INDEX
Explanations
references to ballet and dance education
New Auto-Interp
Negative Logits
ASC
-0.17
оÑĢони
-0.17
utsch
-0.16
Babe
-0.15
eel
-0.15
rán
-0.15
isay
-0.14
ammer
-0.14
Binder
-0.14
ëĿ½
-0.14
POSITIVE LOGITS
ballet
0.38
Ballet
0.34
dancers
0.28
danced
0.27
dancer
0.27
Swan
0.24
Nut
0.24
dancing
0.23
tut
0.23
Dance
0.23
Activations Density 0.019%