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

    Neuronpedia

    INDEX
    Explanations

    commas in data

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    然后
    -0.08
     so
    -0.07
    仰
    -0.07
     stupid
    -0.07
    我认为
    -0.07
    ▂
    -0.07
    Saturday
    -0.07
    rud
    -0.07
    一共
    -0.07
     LOG
    -0.07
    POSITIVE LOGITS
     acomp
    0.07
    uge
    0.07
     Alias
    0.07
    routine
    0.07
     Natasha
    0.06
     Powder
    0.06
    Multiplier
    0.06
    딧
    0.06
    🎶
    0.06
     [%
    0.06
    Act Density 0.004%

    No Known Activations