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

    Neuronpedia

    INDEX
    Explanations

    information related to political, legal, and governmental matters, including discussions about policies, decisions, and public opinions on various issues

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ishable
    -0.60
     jugg
    -0.57
     sequential
    -0.56
     mammoth
    -0.53
     puberty
    -0.53
     transact
    -0.53
    pei
    -0.52
     abyss
    -0.52
    ikuman
    -0.51
    illet
    -0.51
    POSITIVE LOGITS
     However
    1.01
     Instead
    0.99
     Specifically
    0.96
     Because
    0.91
     Nonetheless
    0.91
     Although
    0.91
     Therefore
    0.90
     Though
    0.90
     Moreover
    0.89
     Nevertheless
    0.89
    Act Density 10.092%

    No Known Activations