INDEX
Explanations
words and phrases related to desserts and food descriptions
New Auto-Interp
Negative Logits
Garlic
-0.16
neck
-0.15
alion
-0.15
ÛĮØ´ÙĨ
-0.15
onion
-0.15
GRAY
-0.14
Onion
-0.14
inar
-0.14
garlic
-0.14
_wr
-0.14
POSITIVE LOGITS
cust
0.32
Cust
0.27
cust
0.27
masc
0.26
whipped
0.26
Masc
0.25
cream
0.25
chant
0.25
gel
0.23
stabilized
0.23
Activations Density 0.034%