You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Reproduced the GPT-2 124M parameter LLM from scratch using PyTorch, referencing ”At tention is All You Need” and GPT papers. Optimized performance with Flash Attention, Torch Compile, Gradient Accu…