Fix missing hf_eval_split option in test_convergence_1b_params.sh

[hsuan-lun-chiang]

Description

The hf_eval_split must be set when running end_to_end/tpu/test_convergence_1b_params.sh, otherwise it will throw ValueError: Bad split: . Available splits: ['train'].

Error Log

Tests

~/xpk/xpk.py workload create --cluster=bodaborg-v6e-256-lcscld-c --workload=maxtext-convergence-hf-v6e-2569caf0631 --command='set -xue;export M_RUN_NAME=maxtext-convergence-hf-v6e-256-2026-01-05-06-36-59;bash end_to_end/tpu/test_convergence_1b_params.sh OUTPUT_PATH=gs://ml-auto-solutions/output/maxtext/stable/automated/2026-01-05 DATASET_PATH=gs://max-datasets-rogue LOSS_THRESHOLD=2.7 STEPS=500 PER_DEVICE_BATCH_SIZE=2.0 DATASET_TYPE=hf' --device-type=v6e-256 --num-slices=1 --docker-image=gcr.io/tpu-prod-env-multipod/maxtext_base_image:hf --project=tpu-prod-env-one-vm --zone=southamerica-west1-a --env GCS_OUTPUT=gs://ml-auto-solutions/output/unowned/maxtext-convergence-hf-v6e-256-2026-01-05-06-37-20/ --skip-validation

Log

Checklist

Before submitting this PR, please make sure (put X in square brackets):

• I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
• I have necessary comments in my code, particularly in hard-to-understand areas.
• I have run end-to-end tests tests and provided workload links above if applicable.
• I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[gobbleturk]

Thank you for the fix!