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

    Neuronpedia

    INDEX
    Explanations

    specific days of the week, especially 'Tuesday'

    oai_token-act-pair · gpt-3.5-turbo

    references to the day "Tuesday."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ript
    -0.97
    abet
    -0.90
    ordes
    -0.80
    onductor
    -0.76
    keye
    -0.75
    ris
    -0.75
    aeus
    -0.75
    luster
    -0.74
    agos
    -0.74
    egal
    -0.73
    POSITIVE LOGITS
     morning
    1.53
     afternoon
    1.42
     night
    1.35
     evening
    1.33
     mornings
    1.27
    morning
    1.13
     Night
    1.09
     nights
    1.03
     evenings
    0.97
     Morning
    0.97
    Act Density 0.030%

    No Known Activations