INDEX
Explanations
references to character development and audience connection in films
New Auto-Interp
Negative Logits
phem
-0.15
üh
-0.15
aja
-0.15
azer
-0.14
trom
-0.14
vpn
-0.14
Ministers
-0.14
dsl
-0.14
Boundary
-0.14
Weber
-0.13
POSITIVE LOGITS
Bat
0.49
bat
0.47
Batman
0.45
Bat
0.43
Gotham
0.43
BAT
0.42
bat
0.41
Batman
0.40
BAT
0.40
bats
0.39
Activations Density 0.051%