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

    Neuronpedia

    INDEX
    Explanations

    stages in processes and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    t
    1.59
     alimentare
    1.13
    ്ര
    1.13
    tól
    1.09
     angered
    1.08
    recated
    1.01
    ้
    1.01
    tura
    1.00
    ),
    0.99
    c
    0.98
    POSITIVE LOGITS
    이었
    1.30
    ق
    1.29
    ف
    1.18
    A
    1.15
    ↵↵
    1.14
    بود
    1.13
    с
    1.11
    น
    1.09
     pass
    1.08
    ючи
    1.06
    Act Density 0.023%

    No Known Activations