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

    Neuronpedia

    INDEX
    Explanations

    Hebrew or Greek letters and punctuation symbols used in a specific context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    EStreamFrame
    -0.83
    ournal
    -0.81
    Journal
    -0.81
    aca
    -0.80
    agall
    -0.80
    amar
    -0.78
    inka
    -0.75
    umo
    -0.75
    ollar
    -0.71
    ashtra
    -0.70
    POSITIVE LOGITS
    á½
    1.24
    ο
    1.04
    ι
    1.02
    0.98
    ν
    0.91
    α
    0.90
    ÏĤ
    0.89
    ÏĦ
    0.88
    κ
    0.86
    ¿
    0.82
    Act Density 0.116%

    No Known Activations