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

    Neuronpedia

    INDEX
    Explanations

    phrases related to physical structures or passages, especially corridors

    oai_token-act-pair · gpt-3.5-turbo

    references to physical pathways or corridors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iak
    -0.82
    odium
    -0.76
    itude
    -0.72
    onom
    -0.70
    agi
    -0.69
    lied
    -0.69
    held
    -0.69
     Glory
    -0.68
    orah
    -0.68
    emark
    -0.67
    POSITIVE LOGITS
    ridor
    1.00
     corridors
    0.97
     routes
    0.90
     corridor
    0.89
     Corridor
    0.88
    ways
    0.87
     pathways
    0.83
    ttes
    0.82
     divid
    0.80
     passages
    0.77
    Act Density 0.028%

    No Known Activations