tmac.liang
tmac.liang
user: 最后一行的用户被忽略; crontab:不规范的空格导致panic listen:进程名为空时导致panic
**Describe the bug** we find cudaMemcpyAsync run too much time on torch profiler: 1. on aten::_local_scalar_dense  2. on _has_inf_or_nan  **Expected behavior** reduce cudaMemcpyAsync op **ds_report output** ``` [2023-12-25...
HI,Is there plan to upgrade megatron-lm,it can be support more feature and better performance
Long-text scenarios are quite common, and it would be of great help if they could be supported.
Hi,I noticed that you've been running benchmarks on the L20. May I ask if there are targeted optimizations for the ada architecture?
I think that LLM (Large Language Model) has entered a new stage, and the demand for multimodal and reinforcement learning has increased.