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

    Neuronpedia

    INDEX
    Explanations

    locations or places that hold political, cultural, or economic significance

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    potion
    -0.82
    sbm
    -0.79
    AUT
    -0.78
    sters
    -0.75
    wd
    -0.73
    eker
    -0.72
    ĪĴ
    -0.71
    fortune
    -0.71
    akers
    -0.71
    adr
    -0.71
    POSITIVE LOGITS
     metropolitan
    0.95
     city
    0.92
     cities
    0.85
     suburb
    0.80
    itals
    0.78
     metro
    0.75
     punishment
    0.73
     Citiz
    0.72
    omach
    0.71
     Wasteland
    0.70
    Act Density 9.845%

    No Known Activations