INDEX
Explanations
mentions of a particular sweet food item, **honey**
references to honey
New Auto-Interp
Negative Logits
aneous
-0.89
aneously
-0.84
uers
-0.79
andals
-0.76
ative
-0.72
osure
-0.72
================================================================
-0.72
ATIONS
-0.68
igious
-0.67
ivals
-0.66
POSITIVE LOGITS
moon
1.18
bee
1.17
bee
1.17
honey
1.16
Honey
1.09
bees
1.08
bean
0.97
comb
0.96
Bee
0.94
bees
0.94
Activations Density 0.003%