INDEX
Explanations
words and phrases related to gaming strategies and tactics, particularly relating to improvements and modifications in gameplay
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
736
+0.11
0.3%
50
+0.10
0.3%
690
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1551
+0.11
0.07
736
+0.10
0.09
273
+0.10
0.05
Negative Logits
ImageContext
-0.81
<bos>
-0.80
IsContent
-0.71
מבר
-0.70
שוליים
-0.68
<<<<<<<<<<<<<<
-0.68
HtmlAttribute
-0.68
__).
-0.67
hectáreas
-0.65
DoubleQuotes
-0.65
POSITIVE LOGITS
disagre
1.96
impra
1.81
unspeak
1.81
reluct
1.79
shenan
1.76
apprehen
1.75
emphat
1.71
disreg
1.70
ineffec
1.68
accla
1.68
Activations Density 1.314%