Merged
Conversation
|
The documentation is not available anymore as the PR was closed or merged. |
This was referenced Aug 29, 2023
NLLB-MoE 54B multi-GPU inference throws "Expected all tensors to be on the same device" error
#23385
Closed
parambharat
pushed a commit
to parambharat/transformers
that referenced
this pull request
Sep 26, 2023
* fix encoder hook * style
4 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do ?
Fixes #23385. This PR makes sure that we indeed have a hook with
io_same_device=Trueon the encoder which was not the case previously. If we don't have that, the generation will fail as the output of the encoder won't be on the same device asunfinished_sequences = torch.ones(input_ids.shape[0], dtype=torch.long, device=input_ids.device)and this will lead to an error here:next_tokens = next_tokens * unfinished_sequences + pad_token_id * (1 - unfinished_sequences).