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

    Neuronpedia

    INDEX
    Explanations

    code snippets starting with specific keywords and involving mathematical or programming concepts

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    mberg
    -0.86
    zig
    -0.83
    posts
    -0.82
    ppo
    -0.82
    wig
    -0.82
    xual
    -0.82
    Downloadha
    -0.81
     resil
    -0.79
    gren
    -0.79
    skirts
    -0.78
    POSITIVE LOGITS
    expr
    1.19
     std
    0.80
     char
    0.79
    ricting
    0.77
     inline
    0.72
    rain
    0.72
    antine
    0.71
     uint
    0.71
    catentry
    0.69
     float
    0.68
    Act Density 6.493%

    No Known Activations