Arthur Wu
Arthur Wu
hi, Did you find the relevant training code?
> Thanks for immediate response > > what template do you use on the online demo? I also want to know
@krrishdholakia What I mean is that after the health check, it is found that the server has a problem, the route does not request that server, which means that the...
Yes, I think this is very important, you can open update_deployment(), and then give a simple update_deployment() implementation example
> 机器翻译的数据，对小模型可能是一种数据污染。 在数据生成上面，GuanacoDataset迄今已花费$5K在OpenAI。 对中文数据方面，我们公开发布了Alpaca175任务中文版本的185,702条项目，重复度0.7以内；含有中文的对话任务48,967条。 所有中文包含了简体中文、香港繁体和台湾正体，均是本地化的 seed tasks 独立生成，并且未使用蒸馏过的turbo模型。 如果您有新的生成结构，欢迎向社区分享。 ================== 这份数据我看过了，问题在于没有去除包含图片和文件相关的内容，这部分会扰乱模型
> @ArtyomZemlyak that should be better documented or maybe not the default behaviour. Currently working on #771 which will improve this by allowing multiple requests to efficiently be processed in...
use third party provider url is https://api.myhispreadnlp.com/ can use gpt4o, but not o1mini,o1 try to access o1, get error: Localai: Bad Request - Unsupported value: 'messages[0].role' does not support 'system'...
in other way , or i can setup proxy to access this thrid party url in configure to replace the url of openai itself? Is this feasible?
Okay, I'll give it a try
2个40g就可以