INDEX
Explanations
references to body parts and physical proximity
New Auto-Interp
Negative Logits
eyed
-0.08
pany
-0.07
roupe
-0.07
umbledore
-0.07
eut
-0.07
заб
-0.07
ÑĥÑĢн
-0.07
egov
-0.07
brero
-0.07
nghiá»ĩp
-0.06
POSITIVE LOGITS
sters
0.06
adel
0.06
olas
0.06
-mounted
0.06
enth
0.06
nic
0.05
fel
0.05
еÑģÑĤи
0.05
/trunk
0.05
alam
0.05
Activations Density 0.003%