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

    Neuronpedia

    INDEX
    Explanations

    reviewing documents, data, or types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ța
    -1.22
    ție
    -1.16
    ácia
    -1.13
    ведений
    -1.10
    Hva
    -1.10
    ATIVA
    -1.10
     καλύτε
    -1.09
    lawesi
    -1.09
     Allociné
    -1.09
    )',
    -1.07
    POSITIVE LOGITS
     -
    1.28
     "
    1.22
     '
    1.20
    </em>
    1.19
     $
    1.18
     במה
    1.07
     is
    1.06
    눙
    1.05
    </u>
    1.05
    besides
    1.03
    Act Density 0.012%

    No Known Activations