Skip to content

Latest commit

 

History

History
10 lines (7 loc) · 390 Bytes

README.md

File metadata and controls

10 lines (7 loc) · 390 Bytes

learn-llm

This project is dedicated to learning and developing large language models (LLMs). It encompasses various stages of training and testing, including pretraining, supervised fine-tuning (SFT), and reinforcement learning (RL).

Repos

  • pretrain: pretrain code
  • sft: Supervised fine tunning
  • rl: Reinforcement learning
  • agents: some agents