INDEX
Explanations
references to significant events and performances, particularly related to music and competitions
New Auto-Interp
Negative Logits
fully
-0.19
elian
-0.17
ably
-0.16
edly
-0.16
elijke
-0.16
roll
-0.16
lessly
-0.15
лев
-0.15
fulness
-0.15
alim
-0.15
POSITIVE LOGITS
ity
0.24
ities
0.24
most
0.21
mente
0.20
itarian
0.19
gado
0.19
ogy
0.17
zeitig
0.16
ogl
0.16
-minded
0.16
Activations Density 0.318%