INDEX
Explanations
references to the name Winston Churchill
New Auto-Interp
Negative Logits
adge
-0.15
et
-0.15
wat
-0.15
Abrams
-0.14
erie
-0.14
elit
-0.13
asje
-0.13
eten
-0.13
.aspx
-0.13
agne
-0.13
POSITIVE LOGITS
Churchill
0.28
ipeg
0.22
church
0.19
Winston
0.18
icism
0.17
peg
0.16
Salem
0.16
ilig
0.16
alem
0.15
Church
0.15
Activations Density 0.012%