INDEX
Explanations
references to birds or bird-related subjects
New Auto-Interp
Negative Logits
ucz
-0.15
ithe
-0.15
aub
-0.14
entes
-0.14
ocha
-0.14
dia
-0.13
ceb
-0.13
ei
-0.13
iren
-0.13
erule
-0.13
POSITIVE LOGITS
arium
0.19
ong
0.15
case
0.15
.ud
0.15
.sy
0.15
HELL
0.15
Excell
0.14
orr
0.14
odd
0.14
è¿Ķ
0.14
Activations Density 0.007%