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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "sugar."

    oai_token-act-pair · gpt-3.5-turbo

    instances related to sugar content and its various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ynski
    -0.77
    atche
    -0.74
    naire
    -0.73
    Downloadha
    -0.72
    acular
    -0.71
    agall
    -0.70
    agame
    -0.70
    ahime
    -0.69
    ļé
    -0.68
    ovie
    -0.68
    POSITIVE LOGITS
     syrup
    1.13
     beet
    1.12
     cane
    1.05
     sugar
    0.89
     sweetness
    0.80
    ptive
    0.78
     starch
    0.76
     fructose
    0.76
     coating
    0.75
     desserts
    0.75
    Act Density 0.016%

    No Known Activations