---
inference: false
license: other
---

# AlpaGasus-7B Model Card

## Model Details

This is an **unofficial** implementation of AlpaGasus-7B, which is a chat assistant trained by fine-tuning LLaMA on a Claud-filtered Alpaca dataset with around 5K triplets.

- **Developed by:** [gpt4life](https://github.com/gpt4life)
- **Model type:** An auto-regressive language model based on the transformer architecture.
- **License:** Non-commercial license
- **Finetuned from model:** [LLaMA-7B](https://huggingface.co/elinas/llama-7b-hf-transformers-4.29).

Please see the original LLaMA [license](https://github.com/facebookresearch/llama/blob/main/LICENSE) before using this model.

### Model Sources

- **Repository:** https://github.com/gpt4life/alpagasus
- **Paper:** https://arxiv.org/pdf/2307.08701.pdf 

## Training Details

AlpaGasus-7B is fine-tuned from LLaMA-7B with supervised instruction fine-tuning on the filtered [Alpaca dataset](https://github.com/gpt4life/alpagasus/blob/main/rating/alpaca_filtered_data.json).