INDEX
Explanations
structural elements or syntax within code snippets
Code snippets ending with curly brackets
closing braces and punctuation
New Auto-Interp
Negative Logits
end
-1.46
End
-1.31
END
-1.23
end
-1.16
End
-1.15
END
-1.06
ends
-0.99
Ends
-0.83
ended
-0.82
ending
-0.79
POSITIVE LOGITS
}
0.80
}
0.63
.}
0.61
()}
0.60
!}
0.57
''}
0.57
("}\0.57
}
0.56
?}
0.56
'}
0.55
Activations Density 0.236%