使用flash attention会报错

ymcui / Chinese-LLaMA-Alpaca-2

中文LLaMA-2 & Alpaca-2大模型二期项目 + 64K超长上下文模型 (Chinese LLaMA-2 & Alpaca-2 LLMs with 64K long context models)

Apache License 2.0

7.06k stars 578 forks source link

Closed Go4miii closed 8 months ago

Go4miii commented 8 months ago

在运行run_sft.sh时使用--flash_attn 会报错：ValueError: Some specified arguments are not used by the HfArgumentParser: ['--flash_attn'] 。请问是什么情况？

iMountTai commented 8 months ago

修改为--use_flash_attention_2

Go4miii commented 8 months ago

修改为--use_flash_attention_2

好的我试试，谢谢！！