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

    Neuronpedia

    INDEX
    Explanations

    phrases related to geographical locations, particularly islands

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific islands and related terms in geopolitical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ilyn
    -0.83
    baugh
    -0.79
    itbart
    -0.76
    anooga
    -0.74
    OUGH
    -0.74
    earable
    -0.72
    orr
    -0.72
    WORK
    -0.71
    IENCE
    -0.71
    kus
    -0.71
    POSITIVE LOGITS
     Islands
    1.52
     islands
    1.03
     Island
    0.88
     Peninsula
    0.85
     peninsula
    0.83
     Strait
    0.82
     Albania
    0.80
     island
    0.77
     Petroleum
    0.76
     Philippines
    0.74
    Act Density 0.042%

    No Known Activations