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

    Neuronpedia

    INDEX
    Explanations

    numeric quantities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    sonian
    -0.86
    20439
    -0.83
    nen
    -0.74
     Flesh
    -0.74
    enegger
    -0.73
    nee
    -0.73
     Sporting
    -0.68
     Bland
    -0.68
    andowski
    -0.67
     Wine
    -0.67
    POSITIVE LOGITS
     fluctuations
    1.11
     discrepancies
    1.09
     adjustments
    1.07
     adjustment
    1.05
     determination
    1.02
     disparities
    1.02
     regulation
    1.00
     extremes
    0.97
     calcul
    0.97
     measurement
    0.97
    Act Density 0.329%

    No Known Activations