minor fix in mixtral_moe.py for apple silicon #170
Merged
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
There are two fixes that address errors encountered when ran on Apple M2 macbook with Rosetta enabled
cg123
reviewed
Feb 21, 2024
mergekit/scripts/mixtral_moe.py
Outdated
| @@ -113,7 +113,7 @@ def tokenize_prompts( | |||
| prompts: List[str], tokenizer: transformers.PreTrainedTokenizerBase | |||
| ): | |||
| return tokenizer( | |||
| [tokenizer.bos_token + p for p in prompts], | |||
| [str(tokenizer.bos_token) + p for p in prompts], | |||
There was a problem hiding this comment.
It probably would be better to do tokenizer.bos_token or "" here - otherwise it'll be prepending the string "None" to your prompts if the BOS token isn't set.
mergekit/scripts/mixtral_moe.py
Outdated
| @@ -366,6 +366,7 @@ def build( | |||
| ) | |||
| tokenizer.padding_side = "left" | |||
| tokenizer.pad_token_id = tokenizer.bos_token_id | |||
| tokenizer.pad_token = tokenizer.eos_token | |||
There was a problem hiding this comment.
It'd be great to make this conditional - if the pad token is already set it'd be good to not overwrite it.
address review comments
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Outdated suggestions cannot be applied.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.
There are two fixes that address errors encountered when ran on Apple M2 macbook with Rosetta enabled, not sure if it is a universal problem or specific to my environment (therefore I didn't file an issue due to this uncertainty).
Fix 1 : cast str type for tokenizer.bos_token, this aims for a fix of persistent weird error
"typeError: unsupported operand type(s) for +: 'NoneType' and 'str'"although bos_token is indeed str type.Fix 2 : add pad_token value, per suggestion of
padding_strategyraisedValueError "Asking to pad but the tokenizer does not have a padding token. "(Line 2708 in transformers/tokenization_utils_base.py)