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

    Neuronpedia

    INDEX
    Explanations

    phrases describing additional information or items in a list

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ople
    -0.80
    ione
    -0.75
    cko
    -0.75
    Ñı
    -0.74
    ãĤ¢ãĥ«
    -0.72
    ogyn
    -0.71
    èĢħ
    -0.69
    agog
    -0.69
    aucus
    -0.68
    onica
    -0.66
    POSITIVE LOGITS
     besides
    0.69
     also
    0.64
     other
    0.62
     anecd
    0.62
     includ
    0.61
     moreover
    0.60
    mentioned
    0.59
     secretaries
    0.59
    recent
    0.59
     ALSO
    0.58
    Act Density 0.160%

    No Known Activations