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

    Neuronpedia

    INDEX
    Explanations

    namely, introducing specification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     and
    1.14
    ب
    1.04
    يه
    1.03
    大
    1.01
    艐
    0.94
    ليل
    0.93
    х
    0.91
     or
    0.91
    𝘭
    0.85
    一
    0.84
    POSITIVE LOGITS
    '
    1.47
    )
    1.42
    มัน
    0.92
    ),
    0.91
    )،
    0.91
    ↵
    0.90
    "
    0.90
    로
    0.90
    اشی
    0.89
    นี้
    0.89
    Act Density 0.003%

    No Known Activations