INDEX
Explanations
historical references and periods
New Auto-Interp
Negative Logits
243
-0.15
ekim
-0.15
icontrol
-0.15
Thr
-0.15
lien
-0.14
viders
-0.14
244
-0.14
unma
-0.14
linik
-0.13
oku
-0.13
POSITIVE LOGITS
ninth
0.35
ele
0.35
fif
0.32
teenth
0.32
Late
0.31
tenth
0.31
seventh
0.29
Middle
0.28
tw
0.28
sixth
0.28
Activations Density 0.111%