INDEX
Explanations
phrases related to sports events and activities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1601
+0.08
0.2%
11
+0.08
0.2%
562
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1336
+0.08
0.04
11
+0.08
0.03
562
+0.08
0.03
Negative Logits
adal
-0.48
pexpr
-0.47
ModelMap
-0.45
NegativeButton
-0.43
@[+][
-0.43
codiles
-0.43
Ӧ
-0.43
Rptr
-0.42
CURLOPT
-0.42
NoSuch
-0.42
POSITIVE LOGITS
shenan
1.02
impra
0.97
disagre
0.93
disreg
0.93
encomp
0.86
indestru
0.85
intersper
0.85
increa
0.83
inconce
0.83
milf
0.82
Activations Density 0.132%