INDEX
Explanations
references to "House of" and similar phrases or titles
New Auto-Interp
Negative Logits
赤
-0.17
enez
-0.16
ilet
-0.16
resh
-0.15
rello
-0.15
zzo
-0.15
baru
-0.14
vier
-0.14
rame
-0.14
koa
-0.14
POSITIVE LOGITS
Representatives
0.21
House
0.20
house
0.17
representatives
0.16
Commons
0.16
oldem
0.15
/ws
0.15
_taken
0.15
cards
0.15
commons
0.15
Activations Density 0.014%