[mxfp8 moe training] add kernel microbenchmarks table to readme

[danielvegamyhre]

Stacked PRs:

• [mxfp8 moe training] readme updates: update table of contents, organization; remove outdated content #3567
• ->[mxfp8 moe training] add kernel microbenchmarks table to readme #3566
• [mxfp8 moe training] update roofline and bench script to use new cuda kernel #3565

---

[mxfp8 moe training] add kernel microbenchmarks table to readme

This is on NVIDIA B200 with 1000W limit. Suprisingly the % efficiency is actually better on 750W limited device, even though the absolute runtimes are a bit slower. I opted to use the 1000W numbers for the official numbers since it is the full, non-limited power supply.

Preview

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3566

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• B200 runners are down due to network issues

✅ No Failures

As of commit a614ba5 with merge base 8bd3526 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.