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

    Neuronpedia

    INDEX
    Explanations

    references to geographic or political territories

    oai_token-act-pair · gpt-3.5-turbo

    references to territorial concepts or discussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    odcast
    -0.81
    kus
    -0.76
    ever
    -0.76
    apers
    -0.73
    racted
    -0.71
    lder
    -0.69
    oster
    -0.69
    nder
    -0.65
    Episode
    -0.65
    lower
    -0.65
    POSITIVE LOGITS
     territory
    0.89
     boundaries
    0.89
     territories
    0.80
     Territory
    0.79
     Territories
    0.77
     boundary
    0.69
     Crossing
    0.69
     Borders
    0.66
    oslov
    0.65
    trl
    0.65
    Act Density 0.013%

    No Known Activations