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

    Neuronpedia

    INDEX
    Explanations

    instances of physical locations being described in relation to the space around them - either involving boundaries like fences or featuring nearby landmarks, streets, and direction markers

    oai_token-act-pair · gpt-3.5-turbo

    references to locations or properties related to physical spaces

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lance
    -0.59
    umers
    -0.51
    uria
    -0.49
    vertising
    -0.49
    oaded
    -0.47
    negative
    -0.47
    IJ
    -0.46
    Versions
    -0.45
    follow
    -0.45
    software
    -0.45
    POSITIVE LOGITS
     courtyard
    0.89
     driveway
    0.85
     cemetery
    0.83
     creek
    0.83
     plaza
    0.80
     patio
    0.78
     clubhouse
    0.78
     WTC
    0.77
     lake
    0.76
     gardens
    0.75
    Act Density 0.573%

    No Known Activations