junewgl
junewgl
这个错误是在运行 crnn_main.py时出现的。主要问题代码是在dataset.py里面。
请问RLHF进展怎么样了?
Can you describe the content in more detail? @GAN-007
If you are training LLM for the first time, you can first use the most basic `single GPU` to run a smaller LLM, such as `codellama-7b-instruct` @smartforxu
Can you provide specific information about your `export_merge.sh`? Are the weights based on our `train_sft.sh`? You can directly use the [huggingface](https://huggingface.co/Wangzaistone123/CodeLlama-13b-sql-lora) weights we uploaded. @hoangdzung
This seems to be the cause of AMD GPU hardware, you can try using our suggested [AutoDL](https://www.autodl.com/console/instance/list) method. @shouziws
Great, thank you for your issue! You can refer to [PR/88](https://github.com/eosphoros-ai/Awesome-Text2SQL/pull/88) to submit a similar PR. We are looking forward to your contribution and making the community better! @QinYang79