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

    Neuronpedia

    INDEX
    Explanations

    AI, 中国, コード, 海, PC

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    沨
    0.77
    نګه
    0.75
    स्करण
    0.72
    मेरी
    0.72
    मॉडल
    0.72
    嫵
    0.70
     )।
    0.70
    डिंग
    0.69
     tcpHeader
    0.68
    lardı
    0.68
    POSITIVE LOGITS
    を
    1.48
    の
    1.44
    に
    1.28
    が
    1.24
    で
    1.09
    ・
    1.08
    と
    1.07
    을
    1.07
    、
    1.05
    의
    1.05
    Act Density 0.005%

    No Known Activations