xla
Remove the clamp op when we do symmetric quantization on a tensor
#9465

Merged

Remove the clamp op when we do symmetric quantization on a tensor #9465

vanbasten23 merged 3 commits into master from xiowei/optimize_quantized_matmul

vanbasten23 requested a review from

lsy323 116 days ago

vanbasten23 force pushed from 373f7917 to 69fd267a 116 days ago

vanbasten23 commented on 2025-07-09

vanbasten23 marked this pull request as ready for review 116 days ago

lsy323 approved these changes on 2025-07-09

Fix the unit tests. This commit doesnt change any service code

f9963bbf

get rid of clamp

19b4d782

not to conver to bf16. also ran linter

268b19b6

vanbasten23 force pushed from 69fd267a to 268b19b6 116 days ago

vanbasten23 merged cf156c68 into master 116 days ago

Reviewers

lsy323

Assignees

No one assigned

Labels

None yet

Milestone

No milestone