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

    Neuronpedia

    INDEX
    Explanations

    careful plus an action or object

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ?
    1.17
    )
    1.09
    \
    1.06
    ),
    1.05
     ذریع
    1.05
    )،
    1.01
    </i>
    0.98
    !
    0.96
    )’
    0.95
    dır
    0.94
    POSITIVE LOGITS
    n
    1.76
    p
    1.20
    ag
    1.16
    id
    1.14
    a
    1.13
    ن
    1.12
    el
    1.11
    at
    1.09
    ol
    1.07
    ad
    1.06
    Act Density 0.015%

    No Known Activations