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

    Neuronpedia

    INDEX
    Explanations

    optimizing weights and MPC

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    的
    1.20
    2
    1.20
    s
    0.98
    <0x0D>
    0.95
    sp
    0.86
    س
    0.85
    एस
    0.83
    У
    0.82
    Α
    0.81
     breakpoint
    0.80
    POSITIVE LOGITS
    ла
    1.27
    ك
    1.26
    ке
    1.01
    лера
    1.00
    দ
    1.00
    ках
    0.96
    يش
    0.93
    ม
    0.93
    кам
    0.92
    тую
    0.92
    Act Density 0.000%

    No Known Activations