INDEX
Explanations
the game of chess
references to chess
New Auto-Interp
Negative Logits
ient
-0.82
igious
-0.78
aneous
-0.74
unnamed
-0.66
rogen
-0.65
aneously
-0.64
ufact
-0.64
phrine
-0.64
uated
-0.64
regon
-0.63
POSITIVE LOGITS
chess
1.20
bowl
0.92
Chess
0.91
manship
0.86
puzzles
0.85
cube
0.80
Solitaire
0.79
players
0.78
puzzle
0.78
rook
0.75
Activations Density 0.009%