INDEX
Explanations
descriptions of food items
instances of the word "delicious."
New Auto-Interp
Negative Logits
Downloadha
-0.79
vernment
-0.79
Physicians
-0.66
asketball
-0.65
mberg
-0.65
ourced
-0.64
vere
-0.63
yrim
-0.62
patient
-0.62
romy
-0.62
POSITIVE LOGITS
delicious
1.27
ness
1.08
nesses
1.05
Delicious
1.04
tasty
1.03
juicy
0.97
NESS
0.96
meals
0.96
pastry
0.94
nutritious
0.94
Activations Density 0.020%