INDEX
Explanations
summertime-related terms
references to the Summer and Winter Olympics, focusing on their respective years
New Auto-Interp
Negative Logits
atem
-0.90
otomy
-0.80
unal
-0.78
ulhu
-0.77
urally
-0.77
inates
-0.75
ourke
-0.75
akespe
-0.74
ittees
-0.74
pez
-0.73
POSITIVE LOGITS
Olympics
1.05
cit
0.82
bottom
0.81
Olympic
0.81
Festival
0.80
Soldier
0.78
Session
0.77
Pants
0.76
Wonderland
0.76
Ago
0.75
Activations Density 0.020%