INDEX
Explanations
words related to Eastern European names
New Auto-Interp
Negative Logits
decomp
-0.82
enegger
-0.74
dispers
-0.68
fracture
-0.68
terday
-0.66
ifications
-0.64
cavity
-0.63
mosaic
-0.63
ifying
-0.63
killer
-0.63
POSITIVE LOGITS
ĸ
1.51
Ĩ
1.49
Ħ
1.46
Ķ
1.42
¹
1.41
ĺ
1.41
·
1.40
ľ
1.40
ı
1.38
ħ
1.37
Activations Density 0.009%