INDEX
Explanations
instances of ideology and discussions about belief systems
New Auto-Interp
Negative Logits
democrat
-0.17
ekim
-0.15
Wolverine
-0.14
engin
-0.14
asure
-0.14
reed
-0.14
okino
-0.14
EdgeInsets
-0.13
atty
-0.13
828
-0.13
POSITIVE LOGITS
ne
0.20
pure
0.17
post
0.17
neo
0.16
Moderate
0.16
v
0.16
util
0.15
versions
0.15
anti
0.15
orth
0.15
Activations Density 0.314%