INDEX
Explanations
mentions of geographical locations, particularly Idaho and Spokane
New Auto-Interp
Negative Logits
ëij¥
-0.15
cona
-0.15
lev
-0.14
Crus
-0.14
©
-0.14
iyim
-0.14
odst
-0.14
hy
-0.14
orno
-0.14
ilet
-0.14
POSITIVE LOGITS
enco
0.17
greg
0.16
ella
0.15
beck
0.15
ebe
0.15
peer
0.15
Peer
0.15
Becky
0.15
OMETRY
0.14
ola
0.14
Activations Density 0.001%