INDEX
Explanations
proper nouns associated with notable individuals
New Auto-Interp
Negative Logits
worst
-0.16
@nate
-0.16
qt
-0.14
dum
-0.14
arnation
-0.13
shade
-0.13
boss
-0.13
endency
-0.13
adelphia
-0.13
bal
-0.12
POSITIVE LOGITS
ý
0.16
uko
0.15
frank
0.14
pon
0.14
ancode
0.14
heter
0.14
nid
0.14
*)((
0.14
ัà¸Ħร
0.14
/fr
0.13
Activations Density 0.023%