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

    Neuronpedia

    INDEX
    Explanations

    defining functions and variables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    +)
    0.43
    inue
    0.39
     протяжении
    0.39
     ++)
    0.39
     voire
    0.38
    +)$
    0.37
     ""))
    0.37
    +
    0.37
    上的
    0.36
    甚至
    0.36
    POSITIVE LOGITS
    我们要
    0.46
     responsible
    0.44
    我們要
    0.44
     governing
    0.43
     specifying
    0.43
     specifies
    0.42
     our
    0.42
     indica
    0.42
     Specifies
    0.40
    Our
    0.40
    Act Density 0.228%

    No Known Activations