INDEX
Explanations
words associated with historical figures and martyrdom
New Auto-Interp
Negative Logits
*scale
-0.14
WithTitle
-0.14
owitz
-0.14
iston
-0.14
Clone
-0.14
ога
-0.14
unday
-0.14
beim
-0.13
Temple
-0.13
ãĥ¼ãĥ«ãĥī
-0.13
POSITIVE LOGITS
Ńå·ŀ
0.17
unlink
0.15
pend
0.14
uhan
0.14
pic
0.14
uzzi
0.14
vard
0.13
poss
0.13
Lind
0.13
loid
0.13
Activations Density 0.105%