INDEX
Explanations
proper names, specifically the name "George"
instances of the name "George"
New Auto-Interp
Negative Logits
mble
-0.90
ngth
-0.80
crawl
-0.77
odcast
-0.74
eer
-0.72
igslist
-0.71
trak
-0.70
Trend
-0.68
eeper
-0.68
reinforcement
-0.66
POSITIVE LOGITS
George
1.06
Lucas
0.92
Bush
0.90
George
0.89
Zimmerman
0.82
Thor
0.79
Georg
0.79
Soros
0.78
iannopoulos
0.77
Herbert
0.76
Activations Density 0.010%