INDEX
Explanations
words related to death or dying
New Auto-Interp
Negative Logits
ial
-0.17
ëĭ´
-0.15
Born
-0.15
ارÙĩ
-0.15
ãĥ¥
-0.14
inery
-0.14
mutation
-0.14
rette
-0.14
Deadly
-0.14
mt
-0.14
POSITIVE LOGITS
lectric
0.18
intest
0.18
defending
0.18
young
0.17
_slow
0.16
/be
0.16
violent
0.15
-lfs
0.15
slow
0.15
elp
0.15
Activations Density 0.029%