INDEX
Explanations
instances of the letter 'h' and 'v' as prominent characters or symbols
New Auto-Interp
Negative Logits
u
-0.20
iu
-0.18
id
-0.17
oth
-0.16
oi
-0.16
oon
-0.16
oe
-0.16
dio
-0.15
dn
-0.15
iw
-0.15
POSITIVE LOGITS
eel
0.17
e
0.17
iform
0.17
zı
0.17
ch
0.16
A
0.16
Discrim
0.16
arias
0.15
ç«ĭ
0.15
Wilkinson
0.15
Activations Density 0.094%