transformers
fix get_keys_to_not_convert() to return correct modules for full precision inference
#25105

Merged

fix get_keys_to_not_convert() to return correct modules for full precision inference #25105

sgugger merged 3 commits into huggingface:main from ranchlai:fix_get_keys_to_not_convert

ranchlai changed the title ~~fix get_keys_to_not_convert to return correct modules for full precision inference~~ fix get_keys_to_not_convert() to return correct modules for full precision inference 3 years ago

younesbelkada commented on 2023-07-26

younesbelkada commented on 2023-07-27

ranchlai commented on 2023-07-27

younesbelkada commented on 2023-07-27

ranchlai force pushed 3 years ago

younesbelkada commented on 2023-07-28

younesbelkada commented on 2023-07-31

younesbelkada commented on 2023-08-01

add test for `get_keys_to_not_convert`

2dc11e6d

add minimum patch to keep mpt lm_head from 8bit quantization

976248f0

ranchlai force pushed to 976248f0 3 years ago

younesbelkada approved these changes on 2023-08-02

younesbelkada requested a review from

sgugger 3 years ago

sgugger commented on 2023-08-02

add reivsion to

256d4179

sgugger approved these changes on 2023-08-02

sgugger merged 2230d149 into main 3 years ago

Reviewers

sgugger

younesbelkada

Assignees

No one assigned

Labels

None yet

Milestone

No milestone

transformers fix get_keys_to_not_convert() to return correct modules for full precision inference #25105 Merged

fix get_keys_to_not_convert() to return correct modules for full precision inference #25105

transformers
fix get_keys_to_not_convert() to return correct modules for full precision inference
#25105

Merged