INDEX
Explanations
descriptions of trees and their environments
New Auto-Interp
Negative Logits
antor
-0.16
ogie
-0.16
anoi
-0.16
inx
-0.15
awai
-0.15
Arena
-0.15
buat
-0.15
omain
-0.14
astes
-0.14
ivant
-0.14
POSITIVE LOGITS
maple
0.30
oak
0.29
fir
0.29
elm
0.29
pine
0.28
ash
0.28
bir
0.27
Douglas
0.27
con
0.26
Norway
0.26
Activations Density 0.070%