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

    Neuronpedia

    INDEX
    Explanations

    comparisons where something is more significant or prominent than any other item in the set

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing comparison or superiority of one entity or concept over others

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    istries
    -0.73
     Christy
    -0.63
    dal
    -0.63
    ories
    -0.62
    yna
    -0.62
     syndrome
    -0.61
    ulative
    -0.60
    oute
    -0.58
    CBC
    -0.58
    endas
    -0.58
    POSITIVE LOGITS
    THING
    1.08
     imaginable
    0.97
     conceivable
    0.91
    ONE
    0.76
    WHERE
    0.74
     ordinary
    0.73
     else
    0.70
     recip
    0.69
     OTHER
    0.67
     imagined
    0.65
    Act Density 0.054%

    No Known Activations