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

    Neuronpedia

    INDEX
    Explanations

    specific calendar dates expressed in number format

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "dates."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ui
    -0.66
    ack
    -0.66
    immer
    -0.64
     helpless
    -0.63
    morph
    -0.61
     despair
    -0.60
    itudinal
    -0.60
    asm
    -0.60
    ipl
    -0.59
    tackle
    -0.59
    POSITIVE LOGITS
     dates
    4.23
     Dates
    3.12
     date
    2.75
    dates
    2.37
    date
    2.12
     Date
    1.92
     dating
    1.69
    Date
    1.65
     dated
    1.59
    dated
    1.51
    Act Density 0.009%

    No Known Activations