INDEX
Explanations
proper names, especially the name "Felix"
mention of specific individuals named Felix and Fidel
New Auto-Interp
Negative Logits
eds
-0.75
CBC
-0.72
earable
-0.70
ebook
-0.70
ally
-0.70
andals
-0.70
atchewan
-0.69
ript
-0.68
arer
-0.67
our
-0.67
POSITIVE LOGITS
Felix
1.15
Salmon
0.82
odox
0.80
paio
0.78
Hernandez
0.78
etheless
0.76
theless
0.75
entanyl
0.74
ministic
0.70
Fel
0.69
Activations Density 0.012%