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

    Neuronpedia

    INDEX
    Explanations

    words related to physical barriers like barricades and scaffolding

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with barriers or obstructions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ME
    -0.80
    DAY
    -0.76
    Common
    -0.75
    EVA
    -0.75
    common
    -0.73
    day
    -0.73
    OTAL
    -0.73
    20439
    -0.72
    American
    -0.71
    ————————
    -0.70
    POSITIVE LOGITS
     barric
    1.50
     besie
    0.99
     defences
    0.85
     encl
    0.84
     curtains
    0.82
     jams
    0.79
     inval
    0.78
     submer
    0.77
     exting
    0.77
     evac
    0.77
    Act Density 0.006%

    No Known Activations