GitHub - aorko01/Flash_GPT_2: Custom GPT-2 with memory-efficient FlashAttention implemented in Triton for faster training in PyTorch.

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
gpu_profiler.py		gpu_profiler.py
input.txt		input.txt
model.py		model.py
program.py		program.py
requirements.txt		requirements.txt
trainGpt.py		trainGpt.py
train_gpt2.py		train_gpt2.py
train_test.py		train_test.py