INDEX
Explanations
positive sentiments and expressions of approval
New Auto-Interp
Negative Logits
anja
-0.15
icorn
-0.15
inking
-0.14
emoc
-0.14
INK
-0.14
ink
-0.14
loven
-0.13
å²Ĺ
-0.12
acho
-0.12
akk
-0.12
POSITIVE LOGITS
eka
0.16
èĮĤ
0.14
ided
0.14
bson
0.14
Isles
0.14
tilt
0.14
idis
0.13
0.13
šku
0.13
Blasio
0.13
Activations Density 0.074%