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

    Neuronpedia

    INDEX
    Explanations

    instances of actions related to drinking

    oai_token-act-pair · gpt-3.5-turbo

    references to drinking beverages, particularly alcoholic ones

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Velocity
    -0.76
    arte
    -0.74
    Vel
    -0.72
     Catalog
    -0.69
     pivot
    -0.69
     Kinnikuman
    -0.68
    Catalog
    -0.67
    eta
    -0.65
    ilater
    -0.65
    odes
    -0.62
    POSITIVE LOGITS
     drinking
    3.82
     Drinking
    2.86
     drink
    2.24
     drinkers
    2.12
     drank
    2.04
     drinks
    1.85
     drunk
    1.73
     partying
    1.72
     Drink
    1.69
     beverage
    1.65
    Act Density 0.011%

    No Known Activations