INDEX
Explanations
proper nouns, particularly names like "Kelly."
mentions of the name "Kelly."
New Auto-Interp
Negative Logits
xual
-0.77
omething
-0.74
glim
-0.70
eers
-0.68
raints
-0.68
correct
-0.66
ensical
-0.66
nces
-0.66
meaning
-0.65
unres
-0.65
POSITIVE LOGITS
Kelly
0.97
Slater
0.95
Fitzpatrick
0.81
Anne
0.80
Clarkson
0.79
Simpson
0.78
sey
0.77
Sue
0.76
Saunders
0.76
lee
0.75
Activations Density 0.008%