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

    Neuronpedia

    INDEX
    Explanations

    word followed by parenthesis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ["
    1.63
     ['
    1.58
     [[
    1.33
     [['
    1.29
     [\
    1.23
     [{
    1.16
     [*
    1.15
     [[[
    1.10
     [`
    1.08
     [$
    1.07
    POSITIVE LOGITS
    (
    3.27
    '(
    1.80
    (.
    1.66
    }(
    1.65
    (-
    1.56
    (,
    1.56
    (...)
    1.51
    ($
    1.51
    $(
    1.49
    (...
    1.48
    Act Density 0.747%

    No Known Activations