metadata
datasets:
- weqweasdas/ultra_train
base_model:
- OpenRLHF/Llama-3-8b-sft-mixture
Base Model: OpenRLHF/Llama-3-8b-sft-mixture
DPO model: RTO-RL/Llama3-8B-DPO
Reward model: RTO-RL/Llama3.2-1B-RewardModel
Prompt dataset: weqweasdas/ultra_train