INDEX
Explanations
phrases that indicate familial relationships, particularly focusing on the number of children or family members
New Auto-Interp
Negative Logits
sti
-0.19
loo
-0.16
Paginator
-0.15
ios
-0.15
idine
-0.14
io
-0.14
mate
-0.14
andr
-0.14
ins
-0.14
ihan
-0.14
POSITIVE LOGITS
drag
0.17
invention
0.16
slain
0.16
modern
0.15
Kauf
0.14
Modern
0.14
Dragons
0.14
olini
0.14
igers
0.14
cord
0.14
Activations Density 0.032%