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

    Neuronpedia

    INDEX
    Explanations

    phrases that are emphasized with quotation marks

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Ross
    -0.78
     XL
    -0.77
     XI
    -0.77
     Isaac
    -0.76
     Mehran
    -0.76
     rall
    -0.75
     reel
    -0.75
     chant
    -0.74
     Creator
    -0.73
     Amar
    -0.73
    POSITIVE LOGITS
    significant
    1.78
    multiple
    1.68
    serious
    1.66
    many
    1.65
    very
    1.62
    extremely
    1.61
    sever
    1.60
    reasonable
    1.60
    certain
    1.60
    extra
    1.58
    Act Density 0.134%

    No Known Activations