INDEX
Explanations
references to different levels of education, particularly high school
New Auto-Interp
Negative Logits
ean
-0.16
ting
-0.15
anik
-0.14
Sovere
-0.14
볨
-0.14
ffect
-0.14
erken
-0.14
amo
-0.14
ottage
-0.14
usu
-0.13
POSITIVE LOGITS
school
0.46
school
0.44
School
0.37
-school
0.36
SCHOOL
0.31
School
0.29
schools
0.29
_school
0.28
schools
0.28
chool
0.27
Activations Density 0.014%