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

    Neuronpedia

    INDEX
    Explanations

    sudden transitions or events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    contains
    0.93
     yra
    0.90
     imaju
    0.90
     обеспечивает
    0.89
     является
    0.85
     содержит
    0.85
     являются
    0.85
     pueden
    0.84
     peuvent
    0.84
     має
    0.83
    POSITIVE LOGITS
     suddenly
    1.29
     Suddenly
    1.28
     abruptly
    1.20
     Somehow
    1.10
     unexpectedly
    1.10
    Suddenly
    1.07
     miraculously
    1.06
     plötzlich
    1.06
     instead
    1.06
     after
    1.05
    Act Density 0.067%

    No Known Activations