INDEX
Explanations
phrases related to taking off or departing
instances of the phrase "take off."
New Auto-Interp
Negative Logits
Gall
-0.70
NG
-0.64
uddy
-0.63
eller
-0.61
rite
-0.61
ili
-0.60
resy
-0.59
obs
-0.59
abul
-0.59
=]
-0.59
POSITIVE LOGITS
screen
0.79
river
0.76
pist
0.73
takeoff
0.68
hand
0.67
stage
0.65
leash
0.62
site
0.62
wash
0.60
shore
0.60
Activations Density 0.019%