Add mock SFT dataset and generalize SFT padding divisor - #5906
Closed
ilml wants to merge 2 commits into
Closed
Conversation
This was referenced Jul 20, 2026
ilml
marked this pull request as ready for review
July 22, 2026 01:02
ilml
force-pushed
the
split/3386-06-mock-sft-dataset
branch
3 times, most recently
from
July 30, 2026 21:19
93b49e3 to
7a3fbd4
Compare
ilml
force-pushed
the
split/3386-06-mock-sft-dataset
branch
2 times, most recently
from
August 6, 2026 16:44
f0cb3a5 to
e54e802
Compare
ilml
force-pushed
the
split/3386-06-mock-sft-dataset
branch
2 times, most recently
from
August 10, 2026 22:09
a65f451 to
7169c46
Compare
Split 2/10 from NVIDIA#3386 (sequence packing / THD E2E support). Adds the sequence_packing_scheduler knob to ModelParallelConfig, its TransformerConfig validation (TE>=2.9 pin, variable_seq_lengths, alltoall dispatcher), the explicit CLI args, and validate_args checks. Feature-flagged no-op until the scheduler lands. Original changes by @xiaoyao0115 in NVIDIA#3386. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Signed-off-by: ilml <tolong@nvidia.com>
Split 6/10 from NVIDIA#3386 (sequence packing / THD E2E support). Adds MockSFTDataset/MockSFTLowLevelDataset with file- and distribution-mode sequence-length configs, the load_json_arg helper, the --sft-mock-dataset-config-json arg with lognormal default, and generalizes the SFT padding divisor beyond cp>1. Original changes by @xiaoyao0115 in NVIDIA#3386. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Signed-off-by: ilml <tolong@nvidia.com>
6 tasks
Contributor
Author
|
/ok to test 7169c46 |
This was referenced Aug 18, 2026
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
2 participants
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.
What does this PR do?
Adds
MockSFTDataset/MockSFTLowLevelDatasetwith file- and distribution-mode sequence-length configs, theload_json_arghelper,--sft-mock-dataset-config-json(with a lognormal default when--mock-data --sft), and generalizes the SFT padding divisor beyond the previous cp>1-only path.Part 06/10 of splitting #3386 (Add E2E support for THD format; dev-branch PR #2924). Original changes by @xiaoyao0115 in #3386 — split into functionally self-contained PRs to ease review. Hard dependencies (must merge first): #5902, #5903.
Split series (#3386)
Branches are stacked linearly (each on the previous) so every PR shows a clean own-diff once its base is retargeted to the copy-pr-bot
pull-request/<parent>ref; until then the Files-changed view of a stacked PR includes its ancestors — its own change is the last commit.Stacked on #5905.
Review notes
import pandas as pdinsft_dataset.pymakespretrain_gpt.pyhard-require pandas (currently only aciextra) — consider lazy-guarding likevarlen_dataset.pydoes.math/Unionimports insft_dataset.pyare unused.sequence_parallel_size>0or hybrid CP is on.MockSFTDatasethas no dedicated unit test until the varlen PRs exercise it viaMockVarlenDataset.Issue tracking
Linked issue: Related to #3386
Contribution process
Pre-checks
🤖 Generated with Claude Code