INDEX
Explanations
terms related to the Bush administration
references to the Bush administration
New Auto-Interp
Negative Logits
semble
-0.72
Qiao
-0.70
Norn
-0.66
Cth
-0.64
Uriel
-0.64
ymph
-0.63
lihood
-0.63
Harmony
-0.61
Interested
-0.60
feature
-0.60
POSITIVE LOGITS
nell
1.27
ido
1.08
master
0.97
lett
0.87
Bush
0.83
Hussein
0.80
ball
0.78
bour
0.78
Bush
0.77
band
0.76
Activations Density 0.016%