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

    Neuronpedia

    INDEX
    Explanations

    Russian prefixes followed by words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     s
    0.44
     c
    0.42
     l
    0.42
     type
    0.41
     en
    0.41
     be
    0.40
     bero
    0.38
     way
    0.38
     non
    0.37
     sur
    0.37
    POSITIVE LOGITS
    Н
    0.61
    Cadastro
    0.61
    Б
    0.56
     ಮತ್ತು
    0.56
    ayvachi
    0.55
    да
    0.55
    м
    0.55
    ഓം
    0.54
     алгорит
    0.54
    ни
    0.54
    Act Density 0.138%

    No Known Activations