INDEX
Explanations
phrases related to victories and match outcomes in sports
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.28
1.1%
1177
+0.12
0.5%
1842
+0.09
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1261
+0.28
0.03
825
+0.12
0.03
736
+0.09
0.03
Negative Logits
<bos>
-2.22
public
-0.78
for
-0.72
ⓧ
-0.72
if
-0.72
,
-0.71
}
-0.70
算
-0.69
成
-0.68
struct
-0.67
POSITIVE LOGITS
Juf
2.18
accla
1.89
Keny
1.76
Khart
1.72
Minang
1.72
véhic
1.70
Augu
1.67
increa
1.67
aen
1.67
volunte
1.67
Activations Density 0.138%