INDEX
Explanations
references to the Olympic Games and related terminology
New Auto-Interp
Negative Logits
éĿ©
-0.17
rag
-0.15
undi
-0.15
sel
-0.14
srv
-0.14
리그
-0.14
imdi
-0.14
sse
-0.14
605
-0.14
ward
-0.14
POSITIVE LOGITS
Games
0.27
Games
0.21
torch
0.21
Village
0.21
Flame
0.20
Torch
0.20
hopeful
0.20
-sized
0.20
games
0.20
flame
0.19
Activations Density 0.008%