INDEX
Explanations
words related to national security
references to national security
New Auto-Interp
Negative Logits
amaz
-0.92
oven
-0.74
nos
-0.74
bare
-0.73
Torrent
-0.72
Blocks
-0.71
netflix
-0.70
forth
-0.70
ken
-0.68
chrom
-0.68
POSITIVE LOGITS
adviser
1.15
advisor
1.09
Adviser
1.04
Advisor
1.02
Advis
0.95
Agency
0.91
apparatus
0.90
advisers
0.86
advisors
0.85
briefings
0.85
Activations Density 0.043%