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

    Neuronpedia

    INDEX
    Explanations

    Об

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     thrott
    -0.09
    -existing
    -0.08
     ಶ
    -0.07
    .*↵↵
    -0.07
    Thickness
    -0.07
     prevailing
    -0.07
    th
    -0.07
    aj
    -0.07
     conducive
    -0.07
    /th
    -0.07
    POSITIVE LOGITS
    сиа
    0.08
    foundation
    0.08
    」で
    0.08
     loisirs
    0.08
     robin
    0.08
    న్నారు
    0.08
     bagaimana
    0.08
     monumental
    0.07
     Joker
    0.07
    (array
    0.07
    Act Density 0.001%

    No Known Activations