Repository navigation
Configure stock GPT-2 SGDM short speedrun - #171
Merged
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Updates the existing stock GPT-2 Small SGD+momentum arm to a short performance-oriented profile while leaving the architecture unchanged. SGDM uses peak LR 0.3, momentum 0.9, matrix/embedding weight decay 3e-5, 400-update linear warmup, plateau, and 1800-update linear warmdown over 6200 updates. The launcher selects this profile only for
--optimizer sgd_momentum; AdamW/Muon retain the existing llm.c profile. Regression coverage checks schedule endpoints, tied-parameter ownership, decay scope, and momentum update math.Published optimizer provenance: tml-epfl/why-weight-decay, commit dc78bb95c5418d8271cceaa11b67c6193cccb308. Short warmdown structure follows the pinned modded-nanogpt speedrun lineage; no modified architecture is imported.