INDEX
Explanations
references to specific companies and their promotional events
New Auto-Interp
Negative Logits
es
-0.23
a
-0.22
esch
-0.19
pars
-0.19
i
-0.18
an
-0.18
aes
-0.18
и
-0.18
aire
-0.17
bane
-0.17
POSITIVE LOGITS
bling
0.20
plitude
0.20
bole
0.19
mit
0.18
blers
0.17
plit
0.17
بÙĪÙĦ
0.16
eric
0.16
pton
0.16
ming
0.15
Activations Density 0.069%