INDEX
Explanations
references to thunder and lightning phenomena
New Auto-Interp
Negative Logits
gli
-0.17
/bind
-0.16
íĥĪ
-0.15
fp
-0.14
Historic
-0.14
INST
-0.14
enda
-0.14
sla
-0.13
historic
-0.13
zer
-0.13
POSITIVE LOGITS
bolt
0.34
storm
0.33
storms
0.32
bol
0.30
ous
0.30
ously
0.24
bird
0.24
str
0.22
OUS
0.22
cloud
0.22
Activations Density 0.005%