INDEX
Explanations
the presence of punctuation marks and periods in text
New Auto-Interp
Negative Logits
lund
-0.17
ople
-0.15
owie
-0.15
æ¡
-0.15
aska
-0.15
icom
-0.14
lide
-0.14
/of
-0.14
igue
-0.14
cox
-0.14
POSITIVE LOGITS
inus
0.16
ACL
0.16
UnderTest
0.16
csi
0.15
ffen
0.15
arity
0.15
pit
0.15
aad
0.15
elementType
0.14
acco
0.14
Activations Density 0.005%