INDEX
Explanations
references to local news and community events
New Auto-Interp
Negative Logits
comed
-0.18
ogan
-0.17
cular
-0.16
Seen
-0.14
roperties
-0.14
imuth
-0.14
okus
-0.13
ozo
-0.13
neut
-0.13
udiant
-0.13
POSITIVE LOGITS
¥
0.15
yas
0.14
IED
0.14
ertino
0.13
polish
0.13
Reid
0.13
³
0.13
THR
0.13
prepared
0.13
dns
0.13
Activations Density 0.010%