---
license: other
license_name: tongyi-qianwen-license-agreement
license_link: https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT
datasets:
- OpenAssistant/oasst1
- zetavg/ShareGPT-Processed
- augmxnt/ultra-orca-boros-en-ja-v1
language:
- ja
- en
---

<p align="center">
  <img src="https://cdn-uploads.huggingface.co/production/uploads/64c8a2e01c25d2c581a381c1/9CbN4lDGU42c-7DmK_mGM.png" alt="drawing" width="600"/>
</p>

Qwen/Qwen-14B-Chat + Karasu's finetuning datasets

# Demo　・　モデルのデモ

[Model demo ・ モデルのデモ](https://lightblue-qarasu.serveo.net/)

# Blog post・説明の記事

[Blog post・説明の記事](https://note.com/peter_lightblue/n/ne08a7c8cc47a)

# Evaluation

![image/png](https://cdn-uploads.huggingface.co/production/uploads/64b63f8ad57e02621dc93c8b/s3eUP07LOPkxwzNS2p9Yp.png)

In our internal evaluations, we find the Qarasu model to have particularly high performance on the MTーBench benchmark. We are currently awaiting external evaluations.

# How to use


### Hugggingface
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

tokenizer = AutoTokenizer.from_pretrained("lightblue/qarasu-14B-chat-plus-unleashed", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("lightblue/qarasu-14B-chat-plus-unleashed", torch_dtype=torch.bfloat16, device_map="auto", trust_remote_code=True)

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

messages = [{"role": "system", "content": "あなたはAIアシスタントです。"}]
messages.append({"role": "user", "content": "イギリスの首相は誰ですか？"})

prompt = tokenizer.apply_chat_template(conversation=messages, add_generation_prompt=True, tokenize=False)

pipe(prompt, max_new_tokens=100, do_sample=False, temperature=0.0, return_full_text=False)
```


### VLLM
```python
from vllm import LLM, SamplingParams

sampling_params = SamplingParams(temperature=0.0, max_tokens=100)
llm = LLM(model="lightblue/qarasu-14B-chat-plus-unleashed", trust_remote_code=True)

messages = [{"role": "system", "content": "あなたはAIアシスタントです。"}]
messages.append({"role": "user", "content": "イギリスの首相は誰ですか？"})
prompt = llm.llm_engine.tokenizer.apply_chat_template(conversation=messages, add_generation_prompt=True, tokenize=False)
prompts = [prompt]

outputs = llm.generate(prompts, sampling_params)
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
```



# Base checkpoint
[Qwen/Qwen-14B-Chat](https://huggingface.co/Qwen/Qwen-14B-Chat)

# Training datasets (total ~7B)
The same as the 'plus' checkpoint, but with about 6K refusals ("申し訳ありませんが、。。。") filtered out from the category dataset 
* Lightblue's suite of Kujira datasets (unreleased)
* Lightblue's own question-based datasets (unreleased)
* Lightblue's own category-based datasets (unreleased)
* [OASST](https://huggingface.co/datasets/OpenAssistant/oasst1) (Japanese chats only)
* [ShareGPT](https://huggingface.co/datasets/zetavg/ShareGPT-Processed) (Japanese chats only)
* [augmxnt/ultra-orca-boros-en-ja-v1](https://huggingface.co/datasets/augmxnt/ultra-orca-boros-en-ja-v1) (['airoboros', 'slimorca', 'ultrafeedback', 'airoboros_ja_new'] only)

# Developed by

<a href="https://www.lightblue-tech.com">
<img src="https://www.lightblue-tech.com/wp-content/uploads/2023/08/color_%E6%A8%AA%E5%9E%8B-1536x469.png" alt="Lightblue technology logo" width="400"/>
</a>

### Engineers
Peter Devine

Sho Higuchi

### Advisors
Yuuki Yamanaka 

Atom Sonoda

### Project manager
Shunichi Taniguchi

### Dataset evaluator
Renju Aoki