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

    Neuronpedia

    INDEX
    Explanations

    technical descriptions and explanations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Ambro
    -1.02
    cu
    -0.97
     Horses
    -0.96
    aukee
    -0.93
    wr
    -0.92
    byss
    -0.91
     Pistons
    -0.91
     corrid
    -0.91
    enberg
    -0.90
    earable
    -0.88
    POSITIVE LOGITS
     wik
    0.88
    mentioned
    0.88
     reader
    0.88
     wont
    0.86
    ilst
    0.84
     paraph
    0.84
    iries
    0.84
     reviewer
    0.83
     aforementioned
    0.82
    RGB
    0.81
    Act Density 0.508%

    No Known Activations