Ray Wang

Has this issue been fixed? If we want to chat for multiple rounds, which command to use to avoid this UI issue?

Hi, I should be able to help with a AWS EC2 G5 instance with 24GB RAM. Is that enough? @sanagno How could I find you on Discord?

For those who wants to work on expanding the scope of OpenAssistant into Chinese, please DM me and let's chat about how to roll out the first version of feature.

Hi, We got in touch with each other and researched on open APIs from Zhihu to onboard it to the Open Assistant data collection portal. https://open-assistant.io/dashboard We are following https://projects.laion.ai/Open-Assistant/docs/data/datasets...

@ontocord A proof-of-concept dataset in parquet format has been uploaded to https://huggingface.co/datasets/wangrui6/Zhihu-KOL Can you double check the format?

@ontocord Do we need to append the citation at the end of each answer from the website?

@Facico A100 的训练参数组合有经验吗？ ··· ### # optimized for RTX 4090. for larger GPUs, increase some of these? MICRO_BATCH_SIZE = 4 # this could actually be 5 but i like powers...

@Facico 另外想问一下finetune的速度如何？用4090 finetune vicuna13b，100K的samples大概要多久？有可以参考的数据吗？

any update on the error? Experienced this issue on a newly setup EC2 instance.