INDEX
Explanations
movement direction / continuation
New Auto-Interp
Negative Logits
鳎
0.37
señalado
0.36
Ната
0.35
llamado
0.35
潘
0.35
冀
0.35
ελ
0.34
называ
0.34
preferências
0.34
appelé
0.33
POSITIVE LOGITS
upto
0.67
advices
0.61
abit
0.59
thru
0.59
equipments
0.57
appart
0.55
atleast
0.55
unto
0.54
trough
0.53
ppl
0.51
Activations Density 0.067%