[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M74FFVY5B6WFRC84R1NF2MX6":3,"skill-same:01M74FFVY5B6WFRC84R1NF2MX6":98},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":13,"repositoryHtmlUrl":53,"repositoryHomepageUrl":6,"ownerHtmlUrl":54,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":55,"install":56,"resources":69},"01M74FFVY5B6WFRC84R1NF2MX6","fine-tuning-with-trl",null,"TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.","lzy18001500226\u002FMoSim","lzy18001500226","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F152623587?v=4","01M71ZMEX36NK05B642131DEEK","01M71ZMEXSMX72KPVNJ572G0CM",0,"MIT","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: fine-tuning-with-trl\ndescription: \"TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.\"\nversion: 1.0.0\nauthor: Orchestra Research\nlicense: MIT\ndependencies: [trl, transformers, datasets, peft, accelerate, torch]\nplatforms: [linux, macos, windows]\nmetadata:\n  hermes:\n    tags: [Post-Training, TRL, Reinforcement Learning, Fine-Tuning, SFT, DPO, PPO, GRPO, RLHF, Preference Alignment, HuggingFace]\n\n---\n\n# TRL - Transformer Reinforcement Learning\n\n## Quick start\n\nTRL provides post-training methods for aligning language models with human preferences.\n\n**Installation**:\n```bash\npip install trl transformers datasets peft accelerate\n```\n\n**Supervised Fine-Tuning** (instruction tuning):\n```python\nfrom trl import SFTTrainer\n\ntrainer = SFTTrainer(\n    model=\"Qwen\u002FQwen2.5-0.5B\",\n    train_dataset=dataset,  # Prompt-completion pairs\n)\ntrainer.train()\n```\n\n**DPO** (align with preferences):\n```python\nfrom trl import DPOTrainer, DPOConfig\n\nconfig = DPOConfig(output_dir=\"model-dpo\", beta=0.1)\ntrainer = DPOTrainer(\n    model=model,\n    args=config,\n    train_dataset=preference_dataset,  # chosen\u002Frejected pairs\n    processing_class=tokenizer\n)\ntrainer.train()\n```\n\n## Common workflows\n\n### Workflow 1: Full RLHF pipeline (SFT → Reward Model → PPO)\n\nComplete pipeline from base model to human-aligned model.\n\nCopy this checklist:\n\n```\nRLHF Training:\n- [ ] Step 1: Supervised fine-tuning (SFT)\n- [ ] Step 2: Train reward model\n- [ ] Step 3: PPO reinforcement learning\n- [ ] Step 4: Evaluate aligned model\n```\n\n**Step 1: Supervised fine-tuning**\n\nTrain base model on instruction-following data:\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom trl import SFTTrainer, SFTConfig\nfrom datasets import load_dataset\n\n# Load model\nmodel = AutoModelForCausalLM.from_pretrained(\"Qwen\u002FQwen2.5-0.5B\")\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen\u002FQwen2.5-0.5B\")\n\n# Load instruction dataset\ndataset = load_dataset(\"trl-lib\u002FCapybara\", split=\"train\")\n\n# Configure training\ntraining_args = SFTConfig(\n    output_dir=\"Qwen2.5-0.5B-SFT\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=2e-5,\n    logging_steps=10,\n    save_strategy=\"epoch\"\n)\n\n# Train\ntrainer = SFTTrainer(\n    model=model,\n    args=training_args,\n    train_dataset=dataset,\n    tokenizer=tokenizer\n)\ntrainer.train()\ntrainer.save_model()\n```\n\n**Step 2: Train reward model**\n\nTrain model to predict human preferences:\n\n```python\nfrom transformers import AutoModelForSequenceClassification\nfrom trl import RewardTrainer, RewardConfig\n\n# Load SFT model as base\nmodel = AutoModelForSequenceClassification.from_pretrained(\n    \"Qwen2.5-0.5B-SFT\",\n    num_labels=1  # Single reward score\n)\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen2.5-0.5B-SFT\")\n\n# Load preference data (chosen\u002Frejected pairs)\ndataset = load_dataset(\"trl-lib\u002Fultrafeedback_binarized\", split=\"train\")\n\n# Configure training\ntraining_args = RewardConfig(\n    output_dir=\"Qwen2.5-0.5B-Reward\",\n    per_device_train_batch_size=2,\n    num_train_epochs=1,\n    learning_rate=1e-5\n)\n\n# Train reward model\ntrainer = RewardTrainer(\n    model=model,\n    args=training_args,\n    processing_class=tokenizer,\n    train_dataset=dataset\n)\ntrainer.train()\ntrainer.save_model()\n```\n\n**Step 3: PPO reinforcement learning**\n\nOptimize policy using reward model:\n\n```bash\npython -m trl.scripts.ppo \\\n    --model_name_or_path Qwen2.5-0.5B-SFT \\\n    --reward_model_path Qwen2.5-0.5B-Reward \\\n    --dataset_name trl-internal-testing\u002Fdescriptiveness-sentiment-trl-style \\\n    --output_dir Qwen2.5-0.5B-PPO \\\n    --learning_rate 3e-6 \\\n    --per_device_train_batch_size 64 \\\n    --total_episodes 10000\n```\n\n**Step 4: Evaluate**\n\n```python\nfrom transformers import pipeline\n\n# Load aligned model\ngenerator = pipeline(\"text-generation\", model=\"Qwen2.5-0.5B-PPO\")\n\n# Test\nprompt = \"Explain quantum computing to a 10-year-old\"\noutput = generator(prompt, max_length=200)[0][\"generated_text\"]\nprint(output)\n```\n\n### Workflow 2: Simple preference alignment with DPO\n\nAlign model with preferences without reward model.\n\nCopy this checklist:\n\n```\nDPO Training:\n- [ ] Step 1: Prepare preference dataset\n- [ ] Step 2: Configure DPO\n- [ ] Step 3: Train with DPOTrainer\n- [ ] Step 4: Evaluate alignment\n```\n\n**Step 1: Prepare preference dataset**\n\nDataset format:\n```json\n{\n  \"prompt\": \"What is the capital of France?\",\n  \"chosen\": \"The capital of France is Paris.\",\n  \"rejected\": \"I don't know.\"\n}\n```\n\nLoad dataset:\n```python\nfrom datasets import load_dataset\n\ndataset = load_dataset(\"trl-lib\u002Fultrafeedback_binarized\", split=\"train\")\n# Or load your own\n# dataset = load_dataset(\"json\", data_files=\"preferences.json\")\n```\n\n**Step 2: Configure DPO**\n\n```python\nfrom trl import DPOConfig\n\nconfig = DPOConfig(\n    output_dir=\"Qwen2.5-0.5B-DPO\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=5e-7,\n    beta=0.1,  # KL penalty strength\n    max_prompt_length=512,\n    max_length=1024,\n    logging_steps=10\n)\n```\n\n**Step 3: Train with DPOTrainer**\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom trl import DPOTrainer\n\nmodel = AutoModelForCausalLM.from_pretrained(\"Qwen\u002FQwen2.5-0.5B-Instruct\")\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen\u002FQwen2.5-0.5B-Instruct\")\n\ntrainer = DPOTrainer(\n    model=model,\n    args=config,\n    train_dataset=dataset,\n    processing_class=tokenizer\n)\n\ntrainer.train()\ntrainer.save_model()\n```\n\n**CLI alternative**:\n```bash\ntrl dpo \\\n    --model_name_or_path Qwen\u002FQwen2.5-0.5B-Instruct \\\n    --dataset_name argilla\u002FCapybara-Preferences \\\n    --output_dir Qwen2.5-0.5B-DPO \\\n    --per_device_train_batch_size 4 \\\n    --learning_rate 5e-7 \\\n    --beta 0.1\n```\n\n### Workflow 3: Memory-efficient online RL with GRPO\n\nTrain with reinforcement learning using minimal memory.\n\nFor in-depth GRPO guidance — reward function design, critical training insights (loss behavior, mode collapse, tuning), and advanced multi-stage patterns — see **[references\u002Fgrpo-training.md](references\u002Fgrpo-training.md)**. A production-ready training script is in **[templates\u002Fbasic_grpo_training.py](templates\u002Fbasic_grpo_training.py)**.\n\nCopy this checklist:\n\n```\nGRPO Training:\n- [ ] Step 1: Define reward function\n- [ ] Step 2: Configure GRPO\n- [ ] Step 3: Train with GRPOTrainer\n```\n\n**Step 1: Define reward function**\n\n```python\ndef reward_function(completions, **kwargs):\n    \"\"\"\n    Compute rewards for completions.\n\n    Args:\n        completions: List of generated texts\n\n    Returns:\n        List of reward scores (floats)\n    \"\"\"\n    rewards = []\n    for completion in completions:\n        # Example: reward based on length and unique words\n        score = len(completion.split())  # Favor longer responses\n        score += len(set(completion.lower().split()))  # Reward unique words\n        rewards.append(score)\n    return rewards\n```\n\nOr use a reward model:\n```python\nfrom transformers import pipeline\n\nreward_model = pipeline(\"text-classification\", model=\"reward-model-path\")\n\ndef reward_from_model(completions, prompts, **kwargs):\n    # Combine prompt + completion\n    full_texts = [p + c for p, c in zip(prompts, completions)]\n    # Get reward scores\n    results = reward_model(full_texts)\n    return [r[\"score\"] for r in results]\n```\n\n**Step 2: Configure GRPO**\n\n```python\nfrom trl import GRPOConfig\n\nconfig = GRPOConfig(\n    output_dir=\"Qwen2-GRPO\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=1e-5,\n    num_generations=4,  # Generate 4 completions per prompt\n    max_new_tokens=128\n)\n```\n\n**Step 3: Train with GRPOTrainer**\n\n```python\nfrom datasets import load_dataset\nfrom trl import GRPOTrainer\n\n# Load prompt-only dataset\ndataset = load_dataset(\"trl-lib\u002Ftldr\", split=\"train\")\n\ntrainer = GRPOTrainer(\n    model=\"Qwen\u002FQwen2-0.5B-Instruct\",\n    reward_funcs=reward_function,  # Your reward function\n    args=config,\n    train_dataset=dataset\n)\n\ntrainer.train()\n```\n\n**CLI**:\n```bash\ntrl grpo \\\n    --model_name_or_path Qwen\u002FQwen2-0.5B-Instruct \\\n    --dataset_name trl-lib\u002Ftldr \\\n    --output_dir Qwen2-GRPO \\\n    --num_generations 4\n```\n\n## When to use vs alternatives\n\n**Use TRL when:**\n- Need to align model with human preferences\n- Have preference data (chosen\u002Frejected pairs)\n- Want to use reinforcement learning (PPO, GRPO)\n- Need reward model training\n- Doing RLHF (full pipeline)\n\n**Method selection**:\n- **SFT**: Have prompt-completion pairs, want basic instruction following\n- **DPO**: Have preferences, want simple alignment (no reward model needed)\n- **PPO**: Have reward model, need maximum control over RL\n- **GRPO**: Memory-constrained, want online RL\n- **Reward Model**: Building RLHF pipeline, need to score generations\n\n**Use alternatives instead:**\n- **HuggingFace Trainer**: Basic fine-tuning without RL\n- **Axolotl**: YAML-based training configuration\n- **LitGPT**: Educational, minimal fine-tuning\n- **Unsloth**: Fast LoRA training\n\n## Common issues\n\n**Issue: OOM during DPO training**\n\nReduce batch size and sequence length:\n```python\nconfig = DPOConfig(\n    per_device_train_batch_size=1,  # Reduce from 4\n    max_length=512,  # Reduce from 1024\n    gradient_accumulation_steps=8  # Maintain effective batch\n)\n```\n\nOr use gradient checkpointing:\n```python\nmodel.gradient_checkpointing_enable()\n```\n\n**Issue: Poor alignment quality**\n\nTune beta parameter:\n```python\n# Higher beta = more conservative (stays closer to reference)\nconfig = DPOConfig(beta=0.5)  # Default 0.1\n\n# Lower beta = more aggressive alignment\nconfig = DPOConfig(beta=0.01)\n```\n\n**Issue: Reward model not learning**\n\nCheck loss type and learning rate:\n```python\nconfig = RewardConfig(\n    learning_rate=1e-5,  # Try different LR\n    num_train_epochs=3  # Train longer\n)\n```\n\nEnsure preference dataset has clear winners:\n```python\n# Verify dataset\nprint(dataset[0])\n# Should have clear chosen > rejected\n```\n\n**Issue: PPO training unstable**\n\nAdjust KL coefficient:\n```python\nconfig = PPOConfig(\n    kl_coef=0.1,  # Increase from 0.05\n    cliprange=0.1  # Reduce from 0.2\n)\n```\n\n## Advanced topics\n\n**SFT training guide**: See [references\u002Fsft-training.md](references\u002Fsft-training.md) for dataset formats, chat templates, packing strategies, and multi-GPU training.\n\n**DPO variants**: See [references\u002Fdpo-variants.md](references\u002Fdpo-variants.md) for IPO, cDPO, RPO, and other DPO loss functions with recommended hyperparameters.\n\n**Reward modeling**: See [references\u002Freward-modeling.md](references\u002Freward-modeling.md) for outcome vs process rewards, Bradley-Terry loss, and reward model evaluation.\n\n**Online RL methods**: See [references\u002Fonline-rl.md](references\u002Fonline-rl.md) for PPO, GRPO, RLOO, and OnlineDPO with detailed configurations.\n\n**GRPO deep dive**: See [references\u002Fgrpo-training.md](references\u002Fgrpo-training.md) for expert-level GRPO patterns — reward function design philosophy, training insights (why loss increases, mode collapse detection), hyperparameter tuning, multi-stage training, and troubleshooting. Production-ready template in [templates\u002Fbasic_grpo_training.py](templates\u002Fbasic_grpo_training.py).\n\n## Hardware requirements\n\n- **GPU**: NVIDIA (CUDA required)\n- **VRAM**: Depends on model and method\n  - SFT 7B: 16GB (with LoRA)\n  - DPO 7B: 24GB (stores reference model)\n  - PPO 7B: 40GB (policy + reward model)\n  - GRPO 7B: 24GB (more memory efficient)\n- **Multi-GPU**: Supported via `accelerate`\n- **Mixed precision**: BF16 recommended (A100\u002FH100)\n\n**Memory optimization**:\n- Use LoRA\u002FQLoRA for all methods\n- Enable gradient checkpointing\n- Use smaller batch sizes with gradient accumulation\n\n## Resources\n\n- Docs: https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftrl\u002F\n- GitHub: https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl\n- Papers:\n  - \"Training language models to follow instructions with human feedback\" (InstructGPT, 2022)\n  - \"Direct Preference Optimization: Your Language Model is Secretly a Reward Model\" (DPO, 2023)\n  - \"Group Relative Policy Optimization\" (GRPO, 2024)\n- Examples: https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl\u002Ftree\u002Fmain\u002Fexamples\u002Fscripts\n\n\n\n","# TRL - Transformer Reinforcement Learning\n\n## Quick start\n\nTRL provides post-training methods for aligning language models with human preferences.\n\n**Installation**:\n```bash\npip install trl transformers datasets peft accelerate\n```\n\n**Supervised Fine-Tuning** (instruction tuning):\n```python\nfrom trl import SFTTrainer\n\ntrainer = SFTTrainer(\n    model=\"Qwen\u002FQwen2.5-0.5B\",\n    train_dataset=dataset,  # Prompt-completion pairs\n)\ntrainer.train()\n```\n\n**DPO** (align with preferences):\n```python\nfrom trl import DPOTrainer, DPOConfig\n\nconfig = DPOConfig(output_dir=\"model-dpo\", beta=0.1)\ntrainer = DPOTrainer(\n    model=model,\n    args=config,\n    train_dataset=preference_dataset,  # chosen\u002Frejected pairs\n    processing_class=tokenizer\n)\ntrainer.train()\n```\n\n## Common workflows\n\n### Workflow 1: Full RLHF pipeline (SFT → Reward Model → PPO)\n\nComplete pipeline from base model to human-aligned model.\n\nCopy this checklist:\n\n```\nRLHF Training:\n- [ ] Step 1: Supervised fine-tuning (SFT)\n- [ ] Step 2: Train reward model\n- [ ] Step 3: PPO reinforcement learning\n- [ ] Step 4: Evaluate aligned model\n```\n\n**Step 1: Supervised fine-tuning**\n\nTrain base model on instruction-following data:\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom trl import SFTTrainer, SFTConfig\nfrom datasets import load_dataset\n\n# Load model\nmodel = AutoModelForCausalLM.from_pretrained(\"Qwen\u002FQwen2.5-0.5B\")\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen\u002FQwen2.5-0.5B\")\n\n# Load instruction dataset\ndataset = load_dataset(\"trl-lib\u002FCapybara\", split=\"train\")\n\n# Configure training\ntraining_args = SFTConfig(\n    output_dir=\"Qwen2.5-0.5B-SFT\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=2e-5,\n    logging_steps=10,\n    save_strategy=\"epoch\"\n)\n\n# Train\ntrainer = SFTTrainer(\n    model=model,\n    args=training_args,\n    train_dataset=dataset,\n    tokenizer=tokenizer\n)\ntrainer.train()\ntrainer.save_model()\n```\n\n**Step 2: Train reward model**\n\nTrain model to predict human preferences:\n\n```python\nfrom transformers import AutoModelForSequenceClassification\nfrom trl import RewardTrainer, RewardConfig\n\n# Load SFT model as base\nmodel = AutoModelForSequenceClassification.from_pretrained(\n    \"Qwen2.5-0.5B-SFT\",\n    num_labels=1  # Single reward score\n)\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen2.5-0.5B-SFT\")\n\n# Load preference data (chosen\u002Frejected pairs)\ndataset = load_dataset(\"trl-lib\u002Fultrafeedback_binarized\", split=\"train\")\n\n# Configure training\ntraining_args = RewardConfig(\n    output_dir=\"Qwen2.5-0.5B-Reward\",\n    per_device_train_batch_size=2,\n    num_train_epochs=1,\n    learning_rate=1e-5\n)\n\n# Train reward model\ntrainer = RewardTrainer(\n    model=model,\n    args=training_args,\n    processing_class=tokenizer,\n    train_dataset=dataset\n)\ntrainer.train()\ntrainer.save_model()\n```\n\n**Step 3: PPO reinforcement learning**\n\nOptimize policy using reward model:\n\n```bash\npython -m trl.scripts.ppo \\\n    --model_name_or_path Qwen2.5-0.5B-SFT \\\n    --reward_model_path Qwen2.5-0.5B-Reward \\\n    --dataset_name trl-internal-testing\u002Fdescriptiveness-sentiment-trl-style \\\n    --output_dir Qwen2.5-0.5B-PPO \\\n    --learning_rate 3e-6 \\\n    --per_device_train_batch_size 64 \\\n    --total_episodes 10000\n```\n\n**Step 4: Evaluate**\n\n```python\nfrom transformers import pipeline\n\n# Load aligned model\ngenerator = pipeline(\"text-generation\", model=\"Qwen2.5-0.5B-PPO\")\n\n# Test\nprompt = \"Explain quantum computing to a 10-year-old\"\noutput = generator(prompt, max_length=200)[0][\"generated_text\"]\nprint(output)\n```\n\n### Workflow 2: Simple preference alignment with DPO\n\nAlign model with preferences without reward model.\n\nCopy this checklist:\n\n```\nDPO Training:\n- [ ] Step 1: Prepare preference dataset\n- [ ] Step 2: Configure DPO\n- [ ] Step 3: Train with DPOTrainer\n- [ ] Step 4: Evaluate alignment\n```\n\n**Step 1: Prepare preference dataset**\n\nDataset format:\n```json\n{\n  \"prompt\": \"What is the capital of France?\",\n  \"chosen\": \"The capital of France is Paris.\",\n  \"rejected\": \"I don't know.\"\n}\n```\n\nLoad dataset:\n```python\nfrom datasets import load_dataset\n\ndataset = load_dataset(\"trl-lib\u002Fultrafeedback_binarized\", split=\"train\")\n# Or load your own\n# dataset = load_dataset(\"json\", data_files=\"preferences.json\")\n```\n\n**Step 2: Configure DPO**\n\n```python\nfrom trl import DPOConfig\n\nconfig = DPOConfig(\n    output_dir=\"Qwen2.5-0.5B-DPO\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=5e-7,\n    beta=0.1,  # KL penalty strength\n    max_prompt_length=512,\n    max_length=1024,\n    logging_steps=10\n)\n```\n\n**Step 3: Train with DPOTrainer**\n\n```python\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom trl import DPOTrainer\n\nmodel = AutoModelForCausalLM.from_pretrained(\"Qwen\u002FQwen2.5-0.5B-Instruct\")\ntokenizer = AutoTokenizer.from_pretrained(\"Qwen\u002FQwen2.5-0.5B-Instruct\")\n\ntrainer = DPOTrainer(\n    model=model,\n    args=config,\n    train_dataset=dataset,\n    processing_class=tokenizer\n)\n\ntrainer.train()\ntrainer.save_model()\n```\n\n**CLI alternative**:\n```bash\ntrl dpo \\\n    --model_name_or_path Qwen\u002FQwen2.5-0.5B-Instruct \\\n    --dataset_name argilla\u002FCapybara-Preferences \\\n    --output_dir Qwen2.5-0.5B-DPO \\\n    --per_device_train_batch_size 4 \\\n    --learning_rate 5e-7 \\\n    --beta 0.1\n```\n\n### Workflow 3: Memory-efficient online RL with GRPO\n\nTrain with reinforcement learning using minimal memory.\n\nFor in-depth GRPO guidance — reward function design, critical training insights (loss behavior, mode collapse, tuning), and advanced multi-stage patterns — see **[references\u002Fgrpo-training.md](references\u002Fgrpo-training.md)**. A production-ready training script is in **[templates\u002Fbasic_grpo_training.py](templates\u002Fbasic_grpo_training.py)**.\n\nCopy this checklist:\n\n```\nGRPO Training:\n- [ ] Step 1: Define reward function\n- [ ] Step 2: Configure GRPO\n- [ ] Step 3: Train with GRPOTrainer\n```\n\n**Step 1: Define reward function**\n\n```python\ndef reward_function(completions, **kwargs):\n    \"\"\"\n    Compute rewards for completions.\n\n    Args:\n        completions: List of generated texts\n\n    Returns:\n        List of reward scores (floats)\n    \"\"\"\n    rewards = []\n    for completion in completions:\n        # Example: reward based on length and unique words\n        score = len(completion.split())  # Favor longer responses\n        score += len(set(completion.lower().split()))  # Reward unique words\n        rewards.append(score)\n    return rewards\n```\n\nOr use a reward model:\n```python\nfrom transformers import pipeline\n\nreward_model = pipeline(\"text-classification\", model=\"reward-model-path\")\n\ndef reward_from_model(completions, prompts, **kwargs):\n    # Combine prompt + completion\n    full_texts = [p + c for p, c in zip(prompts, completions)]\n    # Get reward scores\n    results = reward_model(full_texts)\n    return [r[\"score\"] for r in results]\n```\n\n**Step 2: Configure GRPO**\n\n```python\nfrom trl import GRPOConfig\n\nconfig = GRPOConfig(\n    output_dir=\"Qwen2-GRPO\",\n    per_device_train_batch_size=4,\n    num_train_epochs=1,\n    learning_rate=1e-5,\n    num_generations=4,  # Generate 4 completions per prompt\n    max_new_tokens=128\n)\n```\n\n**Step 3: Train with GRPOTrainer**\n\n```python\nfrom datasets import load_dataset\nfrom trl import GRPOTrainer\n\n# Load prompt-only dataset\ndataset = load_dataset(\"trl-lib\u002Ftldr\", split=\"train\")\n\ntrainer = GRPOTrainer(\n    model=\"Qwen\u002FQwen2-0.5B-Instruct\",\n    reward_funcs=reward_function,  # Your reward function\n    args=config,\n    train_dataset=dataset\n)\n\ntrainer.train()\n```\n\n**CLI**:\n```bash\ntrl grpo \\\n    --model_name_or_path Qwen\u002FQwen2-0.5B-Instruct \\\n    --dataset_name trl-lib\u002Ftldr \\\n    --output_dir Qwen2-GRPO \\\n    --num_generations 4\n```\n\n## When to use vs alternatives\n\n**Use TRL when:**\n- Need to align model with human preferences\n- Have preference data (chosen\u002Frejected pairs)\n- Want to use reinforcement learning (PPO, GRPO)\n- Need reward model training\n- Doing RLHF (full pipeline)\n\n**Method selection**:\n- **SFT**: Have prompt-completion pairs, want basic instruction following\n- **DPO**: Have preferences, want simple alignment (no reward model needed)\n- **PPO**: Have reward model, need maximum control over RL\n- **GRPO**: Memory-constrained, want online RL\n- **Reward Model**: Building RLHF pipeline, need to score generations\n\n**Use alternatives instead:**\n- **HuggingFace Trainer**: Basic fine-tuning without RL\n- **Axolotl**: YAML-based training configuration\n- **LitGPT**: Educational, minimal fine-tuning\n- **Unsloth**: Fast LoRA training\n\n## Common issues\n\n**Issue: OOM during DPO training**\n\nReduce batch size and sequence length:\n```python\nconfig = DPOConfig(\n    per_device_train_batch_size=1,  # Reduce from 4\n    max_length=512,  # Reduce from 1024\n    gradient_accumulation_steps=8  # Maintain effective batch\n)\n```\n\nOr use gradient checkpointing:\n```python\nmodel.gradient_checkpointing_enable()\n```\n\n**Issue: Poor alignment quality**\n\nTune beta parameter:\n```python\n# Higher beta = more conservative (stays closer to reference)\nconfig = DPOConfig(beta=0.5)  # Default 0.1\n\n# Lower beta = more aggressive alignment\nconfig = DPOConfig(beta=0.01)\n```\n\n**Issue: Reward model not learning**\n\nCheck loss type and learning rate:\n```python\nconfig = RewardConfig(\n    learning_rate=1e-5,  # Try different LR\n    num_train_epochs=3  # Train longer\n)\n```\n\nEnsure preference dataset has clear winners:\n```python\n# Verify dataset\nprint(dataset[0])\n# Should have clear chosen > rejected\n```\n\n**Issue: PPO training unstable**\n\nAdjust KL coefficient:\n```python\nconfig = PPOConfig(\n    kl_coef=0.1,  # Increase from 0.05\n    cliprange=0.1  # Reduce from 0.2\n)\n```\n\n## Advanced topics\n\n**SFT training guide**: See [references\u002Fsft-training.md](references\u002Fsft-training.md) for dataset formats, chat templates, packing strategies, and multi-GPU training.\n\n**DPO variants**: See [references\u002Fdpo-variants.md](references\u002Fdpo-variants.md) for IPO, cDPO, RPO, and other DPO loss functions with recommended hyperparameters.\n\n**Reward modeling**: See [references\u002Freward-modeling.md](references\u002Freward-modeling.md) for outcome vs process rewards, Bradley-Terry loss, and reward model evaluation.\n\n**Online RL methods**: See [references\u002Fonline-rl.md](references\u002Fonline-rl.md) for PPO, GRPO, RLOO, and OnlineDPO with detailed configurations.\n\n**GRPO deep dive**: See [references\u002Fgrpo-training.md](references\u002Fgrpo-training.md) for expert-level GRPO patterns — reward function design philosophy, training insights (why loss increases, mode collapse detection), hyperparameter tuning, multi-stage training, and troubleshooting. Production-ready template in [templates\u002Fbasic_grpo_training.py](templates\u002Fbasic_grpo_training.py).\n\n## Hardware requirements\n\n- **GPU**: NVIDIA (CUDA required)\n- **VRAM**: Depends on model and method\n  - SFT 7B: 16GB (with LoRA)\n  - DPO 7B: 24GB (stores reference model)\n  - PPO 7B: 40GB (policy + reward model)\n  - GRPO 7B: 24GB (more memory efficient)\n- **Multi-GPU**: Supported via `accelerate`\n- **Mixed precision**: BF16 recommended (A100\u002FH100)\n\n**Memory optimization**:\n- Use LoRA\u002FQLoRA for all methods\n- Enable gradient checkpointing\n- Use smaller batch sizes with gradient accumulation\n\n## Resources\n\n- Docs: https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftrl\u002F\n- GitHub: https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl\n- Papers:\n  - \"Training language models to follow instructions with human feedback\" (InstructGPT, 2022)\n  - \"Direct Preference Optimization: Your Language Model is Secretly a Reward Model\" (DPO, 2023)\n  - \"Group Relative Policy Optimization\" (GRPO, 2024)\n- Examples: https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl\u002Ftree\u002Fmain\u002Fexamples\u002Fscripts\n\n\n\n",{"name":5,"author":23,"license":14,"version":24,"metadata":25,"platforms":39,"description":7,"dependencies":43},"Orchestra Research","1.0.0",{"hermes":26},{"tags":27},[28,29,30,31,32,33,34,35,36,37,38],"Post-Training","TRL","Reinforcement Learning","Fine-Tuning","SFT","DPO","PPO","GRPO","RLHF","Preference Alignment","HuggingFace",[40,41,42],"linux","macos","windows",[44,45,46,47,48,49],"trl","transformers","datasets","peft","accelerate","torch","main","e0ec177d00f087cb04e90206b2f7d3aa2f5a1350","References\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim","https:\u002F\u002Fgithub.com\u002Flzy18001500226","2026-08-31 23:29:00",{"source":8,"skillName":5,"commands":57,"prompt":67,"zipAvailable":68},[58,61,64],{"tool":59,"command":60},"npx","npx skills add lzy18001500226\u002FMoSim --skill fine-tuning-with-trl",{"tool":62,"command":63},"bunx","bunx skills add lzy18001500226\u002FMoSim --skill fine-tuning-with-trl",{"tool":65,"command":66},"pnpm","pnpm dlx skills add lzy18001500226\u002FMoSim --skill fine-tuning-with-trl","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002FSKILL.md",true,[70,76,80,84,88,92],{"path":71,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fdpo-variants.md","REFERENCE","参考文档",4297,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Freferences\u002Fdpo-variants.md",{"path":77,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fgrpo-training.md",15901,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Freferences\u002Fgrpo-training.md",{"path":81,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fonline-rl.md",1971,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Freferences\u002Fonline-rl.md",{"path":85,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Freward-modeling.md",2597,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Freferences\u002Freward-modeling.md",{"path":89,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fsft-training.md",3237,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Freferences\u002Fsft-training.md",{"path":93,"resourceType":94,"resourceTypeDesc":95,"sizeBytes":96,"sourceHtmlUrl":97},"templates\u002Fbasic_grpo_training.py","SCRIPT","脚本",6113,"https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FPlatforms\u002Fhermes-agent\u002Foptional-skills\u002Fmlops\u002Ftraining\u002Ftrl-fine-tuning\u002Ftemplates\u002Fbasic_grpo_training.py",[99,109,117,125,133,141],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":103,"licenseFileUrl":104,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":108},"01M78RK6JGRPTD3TDSMTQWDQ9P","skill-installer","Install Codex skills into $CODEX_HOME\u002Fskills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).","Apache-2.0","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fskill-installer\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fskill-installer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":103,"licenseFileUrl":113,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":108},"01M78RK29KVRDQXSMVZX1GQTJN","skill-creator","Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fskill-creator\u002Flicense.txt","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fskill-creator\u002FSKILL.md",[],[],{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":103,"licenseFileUrl":121,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":108},"01M78RJPF3210Y5DA9VJ6JMMG6","openai-docs","Use for Codex models\u002Fpricing, scheduled tasks, skills, settings, setup, troubleshooting, customization, automations, and self-knowledge—including 'you,' 'your,' 'this app,' or 'this coding agent' when they refer to Codex—and for OpenAI APIs\u002Fproducts and ChatGPT Work. Also use for model choice\u002Fmigration, prompting, SDKs, Responses, Realtime, agents, evals, and Chat\u002FWork\u002FCodex comparisons. Do not use for generic app\u002Fsoftware tasks that merely mention Codex.","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fopenai-docs\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fopenai-docs\u002FSKILL.md",[],[],{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":103,"licenseFileUrl":129,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":108},"01M78RJH7X67C5AY8ZVS7RQNWJ","imagegen","Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG\u002Fvector\u002Fcode-native assets, extending an established icon or logo system, or building the visual directly in HTML\u002FCSS\u002Fcanvas.","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fimagegen\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002Fsrc\u002FAgent\u002Fcodex-main\u002Fcodex-rs\u002Fskills\u002Fsrc\u002Fassets\u002Fsamples\u002Fimagegen\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":103,"licenseFileUrl":137,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":108},"01M78RBBAYMJ6T3ZMAYHCFM6G7","skill-creator-cc","Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FWorkflow\u002Fokwinds\u002Fmiscellany\u002Fagent\u002Fskills\u002Fskill-creator-cc\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FWorkflow\u002Fokwinds\u002Fmiscellany\u002Fagent\u002Fskills\u002Fskill-creator-cc\u002FSKILL.md",[],[],{"skillNo":142,"name":143,"nameZh":6,"description":144,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":145,"sourceHtmlUrl":146,"categories":147,"keywords":148,"publishedAt":149},"01M76VV9V2R9FZTVGMJZV7Y996","prd-writing-guide","Write complete, unambiguous PRDs that development teams can implement without guesswork. Includes requirement discovery framework, structured documentation methodology, completeness checklists, and common pitfall avoidance. Use when: writing new PRDs, reviewing PRD drafts, validating requirement completeness, preparing for engineering handoff. Triggers: 'write PRD', '写PRD', '产品需求文档', '需求文档', '需求规格', '需求评审', '完善需求', 'create requirements doc', 'product requirements', 'feature spec', 'requirements document'. Anti-triggers: 'technical design doc', 'architecture design', 'implementation plan', 'API design', '架构设计', '技术方案', '实现方案', '接口设计'.","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FWorkflow\u002Fokwinds\u002Fmiscellany\u002Fagent\u002Fskills\u002Fprd-writing-guide\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Flzy18001500226\u002FMoSim\u002Fblob\u002Fmain\u002FReferences\u002FAgent\u002FWorkflow\u002Fokwinds\u002Fmiscellany\u002Fagent\u002Fskills\u002Fprd-writing-guide\u002FSKILL.md",[],[],"2026-11-12 13:55:43"]