Track 10: Training LLMs
LLM Training Beyond Basics: Preference Tuning, Evals and Deployment
4 lessons
Track 10: Training LLMs, Preference Tuning and Alignment
GRPO and RL for Reasoning
Train a small model to reason with verifiable rewards.
Part of Train and Fine-Tune Models (33 of 45)
Get notified on YouTube Udemy certificate: coming soon
Course
LLM Training Beyond Basics: Preference Tuning, Evals and Deployment
Udemy: coming soon
Tools
- Hugging Face TRL (open-source, free)
- Unsloth (open-source, free)
Watch first
Then try
In this module
- RLHF and Preference Tuning, Explained
- DPO Hands-On (and ORPO, KTO, Reward Models)
- GRPO and RL for Reasoning
The course this lesson belongs to