INDEX
Explanations
instances of the word "bitter" and related concepts
New Auto-Interp
Negative Logits
allows
-0.86
arlane
-0.82
ulhu
-0.78
nesota
-0.78
upload
-0.78
authorized
-0.76
aver
-0.76
alian
-0.75
ioch
-0.75
avers
-0.74
POSITIVE LOGITS
bitter
0.94
disappointment
0.94
bitterness
0.91
parting
0.90
revenge
0.88
irony
0.86
terness
0.84
streaks
0.83
cold
0.83
resentment
0.82
Activations Density 0.008%