INDEX
Explanations
mentions of a specific person's name "Han" along with positive characteristics related to sports
New Auto-Interp
Negative Logits
Thumbnails
-0.72
Colossus
-0.68
destro
-0.68
tsky
-0.67
ENCE
-0.66
utics
-0.66
ktop
-0.65
anwhile
-0.63
llan
-0.62
IMAGES
-0.61
POSITIVE LOGITS
ning
1.08
auer
1.06
wei
0.99
nington
0.98
uman
0.97
hao
0.95
lon
0.93
ako
0.93
Solo
0.93
bang
0.91
Activations Density 0.059%