---
license: llama2
datasets:
- tatsu-lab/alpaca
- OpenAssistant/oasst1
language:
- zh
- en
library_name: transformers
tags:
- baichuan
- lora
pipeline_tag: text-generation
inference: false
---

A bilingual instruction-tuned LoRA model of https://huggingface.co/meta-llama/Llama-2-13b-hf

- Instruction-following datasets used: alpaca, alpaca-zh, open assistant
- Training framework: https://github.com/hiyouga/LLaMA-Efficient-Tuning

Usage:

```python
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer

tokenizer = AutoTokenizer.from_pretrained("hiyouga/Llama-2-Chinese-13b-chat")
model = AutoModelForCausalLM.from_pretrained("hiyouga/Llama-2-Chinese-13b-chat").cuda()
streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

query = "晚上睡不着怎么办"
template = (
    "A chat between a curious user and an artificial intelligence assistant. "
    "The assistant gives helpful, detailed, and polite answers to the user's questions.\n"
    "Human: {}\nAssistant: "
)

inputs = tokenizer([template.format(query)], return_tensors="pt")
inputs = inputs.to("cuda")
generate_ids = model.generate(**inputs, max_new_tokens=256, streamer=streamer)
```

You could also alternatively launch a CLI demo by using the script in https://github.com/hiyouga/LLaMA-Efficient-Tuning

```bash
python src/cli_demo.py --model_name_or_path hiyouga/Llama-2-Chinese-13b-chat
```

---

Loss curve:

![loss](loss.png)