model_finetuning

SKILL.md

TRL provides post-training methods for aligning language models with human preferences.

Workflow 1: Full RLHF pipeline (SFT → Reward Model → PPO)

SFT training guide: See references/sft-training.md for dataset formats, chat templates, packing strategies, and multi-GPU training.

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers. Source: vuralserhat86/antigravity-agentic-skills.

View raw