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

    Neuronpedia

    INDEX
    Explanations

    references to global presence or coverage

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    BIL
    -0.80
    milo
    -0.78
    raped
    -0.74
    Accessory
    -0.74
    HB
    -0.69
    tests
    -0.69
    isons
    -0.68
    Tokens
    -0.67
    Completed
    -0.67
    SpaceEngineers
    -0.64
    POSITIVE LOGITS
     globe
    1.92
     world
    1.73
     country
    1.66
     continent
    1.51
     nation
    1.41
     planet
    1.38
     hemisphere
    1.30
     Americas
    1.21
     WORLD
    1.21
     Continent
    1.20
    Act Density 0.097%

    No Known Activations