wZuck's repositories
DeepSpeed
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Language:PythonApache-2.0000
GPU-Puzzles
Solve puzzles. Learn CUDA.
Language:Jupyter NotebookMIT000
light-llm-grad
This is a lightweight training framework for LLM (Language Model).
Apache-2.0000
Megatron-LM
Ongoing research training transformer models at scale
Language:PythonNOASSERTION000