INDEX
Explanations
information about sports statistics and players
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
906
+0.10
0.3%
1343
+0.10
0.3%
392
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
523
+0.10
0.03
1597
+0.10
0.02
906
+0.10
0.00
Negative Logits
abestanden
-0.64
pathlib
-0.58
wydar
-0.55
hashlib
-0.55
wykonania
-0.54
mieszkań
-0.54
szkole
-0.54
BUILDINGS
-0.54
suchte
-0.53
архивлан
-0.53
POSITIVE LOGITS
emphat
1.36
accla
1.27
applau
1.24
incess
1.23
michelin
1.20
embra
1.19
sobri
1.15
hentai
1.11
suspic
1.09
milf
1.08
Activations Density 0.272%