Large-Language-Models-on-Supercomputers
PublicComprehensive exploration of LLMs, including cutting-edge techniques and tools such as parameter-efficient fine-tuning (PEFT), quantization, zero redundancy optimizers (ZeRO), fully sharded data parallelism (FSDP), DeepSpeed, and Huggingface accelerate.
deepspeedevaluation-metricsfsdphigh-performance-computinghpchuggingfacehuggingface-transformersjupyterllmllm-inference
Creat:2024-09-26T17:08:32
Update:2024-11-08T07:42:01
0
Stars
0
Stars Increase