INDEX
Explanations
academic disciplines and research fields related to science and engineering
New Auto-Interp
Negative Logits
ior
-0.18
etti
-0.17
ORD
-0.15
нен
-0.15
lyon
-0.14
prech
-0.14
Arch
-0.14
isol
-0.14
ilog
-0.14
cob
-0.14
POSITIVE LOGITS
ingleton
0.17
gree
0.15
ninger
0.15
ozÃŃ
0.15
ahun
0.14
еÑĢап
0.14
@show
0.14
stry
0.14
amb
0.13
μμε
0.13
Activations Density 0.017%