INDEX
Explanations
cooking-related instructions and ingredients used in various recipes
New Auto-Interp
Negative Logits
estate
-0.69
funer
-0.69
dfx
-0.67
ighth
-0.64
Labrador
-0.62
utsche
-0.61
anmar
-0.61
Gund
-0.60
landish
-0.59
Os
-0.59
POSITIVE LOGITS
Activate
0.71
(~
0.70
Slowly
0.68
Ingredients
0.68
preferably
0.68
overnight
0.68
Prepare
0.67
Remove
0.66
Bake
0.65
Prep
0.65
Activations Density 0.051%