DeepSeek推理仿真问题
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.2k
- Forks
- 184
- Avg merge
- 13h 4m
- Merged PRs (30d)
- 1
Description
您好,SimAI在描述中说明已经可以适配DeepSeek-V3-671B的仿真,对此,我有三点问题希望能帮忙解答一下,谢谢
(1)在推理仿真中,SimAI说明对于MoE架构的仿真支持还在进行中,并未实现。但是DeepSeek模型是MoE架构的,那请问SimAI是如何仿真DeepSeek的呢,特别是在推理仿真方面?
(2)我在推理仿真中按照Run Llama-3-8B with simai_simulation的步骤运行实验没有问题,但是把--replica_config_model_name改成deepseek后,运行会报错《No such file or directory: './data/profiling/compute/a100/deepseek-671B/mlp.csv'》,请问是否是因为缺乏deepseek的实测数据,那是否可以通过实测补充deepseek的a100数据后进行仿真,还是说目前simai对于deepseek的推理仿真只支持aicb的后端?我目前想要进行完整的deepseek的推理仿真,请问是否有更具体的运行步骤教程。
(3)关于推理仿真的workload是否有更具体的说明,在目前的README似乎没有看到具体关于workload的选择和说明。
感谢解答
Contributor guide
No contributing guide indexed for this repository
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
Start with the README section for “Run Llama-3-8B with simai_simulation” and compare its command with the reported DeepSeek configuration. Inspect the referenced ./data/profiling/compute/a100/deepseek-671B/mlp.csv path and document the supported DeepSeek inference backends, required profiling data, run steps, and workload choices.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- documentation, machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100