Feat: add usage docs for fsdp #1092
Merged
+122
−4
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
Collaborator
|
@Hecate0821 Lets review this together! |
PopSoda2002
reviewed
Dec 12, 2025
docs/en/get_started/usage.md
Outdated
| | **Context Parallel** | `--context-parallel-size` | `--context-parallel-size` | Both support CP | | ||
| | **Initial Learning Rate** | `--lr` | `--lr` | Same parameter | | ||
| | **Learning Rate Decay** | `--lr-decay-style` (linear/cosine) | `--lr-decay-style` (only constant) | | | ||
| | **Warmup** | `--lr-warmup-iters` (steps) | Coming Soon | | |
Collaborator
There was a problem hiding this comment.
Hi I think for learning rate related stuff, it has already being supported, check this out: #1040
Hecate0821
reviewed
Dec 12, 2025
docs/en/get_started/usage.md
Outdated
| | **CPU Backend** | Implemented via distributed optimizer | `--fsdp-cpu-backend` | **FSDP**: Specify CPU backend and use hybrid backend when CPU offload is enabled | | ||
| | **Attention Backend** | Decided by Megatron Core | `--attn-implementation` (flash_attention_2/sdpa/eager) | **FSDP**: Directly passed to HuggingFace | | ||
| | **Mixed Precision** | `--fp16` or `--bf16` | `--fp16` (bf16 inferred automatically) | Basically same | | ||
| | **Offload on Save** | | `--fsdp-state-dict-cpu-offload` (Default True) | **FSDP**: Offload to CPU when saving checkpoint | |
Collaborator
There was a problem hiding this comment.
I think we don't need to mention state-dict-cpu-offload in user doc, it should always be True.
Hecate0821
approved these changes
Dec 12, 2025
PopSoda2002
reviewed
Dec 14, 2025
docs/en/get_started/usage.md
Outdated
| | **Expert Parallel** | `--expert-model-parallel-size` | Coming Soon | | | ||
| | **Context Parallel** | `--context-parallel-size` | `--context-parallel-size` | Both support CP | | ||
| | **Initial Learning Rate** | `--lr` | `--lr` | Same parameter | | ||
| | **Learning Rate Decay** | `--lr-decay-style` (linear/cosine) | `--lr-decay-style` (only constant) | | |
Collaborator
There was a problem hiding this comment.
I think for lr decay, not only constant is supported? You can checkout that PR
Fengzdadi
pushed a commit
to Fengzdadi/slime
that referenced
this pull request
Dec 19, 2025
Feat: add usage docs for fsdp
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Update the docs for lately-supported FSDP backend for slime.