Skip to content
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.

Sign up for GitHub

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

Jump to bottom

DPO speed #34

Open
surcyf123 opened this issue Sep 10, 2023 · 1 comment
Open

DPO speed #34

surcyf123 opened this issue Sep 10, 2023 · 1 comment
Assignees
@surcyf123
Labels
enhancement New feature or request

Comments

@surcyf123
Copy link
Owner

need to benchmark my quantization against the OG model to see variance, determine how much accuracy we sacrifice and how much speed we gain.
@surcyf123 surcyf123 added the enhancement New feature or request label Sep 10, 2023
@surcyf123 surcyf123 self-assigned this Sep 10, 2023
@surcyf123
Copy link
Owner Author

surcyf123 commented Sep 10, 2023 •
edited
Loading

Initial benchmarking:
Using full context length and /root/dataset_enrichment/dataset/benchmarking_completions.json
using code from this commit: 14aa8ad
Run 1: Average time per prompt-completion pair for DirectPreferenceRewardModel: 0.0949 seconds
Run 2: Average time per prompt-completion pair for DirectPreferenceRewardModel: 0.0949 seconds

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

@surcyf123 surcyf123

Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@surcyf123