INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     chert
    -0.68
     limestones
    -0.63
     despotism
    -0.62
     alberto
    -0.61
     sergio
    -0.61
     ecru
    -0.61
     sherds
    -0.60
     feldspar
    -0.58
     blackish
    -0.58
     tubercle
    -0.58
    POSITIVE LOGITS
     véri
    0.66
    expandindo
    0.56
     Sep
    0.54
     écri
    0.51
     exé
    0.51
     rédig
    0.51
     aimer
    0.51
     Apr
    0.49
     Published
    0.49
    Multivariate
    0.49
    Act Density 0.308%

    No Known Activations