Skip to content

Fix (example/brevitas): GPTQ should be applied before calibration.#165

Open
nickfraser wants to merge 1 commit intohuggingface:mainfrom
nickfraser:fix/gptq
Open

Fix (example/brevitas): GPTQ should be applied before calibration.#165
nickfraser wants to merge 1 commit intohuggingface:mainfrom
nickfraser:fix/gptq

Conversation

@nickfraser
Copy link
Copy Markdown
Contributor

When static activation quantization is used, calibration should be applied before GPTQ. Also, under some edge cases, a single inference run needs to occur before GPTQ is applied.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant