by zechenzhangAGI
Expert guidance for Fully Sharded Data Parallel training with PyTorch FSDP - parameter sharding, mixed precision, CPU offloading, FSDP2
2.2
Rating
0
Installs
Machine Learning
Category
No summary available.
zechenzhangAGI
Skill Author
Loading SKILL.md…
ml-pipeline
Jeffallan
sparse-autoencoder-training
huggingface-accelerate
moe-training