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

    Neuronpedia

    INDEX
    Explanations

    completely erasing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    有时候
    0.51
    What
    0.48
    When
    0.46
    鍑
    0.46
    Why
    0.45
    Sometimes
    0.45
    Finding
    0.45
    原来
    0.45
     Something
    0.45
    为什么
    0.45
    POSITIVE LOGITS
    wohner
    0.61
     durchgeführt
    0.52
     Lietuvos
    0.52
     Ghana
    0.51
     adduced
    0.51
    }_{+}
    0.51
     siswa
    0.50
     reactors
    0.49
     歴史
    0.48
     heaters
    0.47
    Act Density 0.002%

    No Known Activations