INDEX
Explanations
sports-related terms and activities, particularly baseball terminology
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.18
0.6%
453
+0.08
0.3%
1842
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1704
+0.18
0.08
1905
+0.08
0.07
701
+0.07
0.05
Negative Logits
<bos>
-2.29
endwhile
-0.71
ⓧ
-0.64
-0.63
writeFieldEnd
-0.59
énégal
-0.59
intios
-0.57
//{
-0.57
writeFileSync
-0.56
MessageOf
-0.54
POSITIVE LOGITS
Juf
1.42
Minang
1.36
Keny
1.34
stockholm
1.31
maneu
1.28
impra
1.25
bangkok
1.24
Confu
1.23
Hez
1.21
philanth
1.19
Activations Density 0.224%