INDEX
Explanations
mentions of screens or devices with screens
instances of the word "screen."
New Auto-Interp
Negative Logits
ghan
-0.76
istan
-0.71
doms
-0.71
Carnegie
-0.69
aternal
-0.69
ETH
-0.68
ortium
-0.67
IGH
-0.67
lesh
-0.63
istani
-0.63
POSITIVE LOGITS
plays
1.23
tops
1.01
printed
0.98
screens
0.97
caps
0.95
TVs
0.90
wip
0.89
reens
0.87
writers
0.85
wallpaper
0.84
Activations Density 0.017%