INDEX
Explanations
proper nouns, names, and titles related to sports and music
New Auto-Interp
Negative Logits
ilde
-0.72
ARA
-0.68
fully
-0.67
ilee
-0.67
orsi
-0.66
ward
-0.66
PsyNetMessage
-0.66
imaru
-0.66
upon
-0.65
BER
-0.65
POSITIVE LOGITS
Ancients
0.86
apocalypse
0.86
Apocalypse
0.85
Americas
0.83
Confederacy
0.80
latter
0.79
Month
0.79
utmost
0.78
same
0.78
highest
0.77
Activations Density 0.106%