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

    Neuronpedia

    INDEX
    Explanations

    references to lighting conditions or fixtures

    oai_token-act-pair · gpt-3.5-turbo

    references to lighting and illumination

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    clair
    -0.81
    sonian
    -0.79
    llah
    -0.79
    berus
    -0.76
    sburgh
    -0.74
    CHAT
    -0.73
    essor
    -0.70
    ileaks
    -0.70
    gery
    -0.69
    irlf
    -0.69
    POSITIVE LOGITS
     torches
    0.99
     bulb
    0.98
     bulbs
    0.95
     lighting
    0.95
     candles
    0.89
     fixtures
    0.88
     poles
    0.87
     lamps
    0.83
     overhead
    0.80
     setups
    0.79
    Act Density 0.014%

    No Known Activations