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

    Neuronpedia

    INDEX
    Explanations

    the word "conventional."

    oai_token-act-pair · gemini-2.0-flash

    conventional

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     
    -1.34
     first
    -1.07
    ,
    -1.06
     most
    -0.93
     in
    -0.91
    ↵
    -0.90
     (
    -0.89
     more
    -0.88
     very
    -0.85
     last
    -0.85
    POSITIVE LOGITS
     Efq
    1.84
     myſelf
    1.79
     itſelf
    1.63
     Theſe
    1.62
     Monfieur
    1.59
     ainfi
    1.57
     Jefus
    1.56
     himſelf
    1.53
     doubtnut
    1.52
     ―――――
    1.52
    Act Density 0.714%

    No Known Activations