INDEX
Explanations
phrases related to religious messages and teachings
New Auto-Interp
Negative Logits
omi
-0.18
iev
-0.15
.ms
-0.14
_$_
-0.14
omin
-0.14
_Internal
-0.14
Vide
-0.14
ainty
-0.14
_Cell
-0.13
sov
-0.13
POSITIVE LOGITS
Walk
0.17
Anth
0.17
çĽ
0.16
usic
0.16
Walk
0.15
Anth
0.15
Matters
0.15
vel
0.15
Anthrop
0.14
anth
0.14
Activations Density 0.017%