Repository navigation
Ground the distillation guide in supported methods and evidence - #223
Merged
sarthakagrawal927 merged 1 commit intoNov 4, 2026
Merged
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
sarthakagrawal927
deleted the
agent/posttrain-distillation-guide-20261104
branch
November 9, 2026 21:09
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The distillation guide now distinguishes logits distillation from response imitation and gives source-verified command shapes, tokenizer prerequisites, a fair-comparison checklist and a result worksheet. It replaces unsupported universal perplexity gains and guessed expected results with dated evidence and explicit limits.
Corrects the Swift comments to match the executable teacher-to-student KL expression. Executable code, dependencies, experiment state and the established docs layout are unchanged.
Validation: Astro/TypeScript, final docs merge, knowledge index and built-link checks passed. Independent source/content review and390/768/1440 headless renders passed with the expected metadata/canonical, one H1 and no document overflow. CLI examples are source-verified illustrations; no training or model run was performed.
Closes #222.