feat(sae): Triton sparse decoder, tensor-parallel training, streaming

[jwilber]

Update SAE to support larger models:

• triton kernels
• producer/consumer streaming for training
• tp

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: ac5cec1e-874f-4f1e-a322-ff808350b2db

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch jwilber/sae-infra

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[jwilber]

/ok to test 9f0f00d