INDEX
Explanations
proper nouns related to names, specifically personal names
New Auto-Interp
Negative Logits
igel
-0.15
trag
-0.15
resh
-0.15
ensis
-0.14
×¢
-0.14
blr
-0.14
vedle
-0.14
letal
-0.14
coni
-0.14
wr
-0.14
POSITIVE LOGITS
ansson
0.23
elyn
0.22
ève
0.17
annes
0.17
venes
0.17
imary
0.16
athan
0.16
ëĤľ
0.15
anna
0.15
anny
0.15
Activations Density 0.029%