---
license: llama3
language:
- ja
- en
pipeline_tag: text-generation
---
# Llama 3 neoAI 8B Chat v0.1


![image/png](https://cdn-uploads.huggingface.co/production/uploads/660965845a3d57c6c5c4e21d/obgFZZ6EjxGz4hiopypOd.png)


# Model Description

**Llama 3 neoAI 8B Chat v0.1**は，[Meta-Llama-3-8B-Instruct](https://huggingface.co./meta-llama/Meta-Llama-3-8B-Instruct)をベースとして日本語能力を強化するために事後学習を行なったモデルです．

詳細はブログ記事を参照してください．


![image/png](https://cdn-uploads.huggingface.co/production/uploads/660965845a3d57c6c5c4e21d/c0UvYA3cYtBEcxaD0rvQp.png)

# Evaluation
![image/png](https://cdn-uploads.huggingface.co/production/uploads/660965845a3d57c6c5c4e21d/FIa9mxyb_37ycqMi_59vK.png)

# Usage

## Huggingface

```python
import transformers
import torch

model_id = "neoai-inc/Llama-3-neoAI-8B-Chat-v0.1"

pipeline = transformers.pipeline(
    "text-generation",
    model=model_id,
    model_kwargs={"torch_dtype": torch.bfloat16},
    device_map="auto",
)

messages = [
    {"role": "system", "content": "あなたは誠実で優秀な日本人のアシスタントです。"},
    {"role": "user", "content": "日本で一番高い山は？"},
]

terminators = [
    pipeline.tokenizer.eos_token_id,
    pipeline.tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = pipeline(
    messages,
    max_new_tokens=256,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9,
)
print(outputs[0]["generated_text"][-1])

```


# License

Llama 3 is licensed under the Meta LLAMA 3 Community License, Copyright (c) Meta Platforms, Inc. All Rights Reserved.

# Developed by
![image/png](https://cdn-uploads.huggingface.co/production/uploads/660965845a3d57c6c5c4e21d/fdqexQ3wlCe1bknANNsnt.png)

# Developers

以下アルファベット順

- Gouki Minegishi
- Koki Itai
- Koshiro Terasawa
- Masaki Otsuki
- Ryo Yagi

# How to Cite

```
@misc{neoAI-Llama3-8B-Chat-v0.1, 
      title={Llama-3-neoAI-8B-Chat-v0.1}, 
      url={https://huggingface.co./neoai-inc/Llama-3-neoAI-8B-Chat-v0.1}, 
      author={Gouki Minegishi and Koki Itai and Koshiro Terasawa and Masaki Otsuki and Ryo Yagi},
      year={2024},
}
```