INDEX
Explanations
food items and their preparation methods
New Auto-Interp
Negative Logits
cond
-0.19
.scalablytyped
-0.19
Cond
-0.17
Sauce
-0.16
cade
-0.16
Sau
-0.16
ape
-0.15
lage
-0.15
Sau
-0.15
pickle
-0.15
POSITIVE LOGITS
til
0.21
ter
0.20
rig
0.19
salmon
0.18
Shepard
0.18
Til
0.18
Ter
0.18
Lingu
0.18
pest
0.17
Rig
0.17
Activations Density 0.122%