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

    Neuronpedia

    INDEX
    Explanations

    words related to seasonal events or holidays

    oai_token-act-pair · gpt-3.5-turbo

    references to Halloween and related festivities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    serv
    -0.80
    ital
    -0.72
    Roberts
    -0.72
    ilit
    -0.68
     oblig
    -0.65
    yi
    -0.65
     UNCLASSIFIED
    -0.64
     interven
    -0.63
    SPA
    -0.63
     lender
    -0.63
    POSITIVE LOGITS
     Halloween
    3.78
    alloween
    2.35
     Thanksgiving
    1.97
     Christmas
    1.93
     Hallow
    1.81
    Christmas
    1.71
     Valentine
    1.61
     Pumpkin
    1.57
     Easter
    1.51
     Haunted
    1.45
    Act Density 0.016%

    No Known Activations