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

    Neuronpedia

    INDEX
    Explanations

    locations or landmarks, especially mountains and volcanoes

    oai_token-act-pair · gpt-3.5-turbo

    names of geographical locations, particularly mountains and forts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    organic
    -0.68
    inant
    -0.65
    staking
    -0.63
    iegel
    -0.63
    itled
    -0.62
     Spiegel
    -0.61
     Hyundai
    -0.61
    vertising
    -0.61
     arrang
    -0.60
    abase
    -0.59
    POSITIVE LOGITS
     Reserv
    0.81
    burgh
    0.78
    llan
    0.74
    arde
    0.74
     Island
    0.73
     Despair
    0.73
     Islands
    0.71
    helm
    0.71
     Cemetery
    0.70
    obyl
    0.70
    Act Density 0.108%

    No Known Activations