PlannedSchedule of parts only
Advanced Training
What happens to a model after pre-training, and how a base model becomes something worth shipping. Training LLMs covered pre-training, fine-tuning and PEFT at a high level. This one goes under that.
| Item | Description |
|---|---|
| 01 | Self-supervised learning: how pre-training gets a training signal with no labels |
| 02 | Post-training: everything done to a base model before it is released |
| 03 | Reinforcement learning, and RLHF |
| 04 | Preference alignment: DPO, PPO and GRPO |
| 05 | Knowledge distillation: teaching a small model from a large one |