INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     due
    0.71
     undes
    0.69
     Shot
    0.68
     quietly
    0.68
     overtaking
    0.67
     dev
    0.67
     Due
    0.67
     complete
    0.66
     weir
    0.66
     shot
    0.65
    POSITIVE LOGITS
    construction
    1.08
    constructing
    1.07
    construct
    1.06
    ceptive
    1.05
    compress
    1.05
     novo
    1.03
     facto
    1.03
    jeuner
    1.01
    jected
    0.96
    velopper
    0.95
    Act Density 0.027%

    No Known Activations