INDEX
Explanations
words related to steaming or cooking methods
New Auto-Interp
Negative Logits
f
-0.18
ogeneous
-0.17
ky
-0.17
itud
-0.15
ieg
-0.15
iej
-0.15
eff
-0.15
standing
-0.15
.dep
-0.14
bal
-0.14
POSITIVE LOGITS
aming
0.27
amed
0.24
aks
0.22
amily
0.19
eps
0.19
ste
0.19
amer
0.18
637
0.18
ers
0.17
edores
0.17
Activations Density 0.003%