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

    Neuronpedia

    INDEX
    Explanations

    matrix factorization, diagonalization, organization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ır
    1.14
    dı
    1.13
    ः
    1.10
    ка
    1.06
    ۲
    1.00
    de
    0.99
    ್ಯ
    0.99
    ного
    0.98
    di
    0.98
    mış
    0.92
    POSITIVE LOGITS
    s
    1.07
     Matrix
    0.99
    ط
    0.97
    ς
    0.95
    να
    0.90
    ח
    0.90
    ัง
    0.89
     It
    0.89
     matriz
    0.89
    for
    0.86
    Act Density 0.015%

    No Known Activations