INDEX
Negative Logits
모델
-0.80
מכל
-0.79
끝
-0.73
космо
-0.73
xét
-0.72
itaji
-0.71
approval
-0.71
ódó
-0.69
Illuminated
-0.69
観察
-0.69
POSITIVE LOGITS
optical
3.33
illusion
3.23
illusions
2.75
optical
2.72
Optical
2.58
Optical
2.56
illusion
2.39
Illusion
2.27
tricks
2.14
опти
2.05
Activations Density 0.050%