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

    Neuronpedia

    INDEX
    Explanations

    words related to politics, policies, and societal issues

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    VIDIA
    -0.59
    WARD
    -0.58
     juven
    -0.57
     confir
    -0.56
     pestic
    -0.55
    enegger
    -0.54
     horm
    -0.52
     elaborated
    -0.52
     tiss
    -0.52
    thumbnails
    -0.52
    POSITIVE LOGITS
     shine
    0.67
     afloat
    0.66
     obsolete
    0.60
     accountable
    0.59
     aside
    0.59
     onto
    0.58
     into
    0.57
     away
    0.55
     mates
    0.55
    neys
    0.55
    Act Density 5.752%

    No Known Activations