INDEX
Explanations
instances of the article "a"
New Auto-Interp
Negative Logits
zell
-0.16
Äįem
-0.16
ir
-0.15
utar
-0.15
pis
-0.15
uner
-0.14
brands
-0.14
bras
-0.14
ates
-0.14
chner
-0.14
POSITIVE LOGITS
float
0.21
hd
0.18
mile
0.17
propos
0.17
sis
0.15
elian
0.15
urally
0.15
/or
0.15
Ch
0.14
nn
0.14
Activations Density 0.074%