INDEX
Explanations
the name "Ka" in various contexts
New Auto-Interp
Negative Logits
utut
-0.18
ya
-0.17
OMIT
-0.16
RIEND
-0.16
ut
-0.16
hare
-0.15
icast
-0.15
yas
-0.15
urm
-0.14
omon
-0.14
POSITIVE LOGITS
ooke
0.16
ie
0.16
meis
0.15
عب
0.15
ovsky
0.14
HEL
0.14
-toggler
0.14
eri
0.14
tub
0.14
oid
0.14
Activations Density 0.012%