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

    Neuronpedia

    INDEX
    Explanations

    yearly ages

    oai_token-act-pair · gpt-3.5-turbo

    references to age in years

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Rocky
    -0.68
     Proc
    -0.65
     gum
    -0.65
     Cassidy
    -0.65
     Prometheus
    -0.64
     Twain
    -0.62
     Rossi
    -0.61
     Swift
    -0.60
     Narc
    -0.59
     Borders
    -0.59
    POSITIVE LOGITS
    degree
    1.34
    year
    1.25
    million
    1.14
    inch
    1.13
    hour
    1.12
    minute
    1.08
    percent
    1.08
    years
    1.03
    member
    0.99
    billion
    0.99
    Act Density 0.073%

    No Known Activations