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

    Neuronpedia

    INDEX
    Explanations

    code documentation or formatting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ℝ
    1.51
     Sincerely
    1.47
     Oops
    1.45
     因此
    1.44
     ****",
    1.42
     Remarkably
    1.41
     ***",
    1.39
     [{\
    1.39
     SmackDown
    1.38
     Aquí
    1.36
    POSITIVE LOGITS
    (
    1.64
    •
    1.48
    $
    1.48
    $\
    1.43
    https
    1.42
    “
    1.40
    http
    1.35
    [
    1.34
    #
    1.32
    «
    1.30
    Act Density 0.052%

    No Known Activations