INDEX
Explanations
various forms of educational and cultural content, particularly exhibitions, series, and research-related topics
New Auto-Interp
Negative Logits
inis
-0.15
612
-0.14
614
-0.13
ptrdiff
-0.13
Tanner
-0.13
posable
-0.13
anst
-0.13
873
-0.13
ween
-0.13
611
-0.13
POSITIVE LOGITS
about
0.49
devoted
0.43
åħ³äºİ
0.41
about
0.38
tentang
0.37
dedicated
0.35
vá»ģ
0.33
concerning
0.33
regarding
0.30
focused
0.30
Activations Density 0.359%