INDEX
Negative Logits
�
-0.08
Word
-0.08
kicking
-0.08
obe
-0.08
Dir
-0.08
şdir
-0.08
Given
-0.08
obar
-0.08
ုပ်
-0.07
ету
-0.07
POSITIVE LOGITS
dumped
0.09
_dump
0.08
SPEED
0.08
Harold
0.08
Dump
0.08
输
0.08
Accordion
0.08
Fernández
0.08
వ్య
0.08
circus
0.08
Activations Density 0.001%