INDEX
    Explanations

    This neuron detects HTML/JSX form markup and input-related attributes (i.e., code tokens tied to form fields and their properties).

    New Auto-Interp
    Negative Logits
     인정
    0.50
    形で
    0.45
     vorgesehen
    0.45
    世紀
    0.44
     Meteorology
    0.44
    0.43
    ResourceManager
    0.43
    <0x06>
    0.41
     हैद
    0.41
    പ്പിക്കുന്ന
    0.40
    POSITIVE LOGITS
     flapping
    0.43
    er
    0.43
    ak
    0.42
     fluid
    0.42
    Sa
    0.42
     flipped
    0.42
     u
    0.41
     above
    0.40
     org
    0.40
    上面
    0.39
    Act Density 0.216%

    No Known Activations