INDEX
Explanations
words related to a specific brand or company name
occurrences of the word "bumble."
New Auto-Interp
Negative Logits
ablishment
-0.88
olphins
-0.82
hips
-0.81
ohn
-0.80
nesota
-0.77
thening
-0.76
asel
-0.72
izations
-0.70
atural
-0.70
afort
-0.70
POSITIVE LOGITS
weed
1.10
Britann
0.83
Doodle
0.79
ble
0.78
bles
0.76
bum
0.76
grass
0.74
Afric
0.74
Upon
0.73
BUG
0.73
Activations Density 0.008%