INDEX
Explanations
references to flights
occurrences of the word "flight."
New Auto-Interp
Negative Logits
zzo
-0.75
tags
-0.71
soType
-0.68
Honest
-0.68
Micha
-0.68
eworks
-0.67
enegger
-0.65
cmp
-0.65
illard
-0.64
Graveyard
-0.64
POSITIVE LOGITS
flight
3.85
Flight
2.92
Flight
2.77
flight
2.70
flights
2.49
flying
1.58
airplane
1.53
plane
1.53
airline
1.51
fly
1.45
Activations Density 0.011%