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

    Neuronpedia

    INDEX
    Explanations

    numerical patterns and data representations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     sclerosis
    -0.83
     unden
    -0.79
    ioned
    -0.71
     Flavoring
    -0.68
    iculty
    -0.68
     jams
    -0.66
     reconc
    -0.66
     backlog
    -0.63
     enthusi
    -0.63
     concess
    -0.63
    POSITIVE LOGITS
    9
    0.80
    8
    0.78
    Series
    0.77
    0
    0.74
    antage
    0.73
    Chapter
    0.72
    19
    0.72
    6
    0.71
    apple
    0.69
    \"
    0.69
    Act Density 7.979%

    No Known Activations