INDEX
Explanations
instances of the term "RT" indicating retweets in social media context
New Auto-Interp
Negative Logits
hiro
-0.15
hausen
-0.14
splash
-0.14
lava
-0.14
heit
-0.14
abi
-0.14
Neutral
-0.14
pare
-0.14
achu
-0.14
rych
-0.14
POSITIVE LOGITS
RT
0.20
(rt
0.18
iers
0.17
gers
0.16
anic
0.16
rt
0.15
ificial
0.15
Ãī
0.15
ÑĥÑĤÑĮ
0.14
_bitmap
0.14
Activations Density 0.008%