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

    Neuronpedia

    INDEX
    Explanations

    say "capitalized word"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ίες
    -0.07
     NHL
    -0.07
    比
    -0.06
    _PLL
    -0.06
     Ratio
    -0.06
     ers
    -0.06
    iscard
    -0.06
    τηση
    -0.06
     bean
    -0.06
     каждый
    -0.06
    POSITIVE LOGITS
    ी।
    0.06
     Coming
    0.06
    .createClass
    0.06
    Browsable
    0.06
     Judaism
    0.06
    ()↵
    0.06
     فع
    0.06
    اضي
    0.06
     fiz
    0.06
    	↵↵↵
    0.06
    Act Density 0.017%

    No Known Activations