Swallow-MX-8x7b-NVE-v0.1

Our Swallow-MX-8x7b-NVE-v0.1 model has undergone continuous pre-training from the Mixtral-8x7B-Instruct-v0.1, primarily with the addition of Japanese language data.

Model Details

• Model type: Please refer to Mixtral technical report for details on the model architecture.
• Language(s): Japanese English
• Tokenizer: This model utilizes the same tokenizer as employed by Mixtral-8x7B-Instruct-v0.1.
• Contact: swallow[at]nlp.c.titech.ac.jp

Base Model Performance

Japanese version

Model	Size	JCommonsenseQA	JEMHopQA	NIILC	JSQuAD	XL-Sum	MGSM	WMT20-en-ja	WMT20-ja-en
		4-shot	4-shot	4-shot	4-shot	1-shot	4-shot	4-shot	4-shot
Llama 2	7B	0.3852	0.4240	0.3410	0.7917	0.1905	0.0760	0.1783	0.1738
Swallow	7B	0.4808	0.5078	0.5968	0.8573	0.1830	0.1240	0.2510	0.1511
Swallow-Plus	7B	0.5478	0.5493	0.6030	0.8544	0.1806	0.1360	0.2568	0.1441
Swallow-NVE	7B	0.5433	0.5425	0.5729	0.8684	0.2117	0.1200	0.2405	0.1512
Mistral-7B-v0.1	7B	0.7301	0.4245	0.2722	0.8563	0.2006	0.1760	0.1405	0.1733
Swallow-MS-7b-v0.1	7B	0.8570	0.4915	0.5519	0.8802	0.1988	0.2240	0.2494	0.1667
Llama 2	13B	0.6997	0.4415	0.4170	0.8533	0.2139	0.1320	0.2146	0.1982
Swallow	13B	0.7837	0.5063	0.6398	0.9005	0.2168	0.2040	0.2720	0.1771
Swallow-NVE	13B	0.7712	0.5438	0.6351	0.9030	0.2294	0.2120	0.2735	0.1817
Llama 2	70B	0.8686	0.4656	0.5256	0.9080	0.2361	0.3560	0.2643	0.2398
Swallow	70B	0.9348	0.6290	0.6960	0.9176	0.2266	0.4840	0.3043	0.2298
Swallow-NVE	70B	0.9410	0.5759	0.7024	0.9254	0.2758	0.4720	0.3042	0.2322
Mixtral-8x7B-v0.1	8x7B	0.8347	0.5335	0.3549	0.8847	0.2192	0.3120	0.1970	0.1987
Swallow-MX-8x7b-NVE-v0.1	8x7B	0.9258	0.5843	0.5687	0.9148	0.2589	0.4360	0.2705	0.2074

English version

Model	Size	OpenBookQA	TriviaQA	HellaSwag	SQuAD2.0	XWINO	GSM8K
		8-shot	8-shot	8-shot	8-shot	8-shot	8-shot
Llama 2	7B	0.3580	0.6265	0.5860	0.3207	0.9049	0.1410
Swallow	7B	0.3180	0.4836	0.5308	0.3125	0.8817	0.1130
Swallow-Plus	7B	0.3280	0.4558	0.5259	0.3134	0.8929	0.1061
Swallow-NVE	7B	0.3180	0.5079	0.5329	0.2919	0.8817	0.0986
Mistral-7B-v0.1	7B	0.3660	0.7050	0.6264	0.3799	0.9157	0.3533
Swallow-MS-7b-v0.1	7B	0.3440	0.5976	0.5810	0.3364	0.9037	0.2623
Llama 2	13B	0.3760	0.7255	0.6148	0.3681	0.9140	0.2403
Swallow	13B	0.3500	0.5852	0.5660	0.3406	0.9075	0.2039
Swallow-NVE	13B	0.3460	0.6025	0.5700	0.3478	0.9006	0.1751
Llama 2	70B	0.4280	0.8239	0.6742	0.3770	0.9290	0.5284
Swallow	70B	0.4220	0.7756	0.6458	0.3745	0.9204	0.4867
Swallow-NVE	70B	0.4240	0.7817	0.6439	0.3451	0.9256	0.4943
Mixtral-8x7B-v0.1	8x7B	0.3960	0.7989	0.6678	0.3842	0.9204	0.5747
Swallow-MX-8x7b-NVE-v0.1	8x7B	0.3740	0.7847	0.6520	0.3801	0.9170	0.5694

Please note that Swallow-MX-8x7b-NVE-v0.1 is not derived from Mixtral-8x7B-v0.1, but rather underwent continued pre-training from Mixtral-8x7B-Instruct-v0.1.

Usage

First install additional dependencies in requirements.txt:

pip install -r requirements.txt

Use the base model

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1"
tokenizer = AutoTokenizer.from_pretrained(model_name)

model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
prompt = "東京工業大学の主なキャンパスは、"
input_ids = tokenizer.encode(
    prompt,
    add_special_tokens=False,
    return_tensors="pt"
)
tokens = model.generate(
    input_ids.to(device=model.device),
    max_new_tokens=128,
    temperature=0.99,
    top_p=0.95,
    do_sample=True,
)

out = tokenizer.decode(tokens[0], skip_special_tokens=True)
print(out)

Training Datasets

Continual Pre-Training

The following datasets were used for continual pre-training.

• Algebraic Stack
• Japanese Wikipedia
• RefinedWeb
• Swallow Corpus
• The Pile
• The Vault

Risks and Limitations

The models released here are still in the early stages of our research and development and have not been tuned to ensure outputs align with human intent and safety considerations.

Acknowledgements

We thank Mistral AI for releasing Mixtral-8x7B-Instruct-v0.1 under an open license for others to build on.

Our project is supported by the ABCI Large-scale Language Model Building Support Program of the National Institute of Advanced Industrial Science and Technology.

License

apache-2.0

Authors

Here are the team members:

• From Okazaki Laboratory, the following members:
  • Naoaki Okazaki
  • Sakae Mizuki
  • Hiroki Iida
  • Mengsay Loem
  • Shota Hirai
  • Kakeru Hattori
  • Masanari Ohi
• From YOKOTA Laboratory, the following members:
  • Rio Yokota
  • Kazuki Fujii
  • Taishi Nakamura

How to cite

If you find our work helpful, please feel free to cite us.

@inproceedings{Fujii:COLM2024,
   title={Continual Pre-Training for Cross-Lingual LLM Adaptation:
Enhancing Japanese Language Capabilities},
   author={Kazuki Fujii and Taishi Nakamura and Mengsay Loem and Hiroki
Iida and Masanari Ohi and Kakeru Hattori and Hirai Shota and Sakae
Mizuki and Rio Yokota and Naoaki Okazaki},
   booktitle="Proceedings of the First Conference on Language Modeling",
   series={COLM},
   pages="(to appear)",
   year="2024",
   month=oct,
   address={University of Pennsylvania, USA},
}

@inproceedings{Okazaki:COLM2024,
   title={Building a Large Japanese Web Corpus for Large Language Models},
   author={Naoaki Okazaki and Kakeru Hattori and Hirai Shota and Hiroki
Iida and Masanari Ohi and Kazuki Fujii and Taishi Nakamura and Mengsay
Loem and Rio Yokota and Sakae Mizuki},
   booktitle="Proceedings of the First Conference on Language Modeling",
   series={COLM},
   pages="(to appear)",
   year="2024",
   month=oct,
   address={University of Pennsylvania, USA},
}