INDEX
Explanations
references to a specific sports team named the Washington Wizards
mentions of the "Wizards" team
New Auto-Interp
Negative Logits
ffer
-0.74
clair
-0.73
rontal
-0.72
lly
-0.70
vent
-0.67
equ
-0.67
shore
-0.67
conviction
-0.65
undai
-0.65
verts
-0.65
POSITIVE LOGITS
Wizards
1.02
sonian
0.87
DragonMagazine
0.83
nesday
0.83
atche
0.77
izards
0.77
pace
0.76
Hots
0.75
haus
0.75
hip
0.75
Activations Density 0.006%