INDEX
Explanations
references to political relationships and defense spending issues
New Auto-Interp
Negative Logits
Ñıз
-0.17
innoc
-0.15
irie
-0.14
rsa
-0.14
IED
-0.14
اØ
-0.14
rani
-0.14
isu
-0.14
Gst
-0.13
IFn
-0.13
POSITIVE LOGITS
lazy
0.29
Paras
0.25
paras
0.25
laz
0.24
lazy
0.22
loaf
0.22
parasites
0.20
contributor
0.20
-contrib
0.20
contributing
0.20
Activations Density 0.197%