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

    Neuronpedia

    INDEX
    Explanations

    email addresses listed for contact purposes

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     tracts
    -0.59
     Elements
    -0.58
     Starr
    -0.58
     fragments
    -0.57
     Souls
    -0.57
    ãĤ¡
    -0.57
     Heist
    -0.56
     Friendship
    -0.55
     Pieces
    -0.55
     distances
    -0.54
    POSITIVE LOGITS
    ]+
    0.68
    &&
    0.65
     ||
    0.65
     UNCLASSIFIED
    0.64
    peror
    0.64
    azeera
    0.64
    amiya
    0.63
    edin
    0.62
    ãĥĦ
    0.62
    }
    0.62
    Act Density 0.038%

    No Known Activations