INDEX
Explanations
mentions of the term "Duke"
references to the term "Duke."
New Auto-Interp
Negative Logits
ional
-0.80
heny
-0.68
telling
-0.64
deduction
-0.62
aging
-0.62
sbm
-0.61
atically
-0.61
Piercing
-0.60
ãģĻ
-0.60
served
-0.60
POSITIVE LOGITS
hyde
0.89
halla
0.82
lac
0.81
University
0.77
istry
0.74
aimon
0.72
ball
0.71
Fly
0.71
ongyang
0.69
Chapel
0.69
Activations Density 0.045%