INDEX
Explanations
phrases related to Olympic events and competitions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
537
+0.15
0.5%
32
+0.12
0.4%
1870
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
537
+0.15
0.03
1065
+0.12
0.03
1351
+0.11
0.02
Negative Logits
shewn
-0.57
anhyd
-0.55
hydrochlor
-0.54
malheure
-0.54
pollut
-0.54
citroen
-0.53
aimable
-0.51
chiare
-0.51
affez
-0.50
ویکیپدیا
-0.50
POSITIVE LOGITS
Olympic
1.19
Olympics
1.12
Olympic
1.00
Olímp
0.84
Olymp
0.82
Olympian
0.82
olymp
0.77
olympic
0.75
Olimp
0.71
Olymp
0.67
Activations Density 0.059%