INDEX
Explanations
discussion of military tactics and historical battles
New Auto-Interp
Negative Logits
acro
-0.15
zcze
-0.15
alem
-0.15
umas
-0.14
chein
-0.14
ecture
-0.13
ecta
-0.13
errick
-0.13
lei
-0.13
stash
-0.13
POSITIVE LOGITS
mus
0.28
cu
0.26
Horse
0.26
foot
0.26
horse
0.26
Foot
0.25
pik
0.25
cav
0.25
Musk
0.24
mounted
0.23
Activations Density 0.050%