INDEX
Explanations
mentions of peanut-related words
references to peanuts and related terms
New Auto-Interp
Negative Logits
Cors
-0.75
fare
-0.73
Desc
-0.72
Contin
-0.70
learn
-0.68
Anonymous
-0.68
Glas
-0.67
Angels
-0.67
CV
-0.66
Supporting
-0.66
POSITIVE LOGITS
peanut
1.30
anut
1.30
peanuts
1.21
butter
1.09
brittle
0.94
heon
0.92
bean
0.86
popcorn
0.86
anuts
0.83
ricia
0.82
Activations Density 0.002%