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

    Neuronpedia

    INDEX
    Explanations

    references to a specific concept or idea in an academic or technical context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    Finish
    -0.63
     Dying
    -0.61
     Toast
    -0.59
     Gifts
    -0.58
     Codec
    -0.56
    Exit
    -0.55
     Deaths
    -0.55
    transfer
    -0.55
    76561
    -0.54
     Cowboy
    -0.54
    POSITIVE LOGITS
     beh
    1.26
     seems
    1.14
    unes
    1.08
    alian
    1.01
     appears
    1.01
     boils
    0.98
    iner
    0.96
     transpired
    0.96
     begs
    0.94
     emerges
    0.93
    Act Density 0.287%

    No Known Activations