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

    Neuronpedia

    INDEX
    Explanations

    class name definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    apayati
    0.47
     når
    0.46
     goût
    0.44
    鐲
    0.42
    ätta
    0.40
    retching
    0.40
    ර්ග
    0.40
    westen
    0.39
    urname
    0.39
    indsay
    0.39
    POSITIVE LOGITS
     class
    0.72
    class
    0.63
     struct
    0.57
     класс
    0.57
    Base
    0.57
    Simple
    0.57
     Class
    0.56
     extends
    0.55
    Wrapper
    0.54
     subclass
    0.54
    Act Density 0.122%

    No Known Activations