INDEX
Explanations
references to "soul" and related concepts
New Auto-Interp
Negative Logits
oons
-0.17
antino
-0.15
aldi
-0.15
ellas
-0.14
oon
-0.14
etro
-0.14
Bed
-0.14
egas
-0.14
ylon
-0.14
fsp
-0.14
POSITIVE LOGITS
stice
0.24
ution
0.22
utions
0.21
ful
0.21
UTION
0.21
mate
0.20
fulness
0.20
less
0.18
FUL
0.18
/body
0.18
Activations Density 0.013%