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

    Neuronpedia

    INDEX
    Explanations

    references to special or distinctive entities or concepts

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    76561
    -1.12
     Twain
    -1.01
    ullivan
    -0.91
    anon
    -0.88
    Ķ
    -0.87
     Sweeney
    -0.86
     Ri
    -0.86
     Cah
    -0.84
    conom
    -0.84
    ·
    -0.84
    POSITIVE LOGITS
    ised
    1.47
    ties
    1.30
    izations
    1.21
    isations
    1.20
    ities
    1.19
    ized
    1.15
    arily
    1.11
    marine
    1.11
    atural
    1.10
    isable
    1.09
    Act Density 0.643%

    No Known Activations