INDEX
Explanations
food-related concepts and experiences
New Auto-Interp
Negative Logits
azz
-0.16
chef
-0.16
achen
-0.16
munch
-0.16
pent
-0.15
itchen
-0.15
chefs
-0.15
_recipe
-0.15
.rx
-0.14
binh
-0.14
POSITIVE LOGITS
Spam
0.18
chips
0.18
flank
0.18
ice
0.18
Sprite
0.17
advoc
0.17
mac
0.17
,eg
0.17
Chips
0.17
rehe
0.17
Activations Density 0.993%