INDEX
Explanations
proper nouns referring to people's names
variations of the word "ather."
New Auto-Interp
Negative Logits
aepernick
-0.70
unreal
-0.62
ako
-0.61
hao
-0.61
sure
-0.60
apo
-0.59
Bravo
-0.59
lez
-0.59
elman
-0.59
Played
-0.58
POSITIVE LOGITS
apy
1.04
ivities
1.03
sburg
1.03
ings
0.86
s
0.84
weights
0.82
aith
0.79
atform
0.78
athering
0.78
illac
0.78
Activations Density 0.026%