INDEX
Explanations
mentions of friends or relationships in various contexts
phrases that describe relationships or connections between people
New Auto-Interp
Negative Logits
essen
-0.71
reperto
-0.67
ItemImage
-0.66
OGR
-0.65
PF
-0.64
percentages
-0.63
largeDownload
-0.61
erial
-0.61
ulo
-0.61
partName
-0.60
POSITIVE LOGITS
hers
1.10
ours
1.08
theirs
0.97
yours
0.88
sorts
0.85
mine
0.84
irlf
0.81
Mine
0.78
whom
0.75
course
0.69
Activations Density 0.087%