Self-Generated Dataset Creation Method for DPO Learning
#2
by
ehartford
- opened
Would love to check out your methods and data that you discuss on the model card.
Same here, forgive my ignorance, I couldn't find where SGD was mentioned in the nox framework https://github.com/davidkim205/nox
Also, congrates on ranking #1 on the openllm leaderboard. Looking forward releasing the v1.0 version of Rhea
I'm sorry. SGD is under research and cannot be made public yet. nox is a framework for sft and dpo.
davidkim205
changed discussion status to
closed