-
Notifications
You must be signed in to change notification settings - Fork 608
remove short-term quant #305
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
This pull request was exported from Phabricator. Differential Revision: D48761858 |
5955897
to
dea2a5c
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
dea2a5c
to
8133984
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
8133984
to
99b94a2
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
99b94a2
to
897d6c9
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
897d6c9
to
de6f5d2
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
Summary: Deprecating Short Term quant flow in new testing infra Reviewed By: salilsdesai Differential Revision: D48761858
de6f5d2
to
bd5a798
Compare
This pull request was exported from Phabricator. Differential Revision: D48761858 |
This pull request has been merged in 3e78b2f. |
Summary: It handles the case hhen params.json is explicitly given and the json doesn't contain mention about tokenizer config, but the command line does. Test Plan: python torchchat.py generate --device cpu --checkpoint-path /Users/mnachin/models/Meta-Llama-3-8B/original/consolidated.00.pth --params-path=/Users/mnachin/models/Meta-Llama-3-8B/original/params.json --temperature 0 --tiktoken
Summary: Deprecating Short Term quant flow in new testing infra
Reviewed By: salilsdesai
Differential Revision: D48761858