Alvaro Bartolome

Hi here @amritsingh183, thanks for reporting this issue! We'll investigate this, as we did saw problems when installing `flash-attn` before `vllm` for example, but didn't see problems with `llamacpp-python`, but...

Hi @amritsingh183, do you have any update on this? Internally we tried to reproduce and I was unable while both @gabrielmbmb and @plaguss could indeed reproduce and got similar issues...

To reproduce this @plaguss you can just create an infinite running step as: ```python import time from distilabel.steps import StepInput, step from distilabel.steps.typing import StepOutput @step() def InfiniteStep(*inputs: StepInput) ->...

Hi @davidberenstein1957 can you also share the code from the `Pipeline` or the serialized `Pipeline` instead if that's the only step that fails? i.e. `pipeline.dump()` and share it? Thanks!

Hi @rasdani indeed this is something already in `develop` as completed recently by @plaguss in https://github.com/argilla-io/distilabel/pull/565

Hi here @YueWu0301! Indeed to do that you only need to define an intermediate step within your `Pipeline` so that the `PushToHub` step is called, meaning that it will wait...

Also @YueWu0301 note that we've fully refactored `distilabel` from 0.Y.Z to 1.0.0; so the code you're using does not match the one you're referencing in the `distilabel` docs. So we...

Hi here @YueWu0301 I'll close this issue for the moment, since it is inactive, and we couldn't reproduce on our end! Let us know if you still experience this issue...

Maybe adding default annotation guidelines can be complex if we want this to be extensible, so the easiest may be ~to just add the task definition within the guidelines i.e....

> No. We know is a preference dataset. I'm just talking about something like: > > Rate the quality of the responses to the instructions based on aspects like .......