YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co./docs/hub/model-cards#model-card-metadata)
The Llama3-8b-based Reward Model was trained using OpenRLHF and a combination of datasets available at https://huggingface.co./datasets/OpenLLMAI/preference_dataset_mixture2_and_safe_pku.
Base model: https://huggingface.co./OpenRLHF/Llama-3-8b-sft-mixture
Cosine Scheduler
Learning Rate: 9e-6
Warmup Ratio: 0.03
Batch Size: 256
Epoch: 1
- Downloads last month
- 8,327