INDEX
Explanations
references to specific names, especially "Jennings", and possibly specific actions or contexts associated with those names
mentions of specific individuals, particularly Jennings and Hendricks
New Auto-Interp
Negative Logits
undo
-0.74
unda
-0.72
tered
-0.72
fare
-0.70
tering
-0.67
planes
-0.67
achelor
-0.67
xious
-0.66
unch
-0.64
warts
-0.63
POSITIVE LOGITS
Jennings
1.03
patrick
0.76
yk
0.75
manship
0.73
Jarrett
0.72
Cla
0.71
BUG
0.70
iewicz
0.69
nect
0.69
Jenkins
0.68
Activations Density 0.013%