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

    Neuronpedia

    INDEX
    Explanations

    Twitter usernames and handles

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    arial
    -0.98
    uate
    -0.89
     afore
    -0.86
     gratification
    -0.85
     succeeding
    -0.84
     unpre
    -0.81
    osal
    -0.80
    EStream
    -0.78
    henko
    -0.77
     Entered
    -0.76
    POSITIVE LOGITS
    ITCH
    1.49
    INGS
    1.48
    OW
    1.45
    ITNESS
    1.43
    atts
    1.41
    edge
    1.37
    OOD
    1.37
    atson
    1.35
    aver
    1.35
    ALK
    1.34
    Act Density 1.684%

    No Known Activations