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

    Neuronpedia

    INDEX
    Explanations

    directory where

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     முடியாது
    0.75
    することも
    0.74
     who
    0.70
     silo
    0.69
    了出来
    0.68
     întâm
    0.67
     surgi
    0.66
     losers
    0.66
     silenz
    0.66
     omn
    0.66
    POSITIVE LOGITS
     where
    3.02
    where
    3.00
    Where
    2.80
     Where
    2.69
     donde
    2.62
    에서
    2.60
    에서의
    2.35
     où
    2.35
     WHERE
    2.32
    에서는
    2.29
    Act Density 0.045%

    No Known Activations