INDEX
Explanations
cooking-related actions and instructions
New Auto-Interp
Negative Logits
Tank
-0.17
tanks
-0.15
Fro
-0.15
azo
-0.14
til
-0.14
inou
-0.14
Ivan
-0.14
mu
-0.14
pole
-0.14
oust
-0.14
POSITIVE LOGITS
bake
0.27
Bake
0.26
baking
0.24
oven
0.20
baked
0.18
bak
0.17
475
0.16
baker
0.16
sheet
0.16
дÑĥÑħов
0.15
Activations Density 0.031%