INDEX
Explanations
references to schools and educational contexts
New Auto-Interp
Negative Logits
ViewFeatures
-0.75
Mante
-0.73
\}_{-0.73
вяз
-0.72
réhen
-0.71
complexContent
-0.71
picable
-0.71
idends
-0.70
ANTED
-0.70
wüns
-0.69
POSITIVE LOGITS
school
2.30
School
2.14
school
2.08
School
2.03
schools
1.90
SCHOOL
1.90
SCHOOL
1.79
chool
1.72
Schools
1.70
schools
1.61
Activations Density 0.050%