INDEX
Negative Logits
encompassing
0.44
关注
0.41
('0.40
AMA
0.40
proceder
0.39
turut
0.39
వారం
0.39
necessitating
0.39
ظل
0.39
undergo
0.38
POSITIVE LOGITS
которые
0.53
Which
0.53
۔
0.52
которая
0.49
которое
0.48
जिसको
0.47
который
0.46
which
0.45
Which
0.45
और
0.44
Activations Density 0.002%