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

    Neuronpedia

    INDEX
    Explanations

    references related to the term "Gram" which could include the term "Grammy" as well as other contextual references to "Gram" such as "grammatical" or "program."

    oai_token-act-pair · gpt-3.5-turbo

    references to measurements and units of weight, particularly grams

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Occupations
    -0.75
    Madison
    -0.68
    idth
    -0.68
    quished
    -0.64
    landish
    -0.64
     Draper
    -0.64
    Downloadha
    -0.64
    ño
    -0.62
    olve
    -0.62
    ITNESS
    -0.62
    POSITIVE LOGITS
    matical
    1.49
    matically
    1.32
    ophone
    1.13
    ming
    1.03
    matic
    0.95
    stones
    0.94
    bling
    0.93
    mers
    0.88
    mys
    0.88
    atically
    0.86
    Act Density 0.021%

    No Known Activations