786 Commits

Author	SHA1	Message	Date
Johann-Peter Hartmann	63da6294dd	Merge branch 'hiyouga:main' into main	2024-02-04 12:51:25 +00:00
hiyouga	db0ab4d601	fix reserved label len	2024-02-04 17:54:26 +08:00
hiyouga	19d33ede13	fix #2420	2024-02-04 15:51:47 +08:00
hiyouga	b988ce0a0c	fix #2189	2024-02-04 00:47:37 +08:00
hiyouga	38e63bfd28	bump up transformers version	2024-02-04 00:01:16 +08:00
hiyouga	51df865734	fix #2397	2024-02-03 23:45:31 +08:00
hiyouga	6545c02790	add hint for freeze #2412	2024-02-03 23:38:56 +08:00
hiyouga	4ecadc3512	fix #2376	2024-02-03 23:14:31 +08:00
hiyouga	901faa16cc	support minicpm #2404	2024-02-03 22:36:46 +08:00
Johann-Peter Hartmann	b0ffde6e98	add simple german chatml template chatml_de	2024-02-03 09:01:15 +01:00
Fallen Angel	3399c0d645	fix eos_token_id=0 bug ... when eos_token_id=0, will never add eos_token	2024-02-02 17:34:48 +08:00
hiyouga	39bd5bd524	fix #2388	2024-01-31 17:23:56 +08:00
hiyouga	521ad76552	fix autoset attn impl, update data readme	2024-01-31 11:58:07 +08:00
hiyouga	2bc30763e9	fix #2320	2024-01-24 16:19:18 +08:00
ldwang	c284665425	Add patch_mixtral_replace_moe_impl for full training Mitral using DeepSpeed Zero3. ... Signed-off-by: ldwang <ftgreat@gmail.com>	2024-01-24 15:25:31 +08:00
ldwang	18923b1402	Add patch_mixtral_replace_moe_impl for full training Mitral using DeepSpeed Zero3. ... Signed-off-by: ldwang <ftgreat@gmail.com>	2024-01-24 14:43:16 +08:00
hiyouga	e4ba1deedf	add hint	2024-01-22 23:32:01 +08:00
hoshi-hiyouga	8e4b3a959a	Merge pull request #2283 from A-Cepheus/main ... fix: ZeRO3 does not work with MoE models	2024-01-22 23:28:45 +08:00
hoshi-hiyouga	bdc9eff635	Update patcher.py	2024-01-22 23:27:39 +08:00
hiyouga	b2fb0eca56	fix #2282 and update tool prompt	2024-01-22 22:27:30 +08:00
hiyouga	6fc2d5cc03	add orion models	2024-01-22 21:26:53 +08:00
A-Cepheus	b06a31e76a	🐞 fix: typo	2024-01-22 16:04:39 +08:00
A-Cepheus	319a72b48d	🐞 fix: typo, move MoE fix to patcher	2024-01-22 16:01:58 +08:00
A-Cepheus	e1d5c98519	fix: ZeRO3 does not work with MoE models	2024-01-22 15:21:14 +08:00
hiyouga	486cc8d360	add array param format	2024-01-21 22:17:48 +08:00
hiyouga	f51d48ea08	fix api	2024-01-21 19:15:27 +08:00
hiyouga	e0a717aa3a	fix #2268	2024-01-21 14:11:38 +08:00
hiyouga	3bc301be02	tiny fix	2024-01-21 13:26:12 +08:00
hoshi-hiyouga	2533be5e2a	Merge branch 'main' into fix_export_model_dtype	2024-01-21 12:40:24 +08:00
hoshi-hiyouga	662b9a9dcf	Update tuner.py	2024-01-21 12:39:38 +08:00
hoshi-hiyouga	11f612deab	Merge pull request #2262 from fenglui/main ... fix torch_dtype check of export_model	2024-01-21 12:34:37 +08:00
hiyouga	3c1e256992	format	2024-01-21 12:34:17 +08:00
yhyu13	9cdbd3bfc8	Remove manully set use_cache; torch_dtype is not str, save model as bfloat16 used to fail;	2024-01-21 11:12:15 +08:00
Aleksandr	ace7fd4e0a	add russian lang	2024-01-21 04:28:14 +03:00
fenglui	f7c5b2d105	fix torch_dtype check of export_model	2024-01-21 05:01:53 +08:00
hiyouga	3e982cc714	finish agent	2024-01-21 01:47:33 +08:00
hiyouga	55f707196e	fix api	2024-01-21 00:03:09 +08:00
hiyouga	a9c18255aa	fix internlm2 template	2024-01-20 23:33:50 +08:00
hiyouga	c550987a72	fix cli_demo	2024-01-20 23:27:10 +08:00
hiyouga	cf818a2598	fix #2260	2024-01-20 23:22:09 +08:00
hiyouga	71cfdc2658	release v0.5.0	2024-01-20 20:21:39 +08:00
hiyouga	638234ceee	format style	2024-01-20 20:15:56 +08:00
hiyouga	f6d6e00337	fix tests	2024-01-20 19:58:04 +08:00
hiyouga	38af076a75	support longlora for main branch	2024-01-20 19:25:22 +08:00
hoshi-hiyouga	bb92cdd0db	Merge pull request #2201 from liu-zichen/token_embed_resize ... support resize embed for zero3	2024-01-20 17:45:38 +08:00
hiyouga	8cbe4e9609	add upcast_lmhead option	2024-01-19 23:54:25 +08:00
hiyouga	0ff9a1fb4f	set use_reentrant=False	2024-01-19 23:29:54 +08:00
hiyouga	12043aab9c	fix #2249	2024-01-19 21:44:32 +08:00
hiyouga	b6ec112beb	add bf16 lora option	2024-01-19 16:29:03 +08:00
hiyouga	35aef8b287	fix function formatter	2024-01-18 16:01:07 +08:00