INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     phenomena
    -0.08
     partid
    -0.07
    Residence
    -0.07
     phenomenon
    -0.07
    kern
    -0.07
    Contains
    -0.07
    97
    -0.07
    loading
    -0.07
     To
    -0.07
     residence
    -0.07
    POSITIVE LOGITS
     Antique
    0.08
     осторож
    0.08
    rstrip
    0.08
     plain
    0.08
    /plain
    0.08
    _plain
    0.08
     ткани
    0.08
     ngon
    0.08
     skrif
    0.08
     mayo
    0.08
    Act Density 0.009%

    No Known Activations