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

    Neuronpedia

    INDEX
    Explanations

    function parameter types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    «.
    0.42
     সফদর
    0.41
    ूटूथ
    0.40
    ził
    0.39
    UREG
    0.39
     méth
    0.39
    лій
    0.39
    𒄩
    0.39
    *;
    0.38
    Methoxy
    0.38
    POSITIVE LOGITS
     $
    0.82
    array
    0.71
     array
    0.71
    &$
    0.70
     &$
    0.63
     $_
    0.61
     mixed
    0.60
     callable
    0.60
     string
    0.57
    ($
    0.55
    Act Density 0.005%

    No Known Activations