C-Cuda-GPT2 (Ongoing)

FlashAttention from scratch in CUDA C using cuBLASLt and cuDNN API

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
data_utils		data_utils
include		include
model		model
utils		utils
1_train_gpt.cu		1_train_gpt.cu
README.md		README.md

Provide feedback