Upload B2 DPO PPO linear model soup adapter

216704c verified 4 days ago

840 Bytes

library_name: peft
tags:
  - peft
  - lora
  - visual-question-answering
base_model: chaoyinshe/llava-med-v1.5-mistral-7b-hf

Medical VQA Merged Adapter

Merge method: linear

Adapter	Weight
`/Users/springwang/Library/Mobile Documents/com~apple~CloudDocs/juniorYear/DeepLearning/Final_523H0173_523H0178/DL_MedicalVQA_Project/checkpoints/B2/checkpoint-3355`	0.4000
`/Users/springwang/Library/Mobile Documents/com~apple~CloudDocs/juniorYear/DeepLearning/Final_523H0173_523H0178/DL_MedicalVQA_Project/checkpoints/DPO/final_adapter`	0.3000
`/Users/springwang/Library/Mobile Documents/com~apple~CloudDocs/juniorYear/DeepLearning/Final_523H0173_523H0178/DL_MedicalVQA_Project/checkpoints/PPO/final_adapter`	0.3000

This adapter was created by merging fine-tuned LoRA adapters without additional training.