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

    Neuronpedia

    INDEX
    Explanations

    portrays constant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ة
    2.48
    ing
    2.47
    ၎င်း
    2.27
    britannien
    2.00
    ている
    1.99
    lar
    1.94
    ————
    1.91
    ků
    1.91
    ه‌
    1.88
    ların
    1.88
    POSITIVE LOGITS
    те
    2.64
    т
    2.39
    𝘴
    2.36
    ا
    2.33
    𝙚
    2.19
    й
    2.08
    ಲ್ಲಿ
    2.05
    𝙖
    2.05
    н
    2.03
    ET
    2.02
    Act Density 0.210%

    No Known Activations