INDEX
Explanations
ethnicity and group identities
New Auto-Interp
Negative Logits
Sc
0.74
ம்
0.73
nse
0.71
n
0.70
et
0.67
المؤس
0.67
ాలు
0.67
IndexOf
0.66
cı
0.65
io
0.65
POSITIVE LOGITS
corridors
1.02
।
1.02
ethnicity
0.97
ж
0.97
ethnicities
0.94
ethnic
0.93
étn
0.90
។
0.88
。
0.82
nationalities
0.82
Activations Density 0.004%