INDEX
Explanations
time-related phrases and references to nighttime activities
New Auto-Interp
Negative Logits
sunrise
-0.23
mornings
-0.22
dawn
-0.20
morning
-0.20
sunshine
-0.19
breakfast
-0.18
brunch
-0.18
daytime
-0.18
Morning
-0.17
noon
-0.17
POSITIVE LOGITS
bed
0.42
-bed
0.39
bed
0.36
Bed
0.35
Bed
0.34
_bed
0.30
BED
0.29
bedtime
0.29
beds
0.28
.bed
0.28
Activations Density 0.136%