INDEX
Explanations
phrases indicating time and experience in context
New Auto-Interp
Negative Logits
oppel
-0.14
Ñģли
-0.14
lap
-0.14
crunch
-0.14
harek
-0.14
akk
-0.13
WithEvents
-0.13
adu
-0.13
avn
-0.13
-е
-0.13
POSITIVE LOGITS
small
0.27
pie
0.26
individual
0.23
Pie
0.23
various
0.21
smaller
0.21
å°ı
0.21
small
0.20
Pie
0.20
nhá»ı
0.20
Activations Density 0.233%