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

    Neuronpedia

    INDEX
    Explanations

    references to completeness or entirety

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "entire" as it relates to comprehensive situations or subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chairs
    -0.88
    bits
    -0.83
    friends
    -0.79
    none
    -0.75
    facts
    -0.75
    pots
    -0.73
    =-=-=-=-=-=-=-=-
    -0.72
    bis
    -0.69
    plugins
    -0.69
    Examples
    -0.69
    POSITIVE LOGITS
     spectrum
    1.05
     arsenal
    1.00
     continent
    1.00
     length
    0.94
     lifespan
    0.93
     globe
    0.92
     swath
    0.92
     duration
    0.91
     range
    0.90
     repertoire
    0.89
    Act Density 0.039%

    No Known Activations