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

    Neuronpedia

    INDEX
    Explanations

    words related to geopolitics and international relations

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to countries, organizations, and positions of power

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    prem
    -0.83
    ounded
    -0.78
     Supplemental
    -0.69
    hai
    -0.69
    iceps
    -0.66
    utenberg
    -0.64
    ounding
    -0.64
    bah
    -0.61
    ounds
    -0.61
    irens
    -0.60
    POSITIVE LOGITS
     imaginable
    1.30
     whatsoever
    1.14
     conceivable
    1.01
     except
    0.90
     besides
    0.75
     describ
    0.74
     anywhere
    0.74
    Else
    0.70
     alike
    0.68
    soever
    0.68
    Act Density 0.233%

    No Known Activations