INDEX
Negative Logits
Dutch
0.55
dutch
0.48
Groningen
0.45
Dutch
0.44
溇
0.42
الج
0.41
Unreal
0.41
coff
0.40
coffin
0.40
Belgian
0.39
POSITIVE LOGITS
Wass
0.52
Corpus
0.52
corpus
0.51
Maduro
0.48
Security
0.45
corpus
0.45
Beach
0.44
Du
0.44
Sche
0.43
ressort
0.43
Activations Density 0.003%