Improve model card: add metadata, paper info and links

Create README.md

Delete training_args.bin with huggingface_hub

Delete trainer_state.json with huggingface_hub

Delete reward_func_history.json with huggingface_hub

Upload folder using huggingface_hub

initial commit