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

    Neuronpedia

    INDEX
    Explanations

    Titles and subtitles

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     previamente
    -0.10
     beforehand
    -0.09
    conditionally
    -0.08
    credito
    -0.08
     anteriormente
    -0.08
    -att
    -0.08
     заранее
    -0.08
     eerdere
    -0.08
     predicament
    -0.07
     tegelijk
    -0.07
    POSITIVE LOGITS
    成为
    0.09
    들은
    0.08
     발전
    0.08
     cole
    0.08
    时期
    0.08
     Addison
    0.08
    398
    0.08
     совет
    0.08
     WWII
    0.08
     확대
    0.08
    Act Density 0.064%

    No Known Activations