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

    Neuronpedia

    INDEX
    Explanations

    Context, latest, well, little, see

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused503>
    1.05
    <unused284>
    0.92
    <unused729>
    0.92
    <unused2056>
    0.90
    <unused518>
    0.88
    <unused289>
    0.87
    <unused325>
    0.87
    <unused329>
    0.86
    <unused533>
    0.86
    <unused1803>
    0.86
    POSITIVE LOGITS
    ゙
    1.13
    ́
    0.91
    ̂
    0.89
    ౖ
    0.86
    igned
    0.83
    nbsp
    0.83
    ßt
    0.83
    opically
    0.83
    ে
    0.82
    ̣n
    0.82
    Act Density 0.119%

    No Known Activations