INDEX
Explanations
mentions or discussions related to boxing
New Auto-Interp
Negative Logits
pard
-0.81
vironment
-0.76
ividual
-0.74
Reviewer
-0.74
utter
-0.73
closure
-0.72
Frey
-0.71
steen
-0.71
itizen
-0.70
urion
-0.70
POSITIVE LOGITS
instructor
0.95
competitions
0.92
prowess
0.89
instructors
0.88
nas
0.88
academy
0.87
chops
0.85
lessons
0.85
routines
0.85
training
0.84
Activations Density 0.061%