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

    Neuronpedia

    INDEX
    Explanations

    quantifiers and specific types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.21
     and
    0.15
     ,
    0.15
    nThe
    0.14
     being
    0.14
     exceeding
    0.14
     allowing
    0.14
     where
    0.14
    ↵↵
    0.13
     which
    0.13
    POSITIVE LOGITS
     gewisse
    0.20
     bestimmte
    0.19
     variety
    0.18
     handful
    0.18
     ciertos
    0.18
     bestimmten
    0.17
     few
    0.17
     myriad
    0.17
     ciertas
    0.17
     bepaalde
    0.16
    Act Density 0.527%

    No Known Activations