INDEX
Explanations
references to children and families
New Auto-Interp
Negative Logits
utherford
-0.17
anoia
-0.15
Ø·ÙĦ
-0.15
esin
-0.15
¹Ħ
-0.15
atif
-0.14
Äĥr
-0.14
uteur
-0.14
ÄĻk
-0.14
rijk
-0.13
POSITIVE LOGITS
Peer
0.25
peer
0.25
peer
0.24
peers
0.24
same
0.23
Peer
0.23
same
0.23
OfSize
0.22
similar
0.22
around
0.20
Activations Density 0.130%