INDEX
Explanations
references to "vitamin" and related terms
New Auto-Interp
Negative Logits
ahun
-0.19
azzi
-0.17
ksam
-0.17
aç
-0.15
ĸ
-0.15
otherwise
-0.15
hetto
-0.15
åŁ
-0.15
RS
-0.14
ioned
-0.14
POSITIVE LOGITS
ilig
0.24
ória
0.22
ally
0.20
amins
0.19
ru
0.19
uper
0.19
ello
0.17
rine
0.17
ri
0.17
rect
0.17
Activations Density 0.007%