INDEX
Explanations
phrases related to shooting or gun-related actions
references to the term "Bangladesh" and variations of the word 'Bang'
New Auto-Interp
Negative Logits
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
-0.71
ensional
-0.71
icient
-0.71
haps
-0.69
externalToEVAOnly
-0.64
uded
-0.64
wcsstore
-0.64
eper
-0.64
VICE
-0.61
udes
-0.61
POSITIVE LOGITS
kok
1.40
alore
1.36
Bang
1.15
bang
1.13
Bang
1.12
bang
1.09
adesh
0.99
la
0.91
alter
0.90
sam
0.90
Activations Density 0.023%