Orchestra-Research/AI-Research-SKILLs / 06-post-training/grpo-rl-training
grpo-rl-training
VerifiedExpert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 3.4 KB |
| SKILL.md | skill md | 16.8 KB |
| examples/reward_functions_library.py | script | 11.3 KB |
| templates/basic_grpo_training.py | script | 6.0 KB |
README.md
View on GitHub# GRPO/RL Training Skill
**Expert-level guidance for Group Relative Policy Optimization with TRL**
## π Skill Structure
```
grpo-rl-training/
βββ SKILL.md # Main skill documentation (READ THIS FIRST)
βββ README.md # This file
βββ templates/
β βββ basic_grpo_training.py # Production-ready training template
βββ examples/
βββ reward_functions_library.py # 20+ reward function examples
```
## π Quick Start
1. **Read SKILL.md** - Comprehensive guide with all concepts and patterns
2. **Copy `templates/basic_grpo_training.py`** - Start with working code
3. **Browse `examples/reward_functions_library.py`** - Pick reward functions for your task
4. **Modify for your use case** - Adapt dataset, rewards, and config
## π‘ What's Inside
### SKILL.md (Main Documentation)
- Core GRPO concepts and algorithm fundamentals
- Complete implementation workflow (dataset β rewards β training β deployment)
- 10+ reward function examples with code
- Hyperparameter tuning guide
- Training insights (loss behavior, metrics, debugging)
- Troubleshooting guide
- Production best practices
### Templates
- **basic_grpo_training.py**: Minimal, production-ready training script
- Uses Qwen 2.5 1.5B Instruct
- 3 reward functions (format + correctness)
- LoRA for efficient training
- Fully documented and ready to run
### Examples
- **reward_functions_library.py**: 20+ battle-tested reward functions
- Correctness rewards (exact match, fuzzy match, numeric, code execution)
- Format rewards (XML, JSON, strict/soft)
- Length rewards (ideal length, min/max)
- Style rewards (reasoning quality, citations, repetition penalty)
- Combined rewards (multi-objective optimization)
- Preset collections for common tasks
## π Usage for Agents
When this skill is loaded in your agent's context:
1. **Always read SKILL.md first** before implementing
2. **Start simple** - Use length-based reward to validate setup
3. **Build incrementally** - Add one reward function at a time
4. **Reference examples** - Copy patterns from reward_functions_library.py
5. **Monitor training** - Watch reward metrics (not loss!)
## π― Common Use Cases
| Task Type | Recommended Rewards | Template |
|-----------|---------------------|----------|
| Math reasoning | `MATH_REASONING_REWARDS` preset | basic_grpo_training.py |
| Code generation | `CODE_GENERATION_REWARDS` preset | Modify dataset in template |
| Summarization | `SUMMARIZATION_REWARDS` preset | Adjust prompts + rewards |
| Q&A | `QA_REWARDS` preset | Use fuzzy match + citations |
## β οΈ Critical Reminders
- **Loss goes UP during training** - This is normal (it's KL divergence)
- **Use 3-5 reward functions** - Single rewards often fail
- **Test rewards before training** - Debug each function independently
- **Monitor reward_std** - Should stay > 0.1 (avoid mode collapse)
- **Start with num_generations=4-8** - Scale up if GPU allows
## π External Resources
- [TRL Documentation](https://huggingface.co/docs/trl)
- [DeepSeek R1 Paper](https://arxiv.org/abs/2501.12948)
- [Open R1 Implementation](https://github.com/huggingface/open-r1)
- [Unsloth (2-3x faster)](https://docs.unsloth.ai/)
## π Version
**v1.0.0** - Initial release (January 2025)
## π¨βπ» Maintained By
Orchestra Research
For questions or improvements, see https://orchestra.com
---
**License:** MIT
**Last Updated:** January 2025
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |