[V5] Return a BatchEncoding dict from apply_chat_template by default again#42567
Merged
Rocketknight1 merged 11 commits intomainfrom Dec 4, 2025
Merged
[V5] Return a BatchEncoding dict from apply_chat_template by default again#42567Rocketknight1 merged 11 commits intomainfrom
Rocketknight1 merged 11 commits intomainfrom
Conversation
…nderlying tokenizer
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: blenderbot, bloom, cohere, gpt2, gpt_sw3 |
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Member
Author
|
cc @LysandreJik - this was one of the V5 PRs before, do I need to do anything special with this one, or can we just merge it to |
zucchini-nlp
approved these changes
Dec 4, 2025
Member
zucchini-nlp
left a comment
zucchini-nlp
left a comment
There was a problem hiding this comment.
great, it was already approved once so lgtm 😄
Comment on lines
+3268
to
+3269
| if not tokenize: | ||
| return_dict = False # dicts are only returned by the tokenizer anyway |
Member
There was a problem hiding this comment.
makes me wonder, do we need to support a combination of tokenize=True, return_dict=False or can we deprecate/remove return_dict over time? Can't think of cases when users want a list of tokens as output
Member
Author
There was a problem hiding this comment.
Maybe we can get rid of it over time, but I think it's fine as a backward compatibility flag for now!
Member
There was a problem hiding this comment.
sure, i meant after v5 + several more minor releases, and if users are fine with it
sarathc-cerebras
pushed a commit
to sarathc-cerebras/transformers
that referenced
this pull request
Dec 7, 2025
…again (huggingface#42567) * Flip the default return type for `apply_chat_template` to match the underlying tokenizer * Remove test_tokenization_for_chat tests, which no longer do anything useful * Remove test_tokenization_for_chat tests, which no longer do anything useful * Fix test_encode_message tests * Fix test_encode_message tests * nit fix * Trigger tests * Remove test_tokenization_for_chat * make fixup * Add a little test to make sure that doesn't happen again * make fixup
SangbumChoi
pushed a commit
to SangbumChoi/transformers
that referenced
this pull request
Jan 23, 2026
…again (huggingface#42567) * Flip the default return type for `apply_chat_template` to match the underlying tokenizer * Remove test_tokenization_for_chat tests, which no longer do anything useful * Remove test_tokenization_for_chat tests, which no longer do anything useful * Fix test_encode_message tests * Fix test_encode_message tests * nit fix * Trigger tests * Remove test_tokenization_for_chat * make fixup * Add a little test to make sure that doesn't happen again * make fixup
5 tasks
erictang000
added a commit
to NovaSky-AI/SkyRL
that referenced
this pull request
Mar 20, 2026
#1325) # Overview This PR makes 2 changes for transformers-v5 compatibility: - Sets `return_dict=False` where needed for transformers-v5 compatibility - this can be merged prior to explicitly upgrading to transformers v5 in the pyproject.toml, since vllm still does not technically fully support it in the latest release. (This change should be backwards compatible with transformers v4.*, since the behavior prior to v5 was that the default value for `return_dict` was `None`, which was interpreted as False.) - check if fsdp_transformer_layer_cls_to_wrap is a set for v5 compatibility while maintaining backwards compatibility huggingface/transformers breaking PR `return_dict=false` PR for v5: huggingface/transformers#42567 ## Validation Checked all CPU tests are passing both with transformers < 5.0.0 and for transformers==5.3.0, and checked that `examples/train/gsm8k/run_gsm8k` runs for both old/new transformers. <!-- devin-review-badge-begin --> --- <a href="https://app.devin.ai/review/novasky-ai/skyrl/pull/1325" target="_blank"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://static.devin.ai/assets/gh-open-in-devin-review-dark.svg?v=1"> <img src="https://static.devin.ai/assets/gh-open-in-devin-review-light.svg?v=1" alt="Open with Devin"> </picture> </a> <!-- devin-review-badge-end -->
5 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.
This is basically PR #41626 again! Some of it got clobbered in the tokenizer refactor, but it's just as good the second time.