INDEX
Explanations
instances of the substring "inh"
New Auto-Interp
Negative Logits
pauvreté
-0.45
MessageOf
-0.43
hâte
-0.42
Jerusalén
-0.41
fiabilidad
-0.40
berdayakan
-0.40
toalha
-0.40
vectorielle
-0.40
circledR
-0.40
Mackey
-0.39
POSITIVE LOGITS
inh
2.41
INH
1.73
nh
1.45
anh
1.41
inha
1.11
inho
1.10
Nh
0.94
NH
0.91
ynh
0.90
Anh
0.89
Activations Density 0.010%