INDEX
Explanations
phrases that contain the word "about"
phrases that reference a project or description of a narrative
New Auto-Interp
Negative Logits
iaries
-0.77
reviewed
-0.72
room
-0.70
ãĥ¯
-0.69
oft
-0.68
brate
-0.66
inois
-0.64
Boyle
-0.64
pps
-0.63
endiary
-0.63
POSITIVE LOGITS
impending
0.81
nipple
0.81
evils
0.76
nipples
0.71
sudden
0.71
horizon
0.69
imaginary
0.69
daddy
0.68
topic
0.66
uniqueness
0.66
Activations Density 0.226%