feat: Add Megatron-LM based training #517
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
Co-authored-by: Sahil Jain <sahilj@nvidia.com> Co-authored-by: Yi-Fu Wu <yifu.wu@gmail.com> Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
f5f6c6a to
c086e59
Compare
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
terrykong
reviewed
Jun 14, 2025
terrykong
reviewed
Jun 14, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
terrykong
previously approved these changes
Jun 17, 2025
|
Reverted expandable_segments due to #522 |
parthchadha
reviewed
Jun 17, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com>
terrykong
previously approved these changes
Jun 17, 2025
terrykong
approved these changes
Jun 17, 2025
parthchadha
reviewed
Jun 17, 2025
| enabled: true | ||
| empty_unused_memory_level: 0 | ||
| activation_checkpointing: false | ||
| converter_type: "Qwen2ForCausalLM" |
There was a problem hiding this comment.
Why do users need to provide this? Can we infer from the model name? If not, please add a comment saying how to set this for a new model.
| enabled: false #coming soon | ||
| train_mb_tokens: ${mul:${policy.max_total_sequence_length}, ${policy.train_micro_batch_size}} | ||
| logprob_mb_tokens: ${mul:${policy.max_total_sequence_length}, ${policy.logprob_batch_size}} | ||
| algorithm: "modified_ffd" |
There was a problem hiding this comment.
What are the other options? Lets add a description of these here.
| overlap_param_gather: false | ||
| average_in_collective: true | ||
| use_custom_fsdp: false | ||
| data_parallel_sharding_strategy: "optim_grads_params" |
There was a problem hiding this comment.
What are the other options? If there are no options (or we don't expect folks to use other options), I would prefer not exposing this in yaml file.
| temperature: 1.0 | ||
| top_p: 1.0 | ||
| top_k: null | ||
| vllm_cfg: |
There was a problem hiding this comment.
Can you remove the args that are not being overridden from 1b config?
| monitor_gpus: false # If true, will monitor GPU usage and log to wandb and/or tensorboard | ||
| wandb: | ||
| project: "grpo-dev" | ||
| name: "sj_megatron_1B" |
| The logprob of input token i is specified at position i in the output logprobs tensor. | ||
| """ | ||
| no_grad = torch.no_grad() | ||
| no_grad.__enter__() |
There was a problem hiding this comment.
What's the reason to not use context manager here?
parthchadha
pushed a commit
that referenced
this pull request
Jun 17, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com> Signed-off-by: Terry Kong <terryk@nvidia.com> Co-authored-by: Yi-Fu Wu <yifu.wu@gmail.com> Co-authored-by: Terry Kong <terryk@nvidia.com> Signed-off-by: Parth Chadha <pchadha@nvidia.com>
This was
linked to
issues
Jun 24, 2025
YzjiaoNvd
pushed a commit
to YzjiaoNvd/NeMo-RL
that referenced
this pull request
Jul 14, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com> Signed-off-by: Terry Kong <terryk@nvidia.com> Co-authored-by: Yi-Fu Wu <yifu.wu@gmail.com> Co-authored-by: Terry Kong <terryk@nvidia.com>
KiddoZhu
pushed a commit
that referenced
this pull request
Jul 28, 2025
Signed-off-by: Sahil Jain <sahilj@nvidia.com> Signed-off-by: Terry Kong <terryk@nvidia.com> Co-authored-by: Yi-Fu Wu <yifu.wu@gmail.com> Co-authored-by: Terry Kong <terryk@nvidia.com>
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.
No description provided.