INDEX
Explanations
specific names, likely related to a person named Roh
references to the individual named Roh
New Auto-Interp
Negative Logits
iliary
-0.79
ciating
-0.79
lement
-0.72
hered
-0.70
rights
-0.68
DCS
-0.68
Panther
-0.66
offic
-0.66
eval
-0.65
ships
-0.64
POSITIVE LOGITS
Roh
1.36
sten
0.82
Sod
0.75
rer
0.74
atche
0.71
arty
0.70
tis
0.70
kj
0.70
oh
0.70
owsky
0.69
Activations Density 0.008%