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

    Neuronpedia

    INDEX
    Explanations

    references to closure or closing of physical locations or events

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    icist
    -1.16
    nis
    -1.14
    orum
    -1.11
    ican
    -1.09
    erous
    -1.08
    andum
    -1.04
    eday
    -1.04
     endors
    -1.04
    nik
    -1.01
    ique
    -0.99
    POSITIVE LOGITS
     doors
    1.41
     Doors
    1.32
     shutter
    1.25
     curtains
    1.23
     loopholes
    1.23
    doors
    1.15
     door
    1.09
     throttle
    1.07
    door
    1.03
     wounds
    1.00
    Act Density 0.548%

    No Known Activations