INDEX
Explanations
references to food-related items and vegan cuisine
New Auto-Interp
Negative Logits
rescia
-0.15
zk
-0.15
Sugar
-0.14
arken
-0.14
ossier
-0.14
ucci
-0.14
ffa
-0.14
otel
-0.14
Sugar
-0.14
мали
-0.14
POSITIVE LOGITS
Impossible
0.30
Beyond
0.27
fake
0.26
tofu
0.26
plant
0.26
soy
0.24
Impossible
0.24
Beyond
0.23
Soy
0.23
Fake
0.23
Activations Density 0.010%