INDEX
Explanations
references to film criticism and analysis
New Auto-Interp
Negative Logits
ãn
-0.15
ackage
-0.14
VisualStyle
-0.14
aket
-0.14
ochen
-0.13
_compat
-0.13
prelim
-0.13
upos
-0.13
icros
-0.13
beh
-0.13
POSITIVE LOGITS
preco
0.20
propag
0.19
important
0.18
Contacts
0.17
contacts
0.17
pole
0.17
Contacts
0.17
mec
0.16
merit
0.16
activity
0.15
Activations Density 0.143%