INDEX
Explanations
text related to news headlines or articles, particularly with a focus on specific cities or states
content related to social media interactions and shares
New Auto-Interp
Negative Logits
hurd
-0.68
thal
-0.63
olkien
-0.63
minist
-0.62
intern
-0.61
conflic
-0.60
osate
-0.60
REL
-0.60
schild
-0.60
jong
-0.60
POSITIVE LOGITS
Copy
0.87
0.82
Share
0.77
Hide
0.77
20439
0.76
0.75
Tweet
0.74
Skype
0.73
Flavoring
0.73
0.73
Activations Density 0.050%