deepseek-ai / deepseek-ai/DeepGEMM
[QUESTION] Can fp8_paged_mqa_logits support next_n > 2?
Open
- Dominant language
- Cuda
- Stars
- 7.8k
- Forks
- 1.3k
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 3
Description
I noticed that the [code](https://github.com/deepseek-ai/DeepGEMM/blob/main/csrc/apis/attention.hpp#L166) here in `fp8_paged_mqa_logits` only supports next_n <= 2. May I ask what the reason for this is? Additionally, what modifications would be needed to support next_n > 2?
Any reply would be very helpful.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.