maxtext
maxtext copied to clipboard
Add deepseek c4 convergence test and v5p recipe
Description
- to support different c4 variance version in c4_mperf datatype
- add deepseek v3 convergence configs.
- add deepseek v5p recipe
Tests
validate #2 convergence run and #3 configs
Checklist
Before submitting this PR, please make sure (put X in square brackets):
- [x] I have performed a self-review of my code.
- [x] I have necessary comments in my code, particularly in hard-to-understand areas.
- [x] I have run end-to-end tests tests and provided workload links above if applicable.
- [x] I have made or will make corresponding changes to the doc if needed.
will waiting for PR approval until merge, but need to tag 'pull ready' to unblock internal testing
Oh, one more thing, don't forget to squash commits into 1 commit for clean history. Thank you!