Skip to content
This repository has been archived by the owner on Aug 7, 2024. It is now read-only.

update matmul and linear benchmarks with 7B and 13B shapes #179

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

update matmul and linear benchmarks with 7B and 13B shapes #179

wants to merge 1 commit into from

Commits on Jan 11, 2024

  1. update matmul and linear benchmarks with 7B and 13B shapes 

    Summary:

as titled

also removes float16 for now since it's not super important,
and moves common constants to a reusable place

Test Plan:

```
matmul bench results:
https://gist.github.com/vkuzo/930219d6de1015dd00486c32eac8fa67

linear bench results:
https://gist.github.com/vkuzo/31a063df8d10794497949285319b5ccd
```

Reviewers:

Subscribers:

Tasks:

Tags:
    @vkuzo
    vkuzo committed Jan 11, 2024
    Configuration menu
    Copy the full SHA
    83e1c30 View commit details
    Browse the repository at this point in the history