INDEX
Negative Logits
small
-0.08
small
-0.07
714
-0.07
peas
-0.07
emerg
-0.07
364
-0.07
torque
-0.07
carrots
-0.07
passports
-0.07
abung
-0.07
POSITIVE LOGITS
divul
0.14
disclose
0.12
透露
0.12
divulgação
0.11
disclosed
0.11
공개
0.11
dévo
0.10
divulgar
0.10
prematurely
0.10
divulg
0.10
Activations Density 0.046%