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

    Neuronpedia

    INDEX
    Explanations

    say "unw"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Giving
    -0.07
    Figure
    -0.07
    .plus
    -0.07
     federally
    -0.07
     '../../
    -0.07
     royalty
    -0.06
    from
    -0.06
    ---------
    -0.06
     Directors
    -0.06
    Craft
    -0.06
    POSITIVE LOGITS
     šk
    0.07
    _ER
    0.06
     парт
    0.06
    ETwitter
    0.06
    ate
    0.06
    .rl
    0.06
     zvý
    0.06
    usaha
    0.06
     kå
    0.06
     Bh
    0.06
    Act Density 0.001%

    No Known Activations