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

    Neuronpedia

    INDEX
    Explanations

    references to marathon-related events or activities

    oai_token-act-pair · gpt-3.5-turbo

    references to marathons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    metics
    -0.82
    ongyang
    -0.69
     Aires
    -0.69
    otle
    -0.69
    obyl
    -0.68
    orum
    -0.68
    unin
    -0.66
    choes
    -0.66
    onom
    -0.65
    sheet
    -0.64
    POSITIVE LOGITS
     runner
    1.04
     runners
    0.88
     Runner
    0.86
     marathon
    0.83
     bombing
    0.82
     bombings
    0.81
    athon
    0.77
     Marathon
    0.73
    NING
    0.71
     bombers
    0.70
    Act Density 0.033%

    No Known Activations