support no apllication of chat template for instructions #3
Conversation
This file contains 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
Signed-off-by: Sukriti-Sharma4 <[email protected]>
trl/trainer/sft_trainer.py
Outdated
| @@ -250,7 +251,7 @@ def make_inputs_require_grad(module, input, output): | |||
| if formatting_func is None and dataset_text_field is None: | |||
| # check if dataset has ChatML format or instruction format and is supported | |||
| # if not stays #None | |||
| formatting_func = get_formatting_func_from_dataset(train_dataset, tokenizer) | |||
| formatting_func = get_formatting_func_from_dataset(train_dataset, tokenizer, apply_chat_instruction_template) | |||
|
|
|||
| requires_input_output_keys = False | |||
There was a problem hiding this comment.
This variable name should probably be changed - it doesn't make sense if the keys are prompt / completion
There was a problem hiding this comment.
ya I changed it below and left that one place, thanks for catching it
Signed-off-by: Sukriti-Sharma4 <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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 change exposes a flag
apply_chat_instruction_templateto allow users to control if chat template is applied or not by default to prompt/completion data format supported by SFT Trainer.if flag is set to False, the prompt and completion are just concatenated and masking is applied using length of prompt .