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

    Neuronpedia

    INDEX
    Explanations

    say "news"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    떪
    -0.07
     perder
    -0.07
     Michigan
    -0.07
    微微
    -0.07
    (po
    -0.07
    太过
    -0.07
    braska
    -0.07
     speedy
    -0.07
    ellow
    -0.07
    茶叶
    -0.07
    POSITIVE LOGITS
    .on
    0.08
    şim
    0.07
     MATCH
    0.07
    .Action
    0.07
    衷
    0.07
     Ç
    0.07
    MATCH
    0.06
    ぴ
    0.06
    .sub
    0.06
     DN
    0.06
    Act Density 0.160%

    No Known Activations