INDEX
Explanations
people's first names
the name "Michael" across various contexts
New Auto-Interp
Negative Logits
erent
-0.83
ancial
-0.78
ought
-0.74
hips
-0.71
aways
-0.71
aught
-0.71
ursed
-0.71
assium
-0.70
sylv
-0.69
nesses
-0.68
POSITIVE LOGITS
angelo
1.34
Bie
0.95
Vick
0.89
Angelo
0.87
Archangel
0.87
Hayden
0.86
Phelps
0.84
Bloomberg
0.84
Hastings
0.82
Flynn
0.77
Activations Density 0.032%