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

    Neuronpedia

    INDEX
    Explanations

    words related to physical structures, specifically roofs

    oai_token-act-pair · gpt-3.5-turbo

    references to roofs and roof structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    avez
    -0.80
     MacArthur
    -0.77
    ISTER
    -0.77
     Lauder
    -0.77
     Schumer
    -0.75
    ADRA
    -0.70
    nesia
    -0.70
    bourg
    -0.68
     Norn
    -0.65
     Spiel
    -0.65
    POSITIVE LOGITS
     roof
    1.04
     terr
    0.94
    deck
    0.91
     rack
    0.87
     tops
    0.87
     wip
    0.82
     canopy
    0.82
     curtain
    0.80
     roofs
    0.80
    ilion
    0.80
    Act Density 0.015%

    No Known Activations