INDEX
Explanations
references to types of food or culinary practices, especially those comparing different cuisines
New Auto-Interp
Negative Logits
udic
-0.15
onse
-0.15
<!--[
-0.15
aras
-0.14
pty
-0.14
Claude
-0.14
afil
-0.14
orre
-0.14
onne
-0.13
amt
-0.13
POSITIVE LOGITS
counterpart
0.39
counterparts
0.38
predecessors
0.26
brethren
0.26
peers
0.26
colleagues
0.24
fellow
0.24
predecessor
0.23
neighbours
0.22
predecess
0.21
Activations Density 0.073%