INDEX
Explanations
phrases implying the passage of time
occurrences of the phrase "it has been."
New Auto-Interp
Negative Logits
ives
-0.74
achu
-0.68
ively
-0.66
Jobs
-0.65
rones
-0.65
robe
-0.65
apo
-0.64
lies
-0.64
odder
-0.64
arms
-0.63
POSITIVE LOGITS
bitten
0.90
taken
0.83
able
0.83
given
0.82
proven
0.80
unable
0.80
replaced
0.80
forgotten
0.79
shown
0.78
seen
0.77
Activations Density 0.131%