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

    Neuronpedia

    INDEX
    Explanations

    list differences

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     can
    -0.08
     elementary
    -0.08
     is
    -0.08
     likelihood
    -0.07
     today
    -0.07
     would
    -0.07
     nonsense
    -0.07
     could
    -0.07
     major
    -0.07
     to
    -0.07
    POSITIVE LOGITS
     unmatched
    0.11
     uniques
    0.11
     únicos
    0.10
     exclusivos
    0.10
     einzigart
    0.10
     bhar
    0.09
     distinctly
    0.09
     exclusivo
    0.09
     unheard
    0.09
     uniek
    0.09
    Act Density 0.019%

    No Known Activations