allenaiallenai/open-instruct/.claude/skills/run-gpu-tests/SKILL.md

run-gpu-tests

Skill3.9k starsChanged 3 days agoApache-2.0View on GitHub

Run the GPU test suite on Beaker and add its GPU_TESTS line to the PR body. Use when asked to run the GPU tests.

Create a free account to mount

Files

SKILL.md

Please launch the GPU tests manually on Beaker with:

./scripts/train/build_image_and_launch.sh scripts/test/run_gpu_pytest.sh

Monitor the experiment using the monitor-experiment skill. When the test has passed, update the PR body with the update-pr-body skill to add the experiment ID at the very end of the PR body in this format: GPU_TESTS=[EXPERIMENT_ID](https://beaker.org/ex/EXPERIMENT_ID).

Download now (ZIP)

Jan 7, 2024Code for training llama pro? — TencentARC/LLaMA-Pro2 reactions · 8 comments
Oct 12, 2026Fix sub-block KDA inverse miscompile on TPU v4 — AI-Hypercomputer/maxtext1 reaction · 1 comment
Jul 4, 2026[category] Dataset curation & processing tooling — currentai-org/os-ai-map1 reaction · 1 comment
Feb 3, 2026Add chat template verification step to SFT README — allenai/Olmo-core1 reaction
Apr 25, 2025DPO Trainer for Tulu3 Preference Dataset — swiss-ai/posttraining1 reaction · 1 comment

45 more on GitHub

Feb 22, 2025Add example of RL training using open-instruct — open-thought/reasoning-gym1 reaction · 3 comments
Jan 23, 2025Support Verifiable Rewards like TULU3 — OpenRLHF/OpenRLHF1 reaction · 2 comments
Jul 23, 2024Llama 3 - RuntimeError: shape '[-1, 0]' is invalid for input of size 41041920 — huggingface/transformers1 reaction · 11 comments
Nov 23, 2023How to conduct alpaca eval? — Lightning-AI/litgpt1 reaction
Oct 27, 2026Skip Core optimizer batches with no active loss tokens — allenai/miles-olmo-core0 reactions
Oct 19, 2026Add NGU retries and rejected-group handling for fully async rollouts — allenai/miles-olmo-core0 reactions
Oct 18, 2026Review candidate: github.com/allenai/open-instruct — reacher-z/Awesome-CUA0 reactions · 1 comment
Oct 8, 2026[Integration] allenai/open-instruct: You.com search tool back in the tool registry (you_search) — youdotcom-oss/integration-tracking0 reactions · 4 comments
Sep 18, 2026Hero B300 grouped-expert profiling fails before evaluation — allenai/olmo-eval0 reactions
Aug 12, 2026Keep checkpoint cleanup asynchronous — allenai/Olmo-core0 reactions
May 30, 2026[Dec 2025] Olmo 3 — T41K1/papers0 reactions
May 10, 2026Math verifierをOpen Instruct寄せ + SymPy Lark backendで検証する — T41K1/LLM-jp-4-RL0 reactions
Apr 7, 2026Add BigCode Evaluation Harness and EvalPlus to Category 9 — alvinreal/awesome-opensource-ai0 reactions · 1 comment
Feb 28, 2026rlvr training example scripts를 추가한다. — seopbo/trl-harness0 reactions · 1 comment
Feb 27, 2026adds simple smoke tests — allenai/olmo-eval0 reactions · 2 comments
Feb 3, 2026Add chat template verification step to SFT README — allenai/Olmo-core0 reactions
Jan 8, 2026[Task/Feature/Issue] Investigating tuning techniques — The-AI-Alliance/ai-application-testing0 reactions
Oct 16, 2025[FSDP][ready for review] Reduce peak memory required when gathering log-probs — THUDM/slime0 reactions
Oct 6, 2025Paper Review: Tulu 3: Pushing Frontiers in Open Language Model Post-Training — bits-bytes-nn/bits-bytes-nn.github.io0 reactions
Oct 5, 2025Paper Review: Tulu 3: Pushing Frontiers in Open Language Model Post-Training — bits-bytes-nn/bits-bytes-nn.github.io0 reactions
Sep 9, 2025[Paper Note] Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback, Hamish Ivison+, NeurIPS'24 — AkihikoWatanabe/paper_notes0 reactions · 3 comments
Jul 26, 2025remove_non_tensor_columns — garrett361/open-instruct0 reactions
May 28, 2025Latest 15 Papers - May 29, 2025 — jyyang621/DailyArXiv0 reactions
May 27, 2025Latest 15 Papers - May 28, 2025 — jyyang621/DailyArXiv0 reactions
May 26, 2025Latest 15 Papers - May 27, 2025 — jyyang621/DailyArXiv0 reactions
May 25, 2025Latest 15 Papers - May 26, 2025 — jyyang621/DailyArXiv0 reactions
May 22, 2025Latest 15 Papers - May 23, 2025 — jyyang621/DailyArXiv0 reactions
May 21, 2025Latest 15 Papers - May 22, 2025 — jyyang621/DailyArXiv0 reactions
May 20, 2025Latest 15 Papers - May 21, 2025 — jyyang621/DailyArXiv0 reactions
May 19, 2025Latest 15 Papers - May 20, 2025 — jyyang621/DailyArXiv0 reactions
May 18, 2025Latest 15 Papers - May 19, 2025 — jyyang621/DailyArXiv0 reactions
May 15, 2025Latest 15 Papers - May 16, 2025 — jyyang621/DailyArXiv0 reactions
May 14, 2025Latest 15 Papers - May 15, 2025 — jyyang621/DailyArXiv0 reactions
May 13, 2025Latest 15 Papers - May 14, 2025 — jyyang621/DailyArXiv0 reactions
May 12, 2025Latest 15 Papers - May 13, 2025 — jyyang621/DailyArXiv0 reactions
May 11, 2025Latest 15 Papers - May 12, 2025 — jyyang621/DailyArXiv0 reactions
May 8, 2025Latest 15 Papers - May 09, 2025 — jyyang621/DailyArXiv0 reactions
May 7, 2025Latest 15 Papers - May 08, 2025 — jyyang621/DailyArXiv0 reactions
May 6, 2025Latest 15 Papers - May 07, 2025 — jyyang621/DailyArXiv0 reactions
May 5, 2025Latest 15 Papers - May 06, 2025 — jyyang621/DailyArXiv0 reactions · 1 comment
May 4, 2025Latest 15 Papers - May 05, 2025 — jyyang621/DailyArXiv0 reactions
May 1, 2025Latest 15 Papers - May 02, 2025 — jyyang621/DailyArXiv0 reactions
Apr 30, 2025Latest 15 Papers - May 01, 2025 — jyyang621/DailyArXiv0 reactions
Apr 29, 2025Latest 15 Papers - April 30, 2025 — jyyang621/DailyArXiv0 reactions
Mar 28, 2025Dataset mixers — swiss-ai/posttraining0 reactions · 1 comment

On Bluesky · 6

Nov 24, 2024Reading the TÜLU 3 paper from @ai2.bsky.social. It's refreshing to see a research lab treating AI as a real science with full reports,… — @vivekkalyan.com30 likes and reposts · 1 reply
Nov 28, 2024repo: github.com/allenai/open... — @natolambert.bsky.social12 likes and reposts · 1 reply
Dec 6, 2024New OpenAI RL finetuning API reminds me a lot of RLVR, which we used for Tülu 3 (arxiv.org/abs/2411.15124). Using RL to train against… — @hamishivi.bsky.social8 likes and reposts · 1 reply
Nov 21, 2024Training data: huggingface.co/collections/... Training code: github.com/allenai/open... Eval code: github.com/allenai/olmes — @ai2.bsky.social6 likes and reposts
Nov 26, 2024📦 allenai / open-instruct ⭐ 1,608 (+88) 🗒 Python — @github-trending.bsky.social1 likes and repost

1 more on Bluesky

Nov 21, 2024Models: huggingface.co/collections/... Training data: huggingface.co/collections/... Paper: allenai.org/papers/tulu-... Blog:… — @jacobcares.bsky.social0 likes and reposts · 1 reply

On Hacker News · 1

Nov 26, 2024Training Open Instruction-Following Language Models1 point

In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- monitor-experimentSkill
- run-all-dpoSkill
- run-and-fixSkill
- tpuSkill
- update-pr-bodySkill

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free