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

    Neuronpedia

    INDEX
    Explanations

    cost reduction and optimization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    िन
    1.02
    ה
    0.98
    ן
    0.97
    ణ
    0.93
    ا
    0.90
    لو
    0.89
    ς
    0.88
    ви
    0.84
    ા
    0.84
     Chowdh
    0.81
    POSITIVE LOGITS
    al
    1.52
     for
    1.48
    of
    1.38
    b
    1.32
    ig
    1.27
    ak
    1.27
    n
    1.23
    l
    1.19
    m
    1.16
    UN
    1.09
    Act Density 0.029%

    No Known Activations