INDEX
Explanations
references to folk culture and folk art
New Auto-Interp
Negative Logits
yx
-0.19
chyb
-0.18
eel
-0.18
cpy
-0.17
ghest
-0.17
.forName
-0.16
ãĤº
-0.15
nement
-0.15
eof
-0.15
abbo
-0.15
POSITIVE LOGITS
lor
0.47
lore
0.34
ta
0.33
лоÑĢ
0.28
ore
0.27
ways
0.27
lo
0.27
oric
0.25
tale
0.22
TA
0.21
Activations Density 0.007%