INDEX
Negative Logits
rans
0.69
ধানে
0.64
variable
0.63
ductile
0.63
ర్ప
0.62
ächst
0.61
<0xA2>
0.60
됩니다
0.60
槸
0.60
付き
0.60
POSITIVE LOGITS
appealed
2.20
appeals
2.13
appeal
2.10
excites
2.10
excite
1.98
intrig
1.89
intrigued
1.86
piqu
1.85
Appeal
1.83
inspires
1.80
Activations Density 0.637%