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

    Neuronpedia

    INDEX
    Explanations

    numbers and separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    mathbb
    0.63
    ઙ
    0.58
    grounds
    0.56
     Ω
    0.55
    コンピュー
    0.54
    Subsec
    0.54
    simp
    0.54
    ⁣
    0.53
     simp
    0.53
    ◌
    0.53
    POSITIVE LOGITS
    7
    0.66
    5
    0.65
    8
    0.64
    6
    0.63
    9
    0.62
    2
    0.61
    ७
    0.60
    ۷
    0.60
    <unused1683>
    0.60
    ২
    0.60
    Act Density 0.300%

    No Known Activations