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

    Neuronpedia

    INDEX
    Explanations

    expressions related to winning or being declared as the winner

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of "winner" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    elsius
    -0.70
    uli
    -0.69
    estation
    -0.69
    ugu
    -0.68
    rients
    -0.68
    aeda
    -0.67
    uku
    -0.67
    abb
    -0.66
    uld
    -0.66
    MpServer
    -0.66
    POSITIVE LOGITS
     winner
    0.98
     Winner
    0.92
    Winner
    0.89
     winners
    0.82
     laureate
    0.81
     prize
    0.80
     loser
    0.76
     Hugo
    0.74
     contestant
    0.72
    athon
    0.72
    Act Density 0.012%

    No Known Activations