INDEX
Explanations
occurrences of tags and related terms in the text
New Auto-Interp
Negative Logits
Мексичка
-1.10
itſelf
-0.95
Etr
-0.92
resourceCulture
-0.90
NUMX
-0.89
princesses
-0.89
featureID
-0.88
Савезне
-0.88
mijne
-0.87
violins
-0.87
POSITIVE LOGITS
tag
0.68
or
0.56
the
0.52
to
0.51
combined
0.49
ViewGroup
0.49
(
0.49
and
0.49
these
0.48
through
0.48
Activations Density 0.144%