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

    Neuronpedia

    INDEX
    Explanations

    references to geographical locations, with a strong preference for Europe

    oai_token-act-pair · gpt-3.5-turbo

    references to Europe

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oru
    -0.78
    ymm
    -0.77
    anan
    -0.77
    yright
    -0.75
    atchewan
    -0.72
    LESS
    -0.69
    aron
    -0.68
    yrights
    -0.67
    ledged
    -0.66
    anamo
    -0.66
    POSITIVE LOGITS
     Union
    0.90
     Parliament
    0.89
     countries
    0.82
     continent
    0.78
     nations
    0.78
     capitals
    0.75
     Continent
    0.74
     Galile
    0.73
     Countries
    0.72
    ffen
    0.69
    Act Density 0.042%

    No Known Activations