INDEX
Explanations
actions involving rhythmic or forceful movements
New Auto-Interp
Negative Logits
sid
-0.15
ssi
-0.15
coles
-0.14
enou
-0.14
íĶ
-0.14
ipel
-0.14
sn
-0.13
ilet
-0.13
FI
-0.13
.uni
-0.13
POSITIVE LOGITS
harder
0.17
rhythm
0.17
rhythms
0.16
Tham
0.15
emas
0.15
23
0.15
ety
0.14
hardest
0.14
hard
0.14
awake
0.14
Activations Density 0.139%