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

    Neuronpedia

    INDEX
    Explanations

    hyphenated numerical sequences

    oai_token-act-pair · gpt-3.5-turbo

    numeric representations of time periods, particularly in relation to multi-part constructs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Borders
    -0.70
     Rica
    -0.68
     Lovecraft
    -0.68
     Richards
    -0.67
     Lanka
    -0.67
     McGr
    -0.65
     Gillespie
    -0.64
     Nero
    -0.64
     Proc
    -0.62
     Reyes
    -0.62
    POSITIVE LOGITS
    thirds
    1.45
    legged
    1.41
    year
    1.37
    dimensional
    1.32
    sided
    1.29
    fif
    1.28
    month
    1.28
    footed
    1.26
    digit
    1.25
    bedroom
    1.22
    Act Density 0.059%

    No Known Activations