transformers: Add eager attention option to make Gemma-2 work properly (#6188)

This commit is contained in:
GralchemOz 2024-07-01 23:08:08 +08:00 committed by GitHub
parent 19a56dd538
commit 8a39f579d8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
5 changed files with 7 additions and 0 deletions

View file

@ -69,6 +69,7 @@ def list_model_elements():
'trust_remote_code',
'no_use_fast',
'use_flash_attention_2',
'use_eager_attention',
'load_in_4bit',
'compute_dtype',
'quant_type',