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

    Neuronpedia

    INDEX
    Explanations

    The lists are inconsistent. Cannot determine a pattern

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     οποίος
    1.13
    type
    1.09
    typed
    1.04
    tuple
    0.98
    top
    0.97
     to
    0.93
     or
    0.93
    tops
    0.93
    tos
    0.92
     elems
    0.90
    POSITIVE LOGITS
    ل
    1.90
    ن
    1.83
    ো
    1.54
    ב
    1.48
    ع
    1.43
    س
    1.29
    น
    1.25
    л
    1.24
    n
    1.23
    р
    1.23
    Act Density 1.316%

    No Known Activations