INDEX
Explanations
food-related words, especially related to fruits
names of foods and ingredients
New Auto-Interp
Negative Logits
etheless
-0.80
iably
-0.75
SPONSORED
-0.74
è¡
-0.72
reper
-0.72
aylor
-0.72
iscopal
-0.71
cient
-0.70
nont
-0.69
aboriginal
-0.69
POSITIVE LOGITS
Strip
0.97
Coat
0.96
Shop
0.96
Thing
0.93
Boot
0.93
finger
0.93
Sack
0.92
Powder
0.92
Meadow
0.91
Cake
0.91
Activations Density 0.068%