INDEX
Explanations
sports-related information like player names, injuries, performances, and team updates
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1870
+0.13
0.4%
479
+0.12
0.4%
421
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
479
+0.13
0.04
2036
+0.12
0.03
341
+0.10
0.03
Negative Logits
auguri
-0.61
bewerken
-0.60
particolar
-0.60
javier
-0.59
iscri
-0.59
autunno
-0.58
ohr
-0.58
trovo
-0.58
affez
-0.57
alberto
-0.57
POSITIVE LOGITS
games
1.03
games
0.96
Games
0.90
game
0.90
Games
0.88
GAMES
0.84
game
0.81
Game
0.78
GAME
0.71
Game
0.70
Activations Density 0.090%