train_gpt2.c header

slash star This file trains the G P T two model. This version is the clean, minimal, reference. As such: - it runs on CPU. - it does not make the code too complex; it is readable. - it does not use any processor-specific instructions, intrinsics and such. - it _does_ use a few OpenMP pragmas because this is a large speedup at very low cost There will be other versions of this code that specialize it and make it fast. star slash hash include std I O dot h hash include std lib dot h hash include math dot h hash include time dot h hash include string dot h hash include unistd dot h hash if def O M P hash includeo m p dot h hash end if