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

    Neuronpedia

    INDEX
    Explanations

    Exploitative or explains

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    elled
    1.16
    é
    1.09
    ется
    1.03
    એ
    1.02
    １
    1.02
    ى
    1.02
    あ
    1.01
     powied
    0.99
    ae
    0.98
    le
    0.97
    POSITIVE LOGITS
    0
    1.71
    in
    1.29
    the
    1.29
    8
    1.06
    9
    1.05
    มูล
    1.02
    ים
    1.01
    ه
    0.98
    המ
    0.98
    ร
    0.94
    Act Density 0.683%

    No Known Activations