Foundational course module covering dataset preprocessing, dynamic padding, the Trainer API, custom PyTorch fine-tuning loops, and learning rate schedules for adapting pretrained Transformers.
Chapter 3 of the Hugging Face LLM Course introduces the mechanics of adapting a pretrained Transformer checkpoint to downstream tasks. Crucially, it teaches both the high-level `Trainer` workflow and how to strip away the wrapper to write a native PyTorch + `Accelerate` fine-tuning loop with AdamW and linear warmup/decay.
Course status is tracked independently from roadmap topic checkboxes so you can mark external courses without altering your topic mastery percentage.
Phase 7 Modules 7.1 & 7.2 Fine-Tuning Foundations
100% free interactive web module with runnable Colab notebooks.
Status: Verified (2026-11-09)
Review these foundations before starting so you get the most out of this material:
In Phase 7: Post-training & Alignment
Review during Modules 7.1 and 7.2 for clear coverage of dynamic batch collation, padding masks, and native PyTorch fine-tuning loops.
Base vs. instruction-tuned models, chat templates, prompt-response loss masking, Supervised Fine-Tuning (SFT), and LIMA-style dataset quality.
Full fine-tuning memory costs, Low-Rank Adaptation (LoRA), Quantized LoRA (QLoRA), adapters, and memory-performance tradeoffs.
Practical hands-on course focused on post-training small and medium open-weight language models: chat templates, Supervised Fine-Tuning (SFT), LoRA/QLoRA parameter-efficient adaptation, preference alignment (DPO/ORPO), and evaluation.
Comprehensive open interactive course covering Transformer models, tokenization pipelines, datasets, fine-tuning, and the open-source LLM ecosystem.
Official documentation and technical guides for TRL (Transformer Reinforcement Learning), covering `SFTTrainer`, `RewardTrainer`, `PPOTrainer`, `DPOTrainer`, and `GRPOTrainer`.