INDEX
Explanations
descriptions related to gameplay elements and maneuvers in a video game, particularly focusing on move combinations and character abilities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
453
+0.10
0.3%
1706
+0.09
0.2%
286
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1706
+0.10
0.03
458
+0.09
0.03
286
+0.07
0.03
Negative Logits
kask
-0.65
karton
-0.62
kompati
-0.61
silikon
-0.61
krab
-0.60
kade
-0.60
elek
-0.60
kasa
-0.59
kriminal
-0.57
etik
-0.57
POSITIVE LOGITS
another
1.03
others
1.00
another
0.90
others
0.84
Others
0.78
Another
0.75
Others
0.74
Another
0.74
otra
0.71
另一
0.71
Activations Density 0.151%