INDEX
Explanations
construction-related terms and processes
New Auto-Interp
Negative Logits
fingerprints
-0.16
arters
-0.16
Jacob
-0.16
shaft
-0.15
iasi
-0.15
domic
-0.14
Miy
-0.14
Jeh
-0.14
spin
-0.14
idos
-0.14
POSITIVE LOGITS
membranes
0.24
membrane
0.24
Mem
0.23
geom
0.22
pond
0.22
Pond
0.22
mem
0.21
pond
0.20
.mem
0.20
flashing
0.20
Activations Density 0.031%