Default --audit-conf to audit.config in the reference entry points

[100-JM]

Problem

Seven reference entry points default the LoadGen audit filename to audit.conf:

language/{llama3.1-8b,llama2-70b,llama3.1-405b,mixtral-8x7b,gpt-j}/main.py, language/bert/run.py, speech2text/reference_mlperf.py

but every compliance test under compliance/ ships the file as audit.config (renamed repo-wide in #2127), compliance/TEST06/README.md tells submitters to copy audit.config into the run directory, and both the loadgen Python binding (python_api.cc: audit_config_filename = "audit.config") and text_to_image/main.py already default to audit.config.

loadgen only applies the override when FileExists(audit_config_filename) (loadgen.cc:1232). So a compliance run that follows the README with one of these entry points runs as a plain performance run: no error, no Found Audit Config file line in mlperf_log_detail.txt, and a run_verification.py that then fails or verifies logs that were never produced under audit settings. We noticed this while setting up llama3.1-8b TEST06 — the README's advice to check for that log line is the only thing that exposes it.

Change

default="audit.conf" → default="audit.config" in the seven files. Passing --audit-conf explicitly keeps working; this only changes what happens when the documented procedure is followed without it. No loadgen changes.

autopep8 -a --max-line-length 79 clean on all seven files.

🤖 Generated with Claude Code

[github-actions]

MLCommons CLA bot:
Thank you very much for your submission; we really appreciate it. Before we can accept your contribution,
we ask that you sign the MLCommons CLA (Apache 2). Please submit your GitHub ID to our onboarding form to initiate
authorization. If you are from a MLCommons member organization, we will request that you be added to the CLA.
If you are not from a member organization, we will email you a CLA to sign. For any questions, please contact
support@mlcommons.org.
0 out of 1 committers have signed the MLCommons CLA.
❌ @100-JM
_{You can retrigger this bot by commenting recheck in this Pull Request}

[100-JM]

recheck