feat: DTensorPolicyV2 GPT-OSS support #1470
There are no files selected for viewing
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,29 @@ | ||
| defaults: ../../sft.yaml | ||
| policy: | ||
| model_name: openai/gpt-oss-20b | ||
|
Contributor
There was a problem hiding this comment. I believe you have some plots for the convergence of gpt-oss, can you paste them to the PR? so that others can know this recipe's results. Also do you have tested other models (e.g., llama, qwen) using this PR to make sure this PR won't affect other models? there's a lot of changes in the dtensor v2 worker. |
||
| train_global_batch_size: 128 | ||
| train_micro_batch_size: 8 | ||
| max_total_sequence_length: 512 | ||
| dequantize_base_checkpoint: true | ||
| automodel_model_kwargs: | ||
| backend: | ||
| _target_: nemo_automodel.components.moe.utils.BackendConfig | ||
| attn: te | ||
| linear: te | ||
| rms_norm: te | ||
| enable_deepep: true | ||
| fake_balanced_gate: false | ||
| enable_hf_state_dict_adapter: true | ||
| dtensor_cfg: | ||
| _v2: true | ||
| expert_parallel_size: 8 | ||
| data_parallel_size: 8 | ||
| optimizer: | ||
| name: transformer_engine.pytorch.optimizers.fused_adam.FusedAdam | ||
| kwargs: | ||
| store_param_remainders: true | ||
| master_weights: true | ||
| exp_avg_dtype: bfloat16 | ||
| exp_avg_sq_dtype: bfloat16 | ||
| checkpointing: | ||
| checkpoint_dir: results/sft-gpt-oss-20b-1n8g-fsdp8ep8-automodel | ||
Oops, something went wrong.
Oops, something went wrong.
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Outdated suggestions cannot be applied.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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
can you add the nightly test for this?
you can refer to
tests/test_suites/llm/grpo-deepscaler-1.5b-8K.sh.