Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix: Qwen v1.5 Tokenizer bug #107

Merged
merged 7 commits into from
Aug 2, 2024
Merged

fix: Qwen v1.5 Tokenizer bug #107

merged 7 commits into from
Aug 2, 2024

Conversation

chenghuaWang
Copy link
Contributor

  1. Two packages, re2 and absl, have been integrated into the Qwen1.5 Demo, and the compilation has passed on the ARM platform.
  2. A feature has been added to BPE where it does not split word when encountering a Special Token.

@yirongjie yirongjie self-requested a review August 1, 2024 04:28
@chenghuaWang
Copy link
Contributor Author

chenghuaWang commented Aug 1, 2024 •
edited
Loading

I tested this giant UnicodeData.cpp on 3 Compilers

  1. Clang 17.0.6 Self-build. Debug SUCCESS. Release FAILED to built mllm due to UnicodeData.cpp's const data is too large.
  2. GCC 12.3.0 Pre-build. Both SUCCESS
  3. Clang 18 provided by NDK. Both SUCCESS.

@yirongjie yirongjie merged commit 1b1c3cc into UbiquitousLearning:main Aug 2, 2024
1 check passed
@yirongjie yirongjie changed the title Fix: Qwen v1.5 Tokenizer bug fix: Qwen v1.5 Tokenizer bug Aug 16, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@yirongjie yirongjie yirongjie approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@chenghuaWang @yirongjie