INDEX
Explanations
references to the moon and lunar cycles
New Auto-Interp
Negative Logits
ahl
-0.18
ooter
-0.18
-INF
-0.17
reed
-0.15
ouz
-0.14
jet
-0.14
gren
-0.14
737
-0.14
olver
-0.13
ascar
-0.13
POSITIVE LOGITS
dec
0.15
.uk
0.14
.named
0.14
raki
0.14
.builders
0.14
еÑĢб
0.14
ÏĢη
0.14
orta
0.14
.bi
0.13
decl
0.13
Activations Density 0.015%