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

    Neuronpedia

    INDEX
    Explanations

    references to time-based events occurring in the future

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "tomorrow."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    leeve
    -0.78
    edIn
    -0.78
    onne
    -0.77
    ership
    -0.72
    hip
    -0.72
    hips
    -0.67
    ochet
    -0.67
    ordes
    -0.67
    ounty
    -0.65
    ec
    -0.65
    POSITIVE LOGITS
     morning
    1.60
     afternoon
    1.54
     evening
    1.44
     night
    1.38
    morning
    1.20
     mornings
    1.18
     nights
    1.11
    night
    1.00
     Morning
    0.99
     evenings
    0.97
    Act Density 0.049%

    No Known Activations