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

    Neuronpedia

    INDEX
    Explanations

    Returns:

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mish
    0.43
     oscillating
    0.42
     scorched
    0.41
     %(
    0.40
     soup
    0.39
     mischief
    0.38
     extravagant
    0.38
    透过
    0.38
     subm
    0.37
     tot
    0.36
    POSITIVE LOGITS
     RETURNS
    0.42
     Yine
    0.42
    $&$-
    0.41
    raising
    0.40
    //@
    0.40
    CHINA
    0.40
    횟
    0.40
    #-
    0.39
    earic
    0.39
    中国的
    0.39
    Act Density 0.001%

    No Known Activations