5 Commits

Author	SHA1	Message	Date
LiangSong	f8f4cde228	using huggingface datasets to accelerate training, using open-llama to pretrain	2023-04-24 19:13:53 +08:00
LiangSong	9f140dc99f	update preprocess_instruction, add math/code/multiturn_chat and etc.	2023-04-05 23:51:56 +08:00
LiangSong	a62ac2658f	add instruction-tuning	2023-03-30 23:43:12 +08:00
LiangSong	918a8cdc3d	reformat code with black	2023-03-27 14:34:59 +08:00
LiangSong	73a81a4205	add high-performance Llama pre-train code	2023-03-26 23:59:53 +08:00