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

    Neuronpedia

    INDEX
    Explanations

    fill-in-the-blanks or underscores

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    می
    1.01
    2
    0.94
    은
    0.90
    ная
    0.85
    ת
    0.84
    واصل
    0.82
    は
    0.81
    ני
    0.80
    UM
    0.79
    ı
    0.79
    POSITIVE LOGITS
     in
    1.26
    l
    1.11
    in
    1.02
    t
    0.96
    q
    0.96
    th
    0.92
    r
    0.89
    p
    0.86
     ionization
    0.84
    m
    0.83
    Act Density 0.006%

    No Known Activations