INDEX
Explanations
mentions of the word "Persian" in the text
references to Persian culture or history
New Auto-Interp
Negative Logits
UC
-0.71
FC
-0.71
Hut
-0.69
AW
-0.68
Gear
-0.68
VILLE
-0.67
vans
-0.66
ando
-0.66
jobs
-0.64
NC
-0.62
POSITIVE LOGITS
Persian
3.86
Persia
2.21
Arabian
1.87
Pers
1.81
Pers
1.71
Cyrus
1.64
Iranian
1.42
Zoro
1.41
Darius
1.41
Azerbai
1.36
Activations Density 0.027%