INDEX
Explanations
references to specific individuals named Levy or Levine
New Auto-Interp
Negative Logits
ezvous
-0.84
emic
-0.83
ciating
-0.82
ificant
-0.82
colo
-0.81
astery
-0.79
ework
-0.79
functional
-0.78
psons
-0.78
eworld
-0.77
POSITIVE LOGITS
Levy
0.86
Feldman
0.77
Random
0.76
Hots
0.76
Horowitz
0.72
Restaur
0.72
Katz
0.72
Keys
0.68
Bros
0.67
Oo
0.65
Activations Density 0.007%