INDEX
Explanations
references to beer and brewing
New Auto-Interp
Negative Logits
ors
-0.21
ure
-0.18
ing
-0.17
en
-0.16
ents
-0.16
.
-0.16
y
-0.16
uren
-0.16
ting
-0.15
eil
-0.15
POSITIVE LOGITS
zeug
0.18
gratuiti
0.17
кеÑĤ
0.16
ijken
0.16
asmus
0.16
ifice
0.15
eries
0.15
allel
0.15
coma
0.15
ضÛĮ
0.14
Activations Density 0.023%