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

    Neuronpedia

    INDEX
    Explanations

    common and standard

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    原本
    1.01
     sebenarnya
    0.91
    ar
    0.89
     specifically
    0.86
    상품
    0.86
    기
    0.86
    entire
    0.85
    potentially
    0.85
    Exclusive
    0.84
    Entire
    0.84
    POSITIVE LOGITS
    做法
    1.42
     tropes
    1.41
     occurrence
    1.38
     decency
    1.33
     fare
    1.23
    情况下
    1.23
    isation
    1.19
     wisdom
    1.18
     approach
    1.17
    wisdom
    1.16
    Act Density 0.765%

    No Known Activations