6 Commits

Author	SHA1	Message	Date
Wing Lian	68b227a7d8	Mixtral multipack (#928) ... * mixtral multipack * use mixtral model * sample yml * calculate cu_seqlens properly * use updated flash ettention setting * attn var checks * force use of flash attention 2 for packing * lint * disable future fix for now * update support table	2023-12-09 21:26:30 -05:00
Wing Lian	f162f3c7cc	set transformers cache env var in docker image	2023-07-16 23:03:54 -04:00
Wing Lian	fa8bd14be4	update entrypoint and force min accelerate	2023-05-18 06:25:34 -04:00
Wing Lian	709be5af81	use printf instead of echo in dockerfile for portability	2023-05-08 11:45:38 -04:00
Wing Lian	807cca81c0	fix path name to sorkspace	2023-05-08 11:20:03 -04:00
Wing Lian	79deb35c68	setup runpod images ... use github.ref_name	2023-05-08 10:48:32 -04:00