GRPO & SFT E2E tests #2497
Open
GRPO & SFT E2E tests #2497
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mydatascience
requested review from
gobbleturk,
khatwanimohit,
parambole and
shralex
as code owners
mydatascience
requested review from
A9isha,
NuojCheng,
RissyRan,
SurbhiJainUSC,
aireenmei,
bvandermoon,
gagika,
hengtaoguo,
jiangjy1982,
richjames0,
suexu1025 and
vipannalla
as code owners
mydatascience
force-pushed
the
grpo_sft_tests
branch
2 times, most recently
from
ebbe014 to
480c7e4
Compare
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
mydatascience
force-pushed
the
grpo_sft_tests
branch
2 times, most recently
from
8a115f5 to
c9f10d2
Compare
Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Need scan layers Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Fix Signed-off-by: Vladimir Suvorov <[email protected]> Add sft test as well Signed-off-by: Vladimir Suvorov <[email protected]> Add sft test as well Signed-off-by: Vladimir Suvorov <[email protected]> Add sft test as well Signed-off-by: Vladimir Suvorov <[email protected]> Fix vllm Signed-off-by: Vladimir Suvorov <[email protected]> fix Signed-off-by: Vladimir Suvorov <[email protected]>
There was a problem hiding this comment.
Why do you need a separate file for E2E tests?
You can also run python3 -m MaxText.sft.sft_trainer MaxText/configs/sft.yml \ run_name=$RUN_NAME base_output_directory=$OUTPUT_PATH \ model_name=$MODEL_NAME load_parameters_path=$MODEL_CHECKPOINT_PATH \ hf_access_token=$HF_ACCESS_TOKEN tokenizer_path=$TOKENIZER_PATH"
| XLA_PYTHON_CLIENT_MEM_FRACTION: 0.75 | ||
| TF_FORCE_GPU_ALLOW_GROWTH: false | ||
| HF_TOKEN: ${{ secrets.HF_TOKEN }} | ||
| MAXTEXT_CHECKPOINT_PATH: gs://maxtext-model-checkpoints/llama3.1-8b/2025-01-23-19-04/scanned/0/items |
There was a problem hiding this comment.
Use the instruct checkpoint: gs://maxtext-model-checkpoints/llama3.1_8b_instruct/2025-10-16/scanned/0/items
| XLA_PYTHON_CLIENT_MEM_FRACTION: 0.75 | ||
| TF_FORCE_GPU_ALLOW_GROWTH: false | ||
| HF_TOKEN: ${{ secrets.HF_TOKEN }} | ||
| MODEL_CHECKPOINT_PATH: gs://maxtext-model-checkpoints/llama3.1-8b/2025-01-23-19-04/scanned/0/items |
There was a problem hiding this comment.
Instruct checkpoint: gs://maxtext-model-checkpoints/llama3.1_8b_instruct/2025-10-16/scanned/0/items
| python3 -m pip install -e . --no-dependencies | ||
| - name: Install Tunix vLLM Requirements | ||
| run: | | ||
| bash src/MaxText/examples/install_tunix_vllm_requirement.sh |
There was a problem hiding this comment.
We don't need vllm or tpu-commons to run SFT.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Add tests for e2e grpo & sft
Tests
Added a separate TPU test for grpo
Checklist
Before submitting this PR, please make sure (put X in square brackets):
gemini-reviewlabel.