5 Commits (86bcda5ca9ce73d8f0751f58e80a280d61a1c93f)

Author	SHA1	Message	Date
Wenwen Qu	86bcda5ca9	add default setting for expert parallel size	2023-08-24 18:52:12 +08:00
Wenwen Qu	12c614db94	create expert data group and broadcast moe parameter in expert data group	2023-08-21 11:40:39 +08:00
Wenwen Qu	c357288a8b	feat(XXX): add moe	2023-08-07 20:17:49 +08:00
ytxiong	c219065348	feat(*): support sequence_parallel (#180) ... * support sequence_parallel for no pipeline * sequence_parallel does not support no-flash-attn * support sequence parallel for pipeline * add memory profiler * Update 13B.py * add memory profiler * fix evaluation bug * remove some unnecessary code * remove some unnecessary code * Update parallel_context.py * modify the config * remove memory profiler * modify the config * support selective dropout	2023-08-07 16:42:52 +08:00
Sun Peng	fa7337b37b	initial commit	2023-07-06 12:55:23 +08:00