From c5aba3bead798352d6726c082456a48e76049960 Mon Sep 17 00:00:00 2001 From: Paul Gauthier Date: Tue, 19 Dec 2023 20:42:04 -0600 Subject: [PATCH] copy --- docs/unified-diffs.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/docs/unified-diffs.md b/docs/unified-diffs.md index cfc1e5020..e5d2ca147 100644 --- a/docs/unified-diffs.md +++ b/docs/unified-diffs.md @@ -385,8 +385,9 @@ fine tuning models on aider's simple, high level style of unified diffs. Dropping line numbers from the hunk headers and focusing on diffs of semantically coherent chunks of code -seems to be an important part of successful GPT code editing. +seems to be an important part of successful GPT code editing +(besides the relentless focus on flexibly applying edits). Most LLMs will have already seen plenty of unified diffs in their normal training data, and so should be -very amenable to fining tuning towards this +amenable to fining tuning towards this particular diff style.