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

    Neuronpedia

    INDEX
    Explanations

    method calls starting with get

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (
    0.97
    (“
    0.94
    (+
    0.87
    idla
    0.86
    gebaut
    0.86
    不定
    0.83
     Сури
    0.83
     ఉంది
    0.80
    geht
    0.80
    一提
    0.78
    POSITIVE LOGITS
    า
    1.52
     of
    1.43
    is
    1.25
    ’
    1.23
     and
    1.21
    म
    1.20
     on
    1.17
    of
    1.16
    л
    1.13
    ن
    1.07
    Act Density 0.106%

    No Known Activations