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

    Neuronpedia

    INDEX
    Explanations

    Keeping threats away

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     درصد
    -0.07
     paddingBottom
    -0.07
     можно
    -0.07
    	address
    -0.06
     crime
    -0.06
     데이터
    -0.06
    -0.06
    vection
    -0.06
            ↵        ↵
    -0.06
     Properties
    -0.06
    POSITIVE LOGITS
    avra
    0.07
    PX
    0.07
     bỏ
    0.07
     appl
    0.06
    entifier
    0.06
    (co
    0.06
    quip
    0.06
     eliminating
    0.06
    λευ
    0.05
    öff
    0.05
    Act Density 0.023%

    No Known Activations