INDEX
Negative Logits
s
1.23
ו
1.18
2
1.17
ك
1.17
ل
1.15
യ
1.15
ro
1.07
ra
1.05
с
1.05
ל
1.02
POSITIVE LOGITS
৮
0.99
৭
0.94
a
0.92
্ড
0.91
Victoria
0.89
玼
0.86
Kr
0.85
Victoria
0.83
Đi
0.83
Agustus
0.82
Activations Density 0.002%
s
ו
2
ك
ل
യ
ro
ra
с
ל
৮
৭
a
্ড
Victoria
玼
Kr
Victoria
Đi
Agustus