INDEX
Explanations
intensifiers expressing a high degree of quality or feelings
New Auto-Interp
Negative Logits
vier
-0.16
ongo
-0.16
.scalablytyped
-0.16
ogui
-0.15
TEL
-0.14
frey
-0.14
vim
-0.14
oggler
-0.14
Král
-0.14
arus
-0.14
POSITIVE LOGITS
y
0.16
åĤ
0.15
gia
0.14
abay
0.14
mam
0.14
бÑĥÑĤ
0.13
ÙĦاØŃ
0.13
Jewish
0.13
Hampton
0.13
anto
0.13
Activations Density 0.005%