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

    Neuronpedia

    INDEX
    Explanations

    describing representation or function

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    सा
    0.47
    х
    0.46
    d
    0.45
    t
    0.44
     (
    0.44
    greedy
    0.43
    udere
    0.43
    raded
    0.42
    to
    0.42
    伊
    0.42
    POSITIVE LOGITS
     [,
    0.49
     governs
    0.49
     formulario
    0.47
     escánd
    0.46
     likened
    0.46
    厮
    0.45
    .[[
    0.45
     eure
    0.44
     sever
    0.44
     personalised
    0.44
    Act Density 0.008%

    No Known Activations