INDEX
Explanations
references to extreme heat and temperature conditions
New Auto-Interp
Negative Logits
burgh
-0.16
sworth
-0.15
iced
-0.15
:host
-0.15
sanit
-0.14
winters
-0.14
.mods
-0.14
Maul
-0.14
Troll
-0.14
691
-0.14
POSITIVE LOGITS
heat
0.16
killer
0.16
-Ta
0.15
ROTO
0.15
heat
0.15
ecut
0.15
ldb
0.15
zf
0.14
irt
0.14
lij
0.14
Activations Density 0.137%