
    0.72
     seconds
    0.72
    ¢
    0.71
     weeks
    0.70
    Act Density 0.154%

    No Known Activations