INDEX
Explanations
references to the word "Kyoto" or similar variations of it
New Auto-Interp
Negative Logits
.EXTRA
-0.16
chip
-0.16
onica
-0.15
efeller
-0.14
LANG
-0.14
å·»
-0.14
inery
-0.14
GANG
-0.14
gig
-0.14
æį·
-0.14
POSITIVE LOGITS
rgyz
0.24
rie
0.22
Ky
0.21
ky
0.18
Ky
0.18
riad
0.17
ameleon
0.17
ung
0.17
ehler
0.16
ogle
0.16
Activations Density 0.008%