Skip to content

Latest commit

 

History

History
8 lines (6 loc) · 316 Bytes

README.md

File metadata and controls

8 lines (6 loc) · 316 Bytes

DeepSeek R1 Zero Toy Model Training

Install: !pip install transformers datasets wandb tqdm !pip install bitsandbytes !wandb login --anonymously

Then run the code to train your own small DeepSeek R1 Zero from a small base model. The aim is to be able to run it in limited resources and over a smaller dataset.