INDEX
Explanations
justice, injustices, judiciary
New Auto-Interp
Negative Logits
itzer
-0.12
sWith
-0.09
culpa
-0.08
sm
-0.08
scratch
-0.08
/stdc
-0.08
thic
-0.08
Ñħови
-0.08
pra
-0.08
bis
-0.08
POSITIVE LOGITS
ifiable
0.16
ous
0.14
ifi
0.14
icial
0.12
ices
0.12
/right
0.12
ously
0.12
iciary
0.12
iros
0.12
distrib
0.11
Activations Density 0.021%