INDEX
Explanations
references to the moon and its various qualities or representations
New Auto-Interp
Negative Logits
orns
-0.14
Omni
-0.14
543
-0.14
Ngh
-0.14
Avec
-0.14
inel
-0.13
-minded
-0.13
isible
-0.13
aniel
-0.13
321
-0.13
POSITIVE LOGITS
light
0.26
LIGHT
0.18
sext
0.16
beam
0.16
hack
0.15
rint
0.15
raki
0.15
iez
0.15
landing
0.14
licht
0.14
Activations Density 0.028%