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

    Neuronpedia

    INDEX
    Explanations

    mentions of closures or things being closed

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "closed"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iage
    -0.82
    arat
    -0.75
    oi
    -0.74
    ãĤ¼ãĤ¦ãĤ¹
    -0.73
    aire
    -0.72
    omething
    -0.71
    nik
    -0.71
    Catalog
    -0.69
    Corn
    -0.68
    ican
    -0.68
    POSITIVE LOGITS
     doors
    0.82
     shutter
    0.78
     throttle
    0.77
     curtains
    0.76
     Doors
    0.71
     lid
    0.71
     frontal
    0.70
    closed
    0.70
     closed
    0.69
     scrimmage
    0.69
    Act Density 0.018%

    No Known Activations