Skip to content

Refresh BERT H100 golden values - #5953

Merged
Phlip79 merged 2 commits into
NVIDIA:mainfrom
Phlip79:philip/fix-mcore-513-golden-values
Jul 23, 2026
Merged

Refresh BERT H100 golden values#5953
Phlip79 merged 2 commits into
NVIDIA:mainfrom
Phlip79:philip/fix-mcore-513-golden-values

Conversation

@Phlip79

@Phlip79 Phlip79 commented Jul 21, 2026

Copy link
Copy Markdown
Member
  • I, the PR author, have personally reviewed every line of this PR.

What does this PR do?

Existing golden values were generated when there was a bug causing inf values. Golden values should not have been updated with inf values.

Testing

Validated run.

Signed-off-by: Philip Petrakian <ppetrakian@nvidia.com>
@copy-pr-bot

copy-pr-bot Bot commented Jul 21, 2026

Copy link
Copy Markdown

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

@Phlip79 Phlip79 changed the title Refresh BERT H100 golden values. Refresh BERT H100 golden values Jul 21, 2026
@Phlip79

Phlip79 commented Jul 21, 2026

Copy link
Copy Markdown
Member Author

/ok to test 858226e

@Phlip79

Phlip79 commented Jul 22, 2026

Copy link
Copy Markdown
Member Author

/ok to test 2edfd1c

@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/29953008159

@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/29960718599

@svcnvidia-nemo-ci

Copy link
Copy Markdown
Contributor

🔄 Merge queue validation started!

You can track the progress here: https://github.com/NVIDIA/Megatron-LM/actions/runs/29968253404

Merged via the queue into NVIDIA:main with commit 96d4159 Jul 23, 2026
89 checks passed
@Phlip79
Phlip79 deleted the philip/fix-mcore-513-golden-values branch July 23, 2026 02:20
svcnvidia-nemo-ci pushed a commit that referenced this pull request Jul 24, 2026
The linting job's new 'Validate updated golden values' step (check_golden_values.py,
added on main via #5989) rejects any non-finite value in golden-value JSON files that
differ from the base branch. Three golden files were refreshed on main (#5953 BERT
H100, #5816 MoE GB200) and taken into this sync; each carries a 'nan' at
iteration-time.values['1'] — the warmup iteration.

The functional comparison harness (python_test_utils/common.py) evaluates
iteration-time only over a steady-state window (range(5,21) or range(30,46)), so the
warmup iteration is never read. Replacing the warmup 'nan' with iteration 2's finite
value is inert for the comparison but satisfies the finiteness gate. All of main's
refreshed measured data is preserved (one line changed per file).

Co-Authored-By: Claude <noreply@anthropic.com>
Signed-off-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
svcnvidia-nemo-ci added a commit that referenced this pull request Jul 29, 2026
The main->dev merge adopts main's new golden-value validation
(tools/check_golden_values.py + the linting job's "Validate updated
golden values" step, added by #5989 / AUT-957 on 2026-07-23). That step
runs check_golden_values.py on every golden file that differs from the PR
base and rejects any NaN/Inf. The merge pulled main's versions of three
golden files that predate the check (refreshed by #5953 on 07-22 and #5816
on 07-20), so they still encode the warmup-step iteration timer as the
legacy sentinel "1": "nan".

main's own post-check convention (e.g. the MoE goldens refreshed by #5916
AFTER the check landed, which passed CI) is to omit the step-1
iteration-time cell entirely and set start_step: 2, rather than store
"nan". Eleven files already in the corpus have exactly this shape.

Bring the three stale files into that convention: in the iteration-time
block only, drop "1": "nan" and set start_step 1->2. All other metrics and
all other steps are byte-for-byte unchanged. This regresses no data:
step-1 iteration-time is never read by the functional comparator (it only
evaluates the steady-window steps 5-21 / 30-46) nor by
compare_golden_values_kl.py (which skips non-finite via math.isfinite).

Signed-off-by: svcnvidia-nemo-ci <svcnvidia-nemo-ci@nvidia.com>
shjwudp pushed a commit to shjwudp/Megatron-LM that referenced this pull request Jul 30, 2026
Signed-off-by: Philip Petrakian <ppetrakian@nvidia.com>
terminator123 pushed a commit to 021ai/Megatron-LM that referenced this pull request Aug 3, 2026
Signed-off-by: Philip Petrakian <ppetrakian@nvidia.com>
svcnvidia-nemo-ci pushed a commit to dimapihtar/Megatron-LM that referenced this pull request Aug 4, 2026
Signed-off-by: Philip Petrakian <ppetrakian@nvidia.com>
Signed-off-by: Dmytro Pykhtar <dpykhtar@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants