INDEX
Explanations
references to specific geographical locations and institutions
New Auto-Interp
Negative Logits
Bj
-0.17
oda
-0.17
etre
-0.16
assi
-0.15
BF
-0.15
oined
-0.15
izoph
-0.15
argins
-0.15
omic
-0.14
tangent
-0.14
POSITIVE LOGITS
ALAR
0.15
Head
0.15
kili
0.14
croft
0.14
781
0.14
conform
0.14
Roc
0.14
(Constructor
0.13
trad
0.13
åħ§
0.13
Activations Density 0.029%