INDEX
Explanations
references to crafting and handmade creations
New Auto-Interp
Negative Logits
rael
-0.18
vast
-0.17
ht
-0.16
otes
-0.16
sac
-0.16
uppy
-0.16
phant
-0.15
deki
-0.15
erde
-0.15
oted
-0.15
POSITIVE LOGITS
erson
0.19
bilt
0.16
ificial
0.16
peare
0.16
ivism
0.15
ers
0.15
igli
0.15
McGu
0.15
iness
0.15
ivist
0.14
Activations Density 0.047%