Tagged code-completion
3 write-ups.
A Measured Headroom That Nothing Could Reach
A per-position oracle over my code completer's mixing weight is worth +0.0705 top-1, with a confidence interval excluding zero. Three feature families captured 0.0%, -3.1% and -0.9% of it, and the whole signal available to a per-class router turned out to be the same size as the bias from fitting one.
Code Completion From Your Own Repository's Statistics
One number decides whether it helps: the recital rate, 13.5% to 72.9% across nine corpora. Above 60% recital the n-gram index beats a ten-line frequency table; at 13.5% it beats nothing, the harness prints the null, and one published row was wrong and is corrected in the README.
What a Code Completer's Eval Never Measures: Ghost Text
A hybrid n-gram and 2.48M-parameter code completer that gets 54.6% of tokens right, tuned dial by dial against its own held-out eval. Then I scored the one feature that emits more than a single token and it came back at 0.070, because the eval hands the model a true prefix and the feature hands it its own output.