pytorch / pytorch/TensorRT

🐛 [Bug] Failed to fast refit Llama2-7b with `STRIP_PLAN` + `REFIT`

Open
#3,255 2 comments 0 reactions 1 assignee View on GitHub

@zewenli98 is already working on this.

Since Nov 6, 2024.

bug Story: Runtime & Memory & Serialization
Dominant language
Python
Stars
3k
Forks
410
Avg merge
3d 18h
Merged PRs (30d)
78

Description

Bug Description

If compile llama2-7b with STRIP_PLAN + REFIT and then fast refit (i.e., refit with weight_name_map), the outputs are all zeros. However, slow refit works well.
If compile llama2-7b with STRIP_PLAN + REFIT_IDENTICAL and then fast refit, the outputs are correct.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.