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

    Neuronpedia

    INDEX
    Explanations

    [mention specific skills]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ب
    2.27
    ल
    2.03
    y
    1.90
    на
    1.77
    es
    1.63
    ル
    1.61
    ي
    1.56
    ний
    1.50
    ти
    1.49
    ა
    1.48
    POSITIVE LOGITS
    Архівовано
    1.23
    \}
    1.22
     muque
    1.20
    ٣
    1.15
    ].
    1.14
    ๗
    1.14
    ++]
    1.12
    ສຸດ
    1.12
    hiddenMap
    1.11
    ⎆
    1.11
    Act Density 0.068%

    No Known Activations