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

    Neuronpedia

    INDEX
    Explanations

    references to confectionery items, particularly variations of the word "candy."

    oai_token-act-pair · gpt-3.5-turbo

    references to candy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lihood
    -0.76
    iets
    -0.71
    inen
    -0.70
    yon
    -0.68
     transcript
    -0.68
    chron
    -0.65
     plur
    -0.64
    ugal
    -0.64
    phrine
    -0.63
    ebin
    -0.62
    POSITIVE LOGITS
     cane
    1.13
     candy
    0.98
    strip
    0.97
    mallow
    0.97
    bucks
    0.92
    flake
    0.87
     gum
    0.83
     daddy
    0.82
     sweets
    0.82
     Candy
    0.82
    Act Density 0.021%

    No Known Activations