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

    Neuronpedia

    INDEX
    Explanations

    words related to physical structures or objects like rooftops

    oai_token-act-pair · gpt-3.5-turbo

    mentions of roofs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     MacArthur
    -0.80
    avez
    -0.76
     Sacrament
    -0.69
    Pacific
    -0.65
     Lauder
    -0.65
    Americans
    -0.65
     Western
    -0.64
    ãĢIJ
    -0.63
     Norn
    -0.63
    Clar
    -0.63
    POSITIVE LOGITS
     roof
    1.10
    deck
    0.93
     roofs
    0.88
     rack
    0.86
     loft
    0.84
     ceiling
    0.81
    ©¶æ
    0.81
    ilion
    0.80
     hatch
    0.80
     terr
    0.78
    Act Density 0.009%

    No Known Activations