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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a range or interval of values

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    OGR
    -0.82
    rav
    -0.81
    gow
    -0.75
    jew
    -0.72
    awaru
    -0.68
    insula
    -0.68
    justice
    -0.68
    pot
    -0.68
    eto
    -0.68
    rats
    -0.66
    POSITIVE LOGITS
     halves
    0.91
     sexes
    0.84
     genders
    0.80
     January
    0.80
     1910
    0.77
     1932
    0.75
     1945
    0.74
     midnight
    0.74
     1975
    0.74
     1981
    0.74
    Act Density 0.038%

    No Known Activations