INDEX
Explanations
words related to oils
references to oils and metals
New Auto-Interp
Negative Logits
ership
-0.87
Taliban
-0.64
Plymouth
-0.63
story
-0.62
Translation
-0.60
ee
-0.60
Flyers
-0.58
neck
-0.58
Bird
-0.58
vision
-0.58
POSITIVE LOGITS
mith
1.23
paces
1.13
creen
1.04
uggest
0.99
pace
0.95
oils
0.95
ynthesis
0.93
ystem
0.92
boro
0.92
terday
0.89
Activations Density 0.035%