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

    Neuronpedia

    INDEX
    Explanations

    comparisons or contrasts between different concepts

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that indicate evaluation or judgment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ¼ãĤ¦ãĤ¹
    -0.79
    leen
    -0.74
    awar
    -0.73
    terness
    -0.73
    ulner
    -0.68
    moil
    -0.68
    unks
    -0.67
    pez
    -0.67
    onen
    -0.67
    iple
    -0.66
    POSITIVE LOGITS
     actual
    1.31
     anything
    1.23
     genuine
    1.15
     substantive
    1.11
     conclusive
    1.10
     meaningful
    1.00
     outright
    0.98
     legitimate
    0.98
     definitive
    0.96
     serious
    0.95
    Act Density 0.271%

    No Known Activations