Some of our models are failing internal integration tests where the AutoTokenizer tries to download files even though they exist locally in the cache.
Modifications
This adds the local_files_only kwarg to the .from_pretrained() call in the fast tokenizer conversion in the launcher, to prevent hf from attempting to download tokenizer files
Result
Local tests show this allows the failing models to successfully load up tokenizers
Motivation
Some of our models are failing internal integration tests where the AutoTokenizer tries to download files even though they exist locally in the cache.
Modifications
This adds the
local_files_onlykwarg to the.from_pretrained()call in the fast tokenizer conversion in the launcher, to prevent hf from attempting to download tokenizer filesResult
Local tests show this allows the failing models to successfully load up tokenizers