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

    Neuronpedia

    INDEX
    Explanations

    quantities or measurements, particularly focusing on sizes such as ounces or grams

    oai_token-act-pair · gpt-3.5-turbo

    measurements and quantities related to food and substances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     architecture
    -0.76
    istically
    -0.67
     inev
    -0.64
     sear
    -0.64
    stru
    -0.64
    gotten
    -0.62
     imperson
    -0.61
     afterlife
    -0.61
    neg
    -0.61
     Exile
    -0.61
    POSITIVE LOGITS
     bottles
    1.16
     bottle
    0.98
     cans
    0.93
    Quantity
    0.92
     servings
    0.91
     increments
    0.91
     apiece
    0.90
    cake
    0.86
    poons
    0.84
     jars
    0.82
    Act Density 0.056%

    No Known Activations