INDEX
Explanations
common phrases that indicate the beginning of an interview
instances of the phrase "in an interview with"
New Auto-Interp
Negative Logits
fighter
-0.75
orah
-0.68
bar
-0.68
cakes
-0.67
building
-0.66
gra
-0.65
nesia
-0.64
few
-0.63
amput
-0.63
control
-0.63
POSITIVE LOGITS
regards
0.84
RTX
0.81
Samantha
0.76
Collider
0.74
Wired
0.72
microphones
0.71
rompt
0.70
regard
0.70
Saban
0.70
Cth
0.68
Activations Density 0.075%