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

    Neuronpedia

    INDEX
    Explanations

    phrases referring to different types or degrees of some quality or feature

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    aughs
    -0.98
     Lys
    -0.95
    oons
    -0.94
    sbm
    -0.91
     Emin
    -0.89
     Pigs
    -0.88
     Purs
    -0.88
     Files
    -0.87
    ires
    -0.87
    amples
    -0.86
    POSITIVE LOGITS
     semblance
    1.14
     intermediary
    1.06
     unspecified
    0.98
     whatsoever
    0.98
     resembling
    0.95
     halfway
    0.95
    insula
    0.94
     meaningful
    0.94
     entity
    0.94
     else
    0.92
    Act Density 0.888%

    No Known Activations