|
|
--- |
|
|
license: llama3.1 |
|
|
library_name: transformers |
|
|
pipeline_tag: text-generation |
|
|
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct |
|
|
language: |
|
|
- en |
|
|
- zh |
|
|
tags: |
|
|
- llama-factory |
|
|
- orpo |
|
|
--- |
|
|
|
|
|
> [!CAUTION] |
|
|
> For optimal performance, we refrain from fine-tuning the model's identity. Thus, inquiries such as "Who are you" or "Who developed you" may yield random responses that are not necessarily accurate. |
|
|
|
|
|
|
|
|
# Model Summary |
|
|
|
|
|
llama3.1-8B-Chinese-Chat is an instruction-tuned language model for Chinese & English users with various abilities such as roleplaying & tool-using built upon the Meta-Llama-3.1-8B-Instruct model. |
|
|
|
|
|
Developers: [Shenzhi Wang](https://shenzhi-wang.netlify.app)\*, [Yaowei Zheng](https://github.com/hiyouga)\*, Guoyin Wang (in.ai), Shiji Song, Gao Huang. (\*: Equal Contribution) |
|
|
|
|
|
- License: [Llama-3.1 License](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE) |
|
|
- Base Model: Meta-Llama-3.1-8B-Instruct |
|
|
- Model Size: 8.03B |
|
|
- Context length: 8K |
|
|
|
|
|
# 1. Introduction |
|
|
|
|
|
This is the first model specifically fine-tuned for Chinese & English users based on the [Meta-Llama-3.1-8B-Instruct model](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct). The fine-tuning algorithm used is ORPO [1]. |
|
|
|
|
|
**Compared to the original [Meta-Llama-3.1-8B-Instruct model](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct), our llama3.1-8B-Chinese-Chat model significantly reduces the issues of "Chinese questions with English answers" and the mixing of Chinese and English in responses.** |
|
|
|
|
|
|
|
|
[1] Hong, Jiwoo, Noah Lee, and James Thorne. "Reference-free Monolithic Preference Optimization with Odds Ratio." arXiv preprint arXiv:2403.07691 (2024). |
|
|
|
|
|
Training framework: [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory). |
|
|
|
|
|
Training details: |
|
|
|
|
|
- epochs: 3 |
|
|
- learning rate: 3e-6 |
|
|
- learning rate scheduler type: cosine |
|
|
- Warmup ratio: 0.1 |
|
|
- cutoff len (i.e. context length): 8192 |
|
|
- orpo beta (i.e. $\lambda$ in the ORPO paper): 0.05 |
|
|
- global batch size: 128 |
|
|
- fine-tuning type: full parameters |
|
|
- optimizer: paged_adamw_32bit |
|
|
|
|
|
|
|
|
|
|
|
# 2. Usage |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
|
|
model_id = "shenzhi-wang/Llama3.1-8B-Chinese-Chat" |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
|
model = AutoModelForCausalLM.from_pretrained( |
|
|
model_id, torch_dtype="auto", device_map="auto" |
|
|
) |
|
|
|
|
|
messages = [ |
|
|
{"role": "user", "content": "写一首诗吧"}, |
|
|
] |
|
|
|
|
|
input_ids = tokenizer.apply_chat_template( |
|
|
messages, add_generation_prompt=True, return_tensors="pt" |
|
|
).to(model.device) |
|
|
|
|
|
outputs = model.generate( |
|
|
input_ids, |
|
|
max_new_tokens=8192, |
|
|
do_sample=True, |
|
|
temperature=0.6, |
|
|
top_p=0.9, |
|
|
) |
|
|
response = outputs[0][input_ids.shape[-1]:] |
|
|
print(tokenizer.decode(response, skip_special_tokens=True)) |
|
|
``` |
