beep-bebop
beep-bebop
change the file's name maybe🙂
> @helloworld53 hello, may you show me how to get this checkpoint of expressive?你好，你能告诉我如何获得这个表情检查站吗？ > > I tried this [forum](https://ai.meta.com/resources/models-and-libraries/seamless-downloads/) of meta, but after click the button 'Accept and continue'...
我的感觉是把短样本拼接为一个长样本
> > Hey! Don't you think that ragging the tensor would be more efficient? > > Yes. I didn't describe it well. I updated the description of this PR. The...
> > > we need to consider the scenario where position IDs are not reset between different short samples, especially for LLM pre-training > > > > > > does...
> 注意到这句话—— The model has a long context length (163840). This may cause OOM errors during the initial memory profiling phase, or result in low performance due to small KV...
> LGTM, indeed this makes sense. can you just update the documentation of this datacolator please! Updated! Feel free to edit if needed:) @ArthurZucker
> [@ZhouZineng](https://github.com/ZhouZineng) Thanks for your feedback! We'll investigate this issue感谢您的反馈！我们会调查此问题 Any conclusion or progress? Thanks for your work!
量化时候用的校验集大概有多少条啊，都是什么类型的，我们是观察到不同大小和类型的校验集对模型输出效果的影响还是挺明显的😂