Modelpublic

etri-xainlp/llama2-12.8b_lora-dpo_v1

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes35downloads

Model Card

etri-xainlp/llama2-12.8blora-dpov1

Model Details

Model Developers ETRI xainlp team

Input text only.

Output text only.

Model Architecture

Base Model meta-llama/Llama-13b-hf

Training Dataset

—sft+lora: 710k instruction-following set

—dpo+lora: 90k user preference set

—We use A100 GPU 80GB * 8, when training.