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

    Neuronpedia

    INDEX
    Explanations

    keywords related to underground locations or activities

    oai_token-act-pair · gpt-3.5-turbo

    references to underground environments or structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cific
    -0.86
    otaur
    -0.83
    rar
    -0.79
    atche
    -0.76
    awed
    -0.75
    nia
    -0.75
    ointed
    -0.75
    ettle
    -0.74
    =-=-=-=-=-=-=-=-
    -0.74
    erva
    -0.74
    POSITIVE LOGITS
     underground
    0.95
     Railroad
    0.86
     cavern
    0.82
     railroad
    0.82
     tunnels
    0.81
     labyrinth
    0.81
     bunker
    0.80
     mole
    0.79
     passages
    0.79
     hatch
    0.78
    Act Density 0.013%

    No Known Activations