INDEX
Explanations
instances of the imperative form of verbs
New Auto-Interp
Negative Logits
ears
-0.17
heimer
-0.15
nection
-0.14
erial
-0.14
auth
-0.13
اث
-0.13
avors
-0.13
erah
-0.13
agher
-0.13
åįĩ
-0.13
POSITIVE LOGITS
Modes
0.16
adera
0.15
anza
0.15
ile
0.15
yun
0.15
hs
0.14
sum
0.14
kke
0.14
adil
0.14
ys
0.14
Activations Density 0.043%