INDEX
Explanations
emojis and special characters
special characters or unconventional symbols in text
New Auto-Interp
Negative Logits
Sakuya
-0.80
Gillespie
-0.74
Takeru
-0.71
wagen
-0.71
phrine
-0.69
Akin
-0.68
Bain
-0.68
oidal
-0.67
Abyssal
-0.65
ioned
-0.64
POSITIVE LOGITS
¥ŀ
1.24
ĵĺ
1.22
ĻĤ
1.21
ðŁ
1.20
Ĵ
1.17
Į
1.16
Ķ
1.14
ĵ
1.12
į
1.11
İ
1.10
Activations Density 0.004%