INDEX
Explanations
mentions of food and drink-related terms
New Auto-Interp
Negative Logits
electronics
-0.17
ppy
-0.17
Crafts
-0.16
acrylic
-0.16
helicopt
-0.15
stainless
-0.15
woo
-0.15
polyester
-0.15
welded
-0.14
computer
-0.14
POSITIVE LOGITS
Vict
0.17
jap
0.17
india
0.17
SWG
0.17
surtout
0.15
uzey
0.15
hacks
0.15
japan
0.15
queer
0.15
<article
0.15
Activations Density 0.288%