INDEX
Explanations
terms related to wine, particularly grape varieties and flavor descriptions
New Auto-Interp
Negative Logits
.zh
-0.16
orning
-0.15
ipse
-0.14
esub
-0.14
erchant
-0.14
ues
-0.14
ousel
-0.14
baugh
-0.13
ramer
-0.13
orra
-0.13
POSITIVE LOGITS
unkt
0.16
robe
0.15
ë§Ľ
0.14
chai
0.14
Beaut
0.13
UPI
0.13
ãĥ¬ãĥ¼
0.13
kân
0.13
_NC
0.13
upert
0.13
Activations Density 0.010%