INDEX
Explanations
references to time periods, specifically the word "century"
references to the concept of "century."
New Auto-Interp
Negative Logits
ramid
-0.83
doms
-0.78
inki
-0.76
govtrack
-0.72
liga
-0.72
hod
-0.70
gradient
-0.70
ettings
-0.68
gur
-0.67
vals
-0.67
POSITIVE LOGITS
Ago
0.89
ago
0.84
ocene
0.76
Clicker
0.75
BCE
0.72
Oaks
0.72
Ferdinand
0.72
osaurs
0.70
hindsight
0.69
Daughter
0.68
Activations Density 0.015%