INDEX
Explanations
terms relevant to scientific and medical contexts
New Auto-Interp
Negative Logits
lung
-0.15
ikh
-0.15
hil
-0.14
uth
-0.14
lung
-0.14
zk
-0.14
Lung
-0.14
dÃŃ
-0.13
kup
-0.13
philanth
-0.13
POSITIVE LOGITS
lin
0.24
doen
0.21
sind
0.21
grave
0.21
tum
0.20
farm
0.20
pat
0.20
grave
0.19
oft
0.19
grip
0.19
Activations Density 0.009%