INDEX
Explanations
references to Facebook pages and posts
New Auto-Interp
Negative Logits
fox
-0.15
cratch
-0.15
Fox
-0.15
essages
-0.15
insky
-0.14
omor
-0.14
stime
-0.14
inki
-0.14
Secret
-0.14
ekt
-0.14
POSITIVE LOGITS
scn
0.16
NodeType
0.15
quate
0.15
0.14
Ley
0.14
ÄIJT
0.14
iphy
0.14
newline
0.14
waters
0.14
/mobile
0.14
Activations Density 0.044%