INDEX
Explanations
phrases mentioning the Negev desert
references to geographical locations
New Auto-Interp
Negative Logits
kefeller
-1.09
icum
-0.90
Interstitial
-0.87
ities
-0.77
acca
-0.75
sidel
-0.72
iqueness
-0.70
idental
-0.69
ITIES
-0.68
ideo
-0.66
POSITIVE LOGITS
hog
1.12
orge
1.00
ORGE
0.98
cko
0.88
orget
0.87
rer
0.86
eks
0.82
rers
0.80
orgetown
0.79
geist
0.76
Activations Density 0.014%