github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-06-06 07:04:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

community: Fixed the procedure of initializing pad_token_id (#29434)

Browse Source 
- **Description:** Add to check pad_token_id and eos_token_id of model
config. It seems that this is the same bug as the HuggingFace TGI bug.
In addition, the source code of
libs/partners/huggingface/langchain_huggingface/llms/huggingface_pipeline.py
also requires similar changes.
- **Issue:** #29431
- **Dependencies:** none
- **Twitter handle:** tell14
This commit is contained in:
Teruaki Ishizaki 2025-01-28 04:54:54 +09:00 committed by GitHub
parent dbb6b7b103
commit 3fce78994e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
libs/community/langchain_community/llms/huggingface_pipeline.py
View File

@ -169,7 +169,16 @@ class HuggingFacePipeline(BaseLLM):
) from e ) from e
if tokenizer.pad_token is None: if tokenizer.pad_token is None:
if model.config.pad_token_id is not None:
tokenizer.pad_token_id = model.config.pad_token_id
elif model.config.eos_token_id is not None and isinstance(
model.config.eos_token_id, int
):
tokenizer.pad_token_id = model.config.eos_token_id tokenizer.pad_token_id = model.config.eos_token_id
elif tokenizer.eos_token_id is not None:
tokenizer.pad_token_id = tokenizer.eos_token_id
else:
tokenizer.add_special_tokens({"pad_token": "[PAD]"})
if ( if (
( (