mindspore-lab / mindrlhf

Apache License 2.0
26 stars 12 forks source link

add qwen2 dpo #89

Closed coder-yuzhiwei closed 1 day ago

coder-yuzhiwei commented 2 days ago

增加了qwen2 dpo训练代码

ChessQian commented 1 day ago

add qwen2 7B dpo train and inference