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

    Neuronpedia

    INDEX
    Explanations

    references to physical locations, specifically basements

    oai_token-act-pair · gpt-3.5-turbo

    references to basements or cellar spaces

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Cola
    -0.77
    nir
    -0.75
    acting
    -0.70
    uality
    -0.69
    yt
    -0.69
    uncture
    -0.68
    utan
    -0.67
    NG
    -0.67
    weather
    -0.66
    Population
    -0.65
    POSITIVE LOGITS
     basement
    1.16
    stairs
    0.96
     cellar
    0.94
     lid
    0.87
     Dwell
    0.87
     courtyard
    0.85
    door
    0.85
     attic
    0.84
     closet
    0.84
     apartment
    0.83
    Act Density 0.010%

    No Known Activations