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

    Neuronpedia

    INDEX
    Explanations

    references to various locations specified as "the capital."

    oai_token-act-pair · gpt-3.5-turbo

    references to geographic capitals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    AUT
    -0.81
    sbm
    -0.80
    potion
    -0.75
    bring
    -0.71
    MpServer
    -0.68
    hner
    -0.68
    ĪĴ
    -0.68
    chrom
    -0.66
    fortune
    -0.64
    akers
    -0.64
    POSITIVE LOGITS
     metropolitan
    0.88
    omach
    0.80
     city
    0.77
     suburb
    0.77
    itals
    0.74
    orpor
    0.72
     cities
    0.71
     punishment
    0.71
    ashtra
    0.68
     Citiz
    0.67
    Act Density 0.017%

    No Known Activations