INDEX
Explanations
mentions of boxing matches and related details
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
304
+0.10
0.3%
1842
+0.09
0.3%
1675
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
244
+0.10
0.05
1675
+0.09
0.06
632
+0.08
0.03
Negative Logits
pollut
-0.94
effe
-0.90
vian
-0.90
erec
-0.89
aen
-0.87
dovr
-0.85
levis
-0.83
mef
-0.83
venuto
-0.82
guatemala
-0.81
POSITIVE LOGITS
UFC
0.57
titles
0.53
knockout
0.53
undefeated
0.53
rematch
0.52
brawl
0.51
title
0.51
trainer
0.50
challenger
0.50
challeng
0.50
Activations Density 0.485%