INDEX
Explanations
various forms and contexts of advertising
New Auto-Interp
Negative Logits
dan
-0.17
-door
-0.15
ton
-0.15
votes
-0.14
endale
-0.14
ay
-0.14
outh
-0.14
風
-0.14
ep
-0.13
advertised
-0.13
POSITIVE LOGITS
hoc
0.25
campaigns
0.22
orial
0.21
hoc
0.20
-supported
0.20
/mark
0.20
/prom
0.20
jectives
0.19
orption
0.19
obe
0.19
Activations Density 0.022%