INDEX
Explanations
references to names and places in Hungary
New Auto-Interp
Negative Logits
ksam
-0.18
nder
-0.17
äm
-0.17
ks
-0.16
wealth
-0.16
deal
-0.15
iks
-0.14
rops
-0.14
etty
-0.14
ivery
-0.14
POSITIVE LOGITS
ban
0.22
abb
0.19
ese
0.16
ultan
0.16
oss
0.16
unk
0.16
fel
0.16
felt
0.16
legs
0.15
villa
0.15
Activations Density 0.001%