INDEX
Negative Logits
vtk
-0.08
described
-0.08
images
-0.07
eddi
-0.07
repel
-0.07
sailing
-0.07
blag
-0.07
tweeted
-0.07
definida
-0.07
Img
-0.07
POSITIVE LOGITS
dismant
0.13
拆
0.12
demolition
0.12
guts
0.10
destruct
0.10
ripping
0.10
tearing
0.09
析
0.09
_extract
0.09
الكهرب
0.09
Activations Density 0.085%