INDEX
Explanations
phrases related to game mechanics and strategies, specifically involving characters and their abilities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
663
+0.09
0.3%
1810
+0.09
0.3%
899
+0.09
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1846
+0.09
0.02
1559
+0.09
0.02
899
+0.09
0.02
Negative Logits
unden
-1.35
inev
-1.34
reluct
-1.33
embra
-1.33
desir
-1.31
accla
-1.29
secon
-1.27
oner
-1.26
fta
-1.25
erec
-1.24
POSITIVE LOGITS
compensate
0.99
offset
0.97
compensated
0.90
compensation
0.88
offset
0.82
offsets
0.79
compens
0.76
Offset
0.75
Offset
0.74
compens
0.73
Activations Density 0.156%