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

    Neuronpedia

    INDEX
    Explanations

    strings of capitalized letters

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    roxy
    -0.81
    director
    -0.75
    kus
    -0.73
    Offline
    -0.72
    tank
    -0.71
    ideos
    -0.71
    Kin
    -0.71
    Honest
    -0.70
    romeda
    -0.70
    artisan
    -0.68
    POSITIVE LOGITS
     prefix
    1.18
     alphabet
    1.17
     notation
    1.12
     suffix
    1.11
     denote
    1.09
     encoded
    1.03
     symbols
    1.03
     spelled
    1.00
     symbol
    1.00
     correspond
    0.96
    Act Density 0.174%

    No Known Activations