Skip to content

Question about Figure 2 in PyramidKV paper? #61

@dsj96

Description

@dsj96

Thank you for your research work and code!I'm a bit confused about Figure 2 in the Pyramid KV paper.

Figure 2 shows the attention patterns of one QA example over six different layers.

What I am puzzled about is what exactly this example is and how was it chosen? Does this attention patterns also exist in other samples?

Thanks.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions