Skip to content

[QUESTION] Can fp8_paged_mqa_logits support next_n > 2? #222

@starwang1024

Description

@starwang1024

I noticed that the code 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.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions