hiyouga 767aae4b72 fix #4609 ...

unwrap_model_for_generation(reward_model) is necessary for zero3 training


Former-commit-id: c8d5b21700577cae8d6ca03359bcf1762c8b7cb8

2024-07-03 19:45:51 +08:00

..

dpo

refactor pissa, improve llamaboard

2024-06-28 01:04:24 +08:00

kto

refactor pissa, improve llamaboard

2024-06-28 01:04:24 +08:00

ppo

fix #4609

2024-07-03 19:45:51 +08:00

pt

refactor pissa, improve llamaboard

2024-06-28 01:04:24 +08:00

rm

remove rlhf support for chatglm2&3

2024-07-02 23:03:17 +08:00

sft

improve rlhf

2024-07-02 22:23:08 +08:00

__init__.py

rename package

2024-05-16 18:39:08 +08:00

callbacks.py

loose gemma2 attention

2024-06-29 01:42:14 +08:00

trainer_utils.py

refactor pissa, improve llamaboard

2024-06-28 01:04:24 +08:00

tuner.py

improve rlhf

2024-07-02 22:23:08 +08:00