INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     fourn
    -0.07
    ünd
    -0.07
     deposit
    -0.07
     nhiệt
    -0.07
     QDialog
    -0.06
     coord
    -0.06
    -0.06
    ์บ
    -0.06
    construct
    -0.06
    _radius
    -0.06
    POSITIVE LOGITS
     slow
    0.12
     slowly
    0.11
     slower
    0.11
     Slow
    0.11
    Slow
    0.10
     fast
    0.08
    _slow
    0.08
    _bw
    0.07
    slow
    0.07
    -blind
    0.07
    Act Density 0.012%

    No Known Activations