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

    Neuronpedia

    INDEX
    Explanations

    proper nouns, particularly names, such as "Hoffman."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Hoffman."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arily
    -0.80
    acular
    -0.77
    carry
    -0.70
    ebook
    -0.69
    kered
    -0.69
    cation
    -0.68
    uses
    -0.66
    fully
    -0.66
    inventoryQuantity
    -0.66
     Acts
    -0.64
    POSITIVE LOGITS
     Hoffman
    1.18
    roth
    0.86
    quist
    0.83
    qv
    0.80
    gren
    0.79
    haus
    0.78
     pitched
    0.76
    kson
    0.73
     Est
    0.71
    strom
    0.70
    Act Density 0.009%

    No Known Activations