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

    Neuronpedia

    INDEX
    Explanations

    keywords related to specific locations, specifically "South Africa" in this case

    oai_token-act-pair · gpt-3.5-turbo

    references to the "South" in geographic or political contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     cu
    -0.77
     silence
    -0.70
     result
    -0.70
     non
    -0.68
     rolling
    -0.68
     hex
    -0.67
     battlefield
    -0.67
     Gall
    -0.66
     favor
    -0.65
     conversion
    -0.65
    POSITIVE LOGITS
    South
    3.14
    south
    2.21
    North
    1.75
    West
    1.63
    East
    1.60
    north
    1.48
    Southern
    1.43
    African
    1.29
    Northern
    1.28
    Pak
    1.26
    Act Density 0.023%

    No Known Activations