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

    Neuronpedia

    INDEX
    Explanations

    phrases describing a significant or remarkable instance

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    malink
    -0.61
    semble
    -0.61
    ena
    -0.59
    tsy
    -0.58
     Defin
    -0.57
     Recomm
    -0.57
    aces
    -0.56
    ritz
    -0.56
    STDOUT
    -0.55
    azel
    -0.54
    POSITIVE LOGITS
     where
    1.88
     wherein
    1.75
    where
    1.75
     whereby
    1.48
    when
    1.34
     WHERE
    1.25
     when
    1.23
     Where
    1.15
     WHEN
    1.12
    Where
    1.09
    Act Density 0.757%

    No Known Activations