cannot run qwen3-30B both with ‘router replay’ and 'recompute' on Megatron
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 8.5k
- Forks
- 1.3k
- Avg merge
- 5h 36m
- Merged PRs (30d)
- 22
Description
with task config:
"--use-routing-replay"
"--recompute-granularity full "
"--recompute-method uniform "
"--recompute-num-layers 1 "
there is some errors:
IndexError: list index out of range
I'm not sure if recompute settings affects routing-replay？
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Reproduce the failure with the listed task configuration, including --use-routing-replay and the recompute options. Start from the full IndexError traceback and trace the interaction between routing replay and recomputation; done means the configuration runs without the error or reports a clear incompatibility.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100