INDEX
Explanations
curly brackets with a number inside
closing braces and other symbols used in programming or mathematical expressions
New Auto-Interp
Negative Logits
ateur
-0.83
eline
-0.78
ateurs
-0.75
arios
-0.71
enium
-0.71
omorphic
-0.71
nesday
-0.70
oresc
-0.69
uga
-0.68
elines
-0.68
POSITIVE LOGITS
]}
1.04
else
0.90
}
0.85
},
0.85
.}
0.83
]]
0.82
...]
0.82
},"
0.81
::::::::
0.80
··
0.80
Activations Density 0.011%