INDEX
Explanations
references to baseball teams, players, and related events
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
156
+0.13
0.5%
976
+0.13
0.4%
874
+0.12
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1601
+0.13
0.04
156
+0.13
0.04
1043
+0.12
0.04
Negative Logits
cryst
-0.58
embodi
-0.58
كومونز
-0.57
unlaw
-0.55
PHO
-0.54
toprule
-0.54
sputnik
-0.53
urso
-0.52
ikea
-0.52
pymongo
-0.52
POSITIVE LOGITS
baseball
1.28
Baseball
1.24
Baseball
1.10
baseball
1.09
MLB
0.98
MLB
0.83
pitchers
0.80
pitcher
0.77
isbol
0.69
batting
0.67
Activations Density 0.185%