INDEX
Explanations
possessive forms, particularly related to people's names and their attributes
New Auto-Interp
Negative Logits
ispers
-0.17
igor
-0.16
haft
-0.15
ongs
-0.15
actionTypes
-0.15
ets
-0.14
енÑĮ
-0.14
bellion
-0.14
ings
-0.14
iot
-0.14
POSITIVE LOGITS
Ñĩе
0.18
LENG
0.15
òa
0.14
lege
0.14
Mills
0.14
ازد
0.14
achu
0.14
\Annotation
0.14
msp
0.13
ikip
0.13
Activations Density 0.042%