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

    Neuronpedia

    INDEX
    Explanations

    introducing concepts with following description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    0.39
    いますが
    0.38
    ですが
    0.34
    糉
    0.34
    䄪
    0.33
    tra
    0.32
    اندا
    0.32
     నాలుగు
    0.32
     సంవత్సర
    0.32
    kD
    0.32
    POSITIVE LOGITS
     que
    0.50
     που
    0.44
     continúa
    0.43
     viene
    0.40
     είναι
    0.40
     nació
    0.40
     który
    0.39
    й
    0.39
    ﺮ
    0.38
    E
    0.38
    Act Density 0.632%

    No Known Activations