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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the history or track record of individuals or entities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ishable
    -0.74
    Flow
    -0.71
    wolves
    -0.70
    Sport
    -0.69
    ateurs
    -0.68
    ij士
    -0.66
    uri
    -0.65
    Stars
    -0.64
    idden
    -0.64
    ishers
    -0.63
    POSITIVE LOGITS
     dating
    0.93
     spanning
    0.82
     revolving
    0.81
     associ
    0.77
     of
    0.74
     dealing
    0.74
     supporting
    0.74
     favoring
    0.73
     advocating
    0.73
     exposing
    0.73
    Act Density 0.053%

    No Known Activations