INDEX
Explanations
punctuation marks and certain punctuation patterns
New Auto-Interp
Negative Logits
DockStyle
-0.90
ViewImports
-0.88
-0.84
abbildung
-0.77
مرئيه
-0.77
LayoutConstraint
-0.66
valbard
-0.65
<bos>
-0.64
etna
-0.63
ſmall
-0.63
POSITIVE LOGITS
ErrIntOverflow
0.56
########.
0.53
getRule
0.51
للاسماء
0.50
PhysRev
0.50
canestro
0.50
ija
0.49
Sk
0.48
ఔ
0.48
{.0.48
Activations Density 0.507%