INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     underside
    -1.02
     inside
    -0.87
     upstairs
    -0.86
    内外
    -0.85
     beyond
    -0.80
     under
    -0.80
     речь
    -0.77
     behind
    -0.77
     across
    -0.75
     around
    -0.75
    POSITIVE LOGITS
     preuves
    0.97
    STEL
    0.90
     them
    0.88
     the
    0.88
     niitä
    0.85
    /////////
    0.84
    también
    0.82
    0.82
     arbeid
    0.82
    quê
    0.81
    Act Density 0.035%

    No Known Activations