INDEX
Explanations
phrases related to American football
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1265
+0.14
0.4%
2019
+0.14
0.4%
1103
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1103
+0.14
0.05
1265
+0.14
0.04
1173
+0.11
0.04
Negative Logits
lele
-1.95
aen
-1.81
fte
-1.81
mef
-1.80
fta
-1.70
tew
-1.69
tanga
-1.69
oner
-1.69
lara
-1.69
bangkok
-1.68
POSITIVE LOGITS
according
0.71
therefore
0.69
depending
0.65
if
0.65
except
0.64
importantly
0.64
as
0.63
like
0.63
although
0.63
despite
0.62
Activations Density 0.078%