Hugging Face's logo

tokyotech-llm
/
Swallow-70b-instruct-v0.1

Text Generation
Transformers
Safetensors
English
Japanese
llama
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
Swallow-70b-instruct-v0.1 / README.md
stjohn2007's picture
stjohn2007
Update README.md
fec59af verified
preview code
|
raw
history blame
11.1 kB
metadata
language:
  - en
  - ja
library_name: transformers
pipeline_tag: text-generation
license: llama2
model_type: llama

Swallow

Our Swallow model has undergone continual pre-training from the Llama 2 family, primarily with the addition of Japanese language data. The tuned versions use supervised fine-tuning (SFT). Links to other models can be found in the index.

Model Release Updates

We are excited to share the release schedule for our latest models:

Swallow Model Index

Model Swallow-hf Swallow-instruct-hf Swallow-instruct-v0.1
7B Link Link Link
7B-Plus Link N/A N/A
13B Link Link Link
70B Link Link Link

Swallow Model Index NVE (No Vocabulary Expansion)

Model Swallow-NVE-hf Swallow-NVE-instruct-hf
7B Link Link
13B Link N/A
70B Link Link

logo

This repository provides large language models developed by TokyoTech-LLM.

Model Details

  • Model type: Please refer to LLaMA-2 technical report for details on the model architecture.
  • Language(s): Japanese English
  • Tokenizer: This model employs a tokenizer that features a broadened vocabulary based on Japanese data. This allows for a more efficient representation of text using fewer tokens, leading to a notably faster inference process.
  • Contact: swallow[at]nlp.c.titech.ac.jp

Instruct Model Performance

MT-Bench JA

  • NOTE that the models with the v0.1 suffix are newer versions compared to their original counterparts with the hf.
  • We report overall (i.e., average over scores of the first and second turns), first, and second turn scores.

Overall

Model Average Writing Roleplay Reasoning Math Coding Extraction STEM Humanities
Swallow-7b-instruct-v0.1 0.3435 0.4450 0.4720 0.1853 0.1920 0.2204 0.3015 0.4594 0.4720
Swallow-7b-instruct-hf 0.1833 0.2205 0.1975 0.1593 0.1045 0.1282 0.2672 0.1908 0.1980
Swallow-13b-instruct-v0.1 0.3669 0.4816 0.5562 0.2769 0.1020 0.1505 0.4179 0.4347 0.5150
Swallow-13b-instruct-hf 0.2004 0.1932 0.2552 0.1507 0.1184 0.1285 0.2641 0.2434 0.2500
Swallow-70b-instruct-v0.1 0.4513 0.4822 0.5353 0.3497 0.3492 0.2668 0.5553 0.4955 0.5767
Swallow-70b-instruct-hf 0.3259 0.2925 0.4283 0.3447 0.1562 0.1856 0.5634 0.3315 0.3071

First Turn

Model Average Writing Roleplay Reasoning Math Coding Extraction STEM Humanities
Swallow-7b-instruct-v0.1 0.3829 0.4960 0.4800 0.2220 0.2820 0.2164 0.3220 0.5440 0.4980
Swallow-7b-instruct-hf 0.2216 0.2830 0.2150 0.1590 0.1080 0.1470 0.3542 0.2450 0.2650
Swallow-13b-instruct-v0.1 0.3948 0.5400 0.5220 0.3020 0.1040 0.1760 0.5040 0.5180 0.4920
Swallow-13b-instruct-hf 0.2304 0.2460 0.2640 0.1610 0.1360 0.1330 0.3070 0.3010 0.2950
Swallow-70b-instruct-v0.1 0.4849 0.5720 0.5020 0.4780 0.3680 0.2467 0.5400 0.5720 0.5960
Swallow-70b-instruct-hf 0.3631 0.3420 0.4007 0.4220 0.1580 0.2044 0.6120 0.4280 0.3360

Second Turn

Model Average Writing Roleplay Reasoning Math Coding Extraction STEM Humanities
Swallow-7b-instruct-v0.1 0.3059 0.3940 0.4640 0.1441 0.1000 0.2253 0.2811 0.3724 0.4449
Swallow-7b-instruct-hf 0.1432 0.1567 0.1798 0.1603 0.1010 0.1085 0.1767 0.1343 0.1295
Swallow-13b-instruct-v0.1 0.3353 0.4213 0.5911 0.2516 0.1000 0.1244 0.3194 0.3473 0.5394
Swallow-13b-instruct-hf 0.1692 0.1364 0.2453 0.1401 0.1000 0.1237 0.2199 0.1850 0.2050
Swallow-70b-instruct-v0.1 0.4179 0.3913 0.5689 0.2184 0.3280 0.2884 0.5711 0.4171 0.5562
Swallow-70b-instruct-hf 0.2872 0.2398 0.4564 0.2647 0.1540 0.1676 0.5118 0.2311 0.2762

Evaluation Benchmarks

MT-Bench JA

We used Japanese MT-Bench to assess the instruction-following capabilities of models. We utilized the following settings:

Usage

First install additional dependencies in requirements.txt:

pip install -r requirements.txt

Instruction format Ver0.1

This format must be adhered to strictly, as deviations may result in less optimal outputs from the model.

The template used to construct a prompt for the Instruct model is specified as follows:

<s>[INST] <<SYS>>\n{Instruction}\n<</SYS>>\n\n{USER_MESSAGE_1} [INST] {BOT_MESSAGE_1} </s>[INST] {USER_MESSAGE_2}[/INST]

Please be aware that <s> and </s> are special tokens used for the beginning of string (BOS) and end of string (EOS), respectively, while [INST] and [/INST] are considered regular strings.

For the "{Instruction}" part, We recommend using "あなたは誠実で優秀な日本人のアシスタントです。"

Use the instruct model Ver0.1 

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "tokyotech-llm/Swallow-70b-instruct-v0.1"
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_name)

device = "cuda"

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": "東京工業大学の主なキャンパスについて教えてください"}
]

encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")

model_inputs = encodeds.to(device)
model.to(device)

generated_ids = model.generate(model_inputs, max_new_tokens=128, do_sample=True)
decoded = tokenizer.batch_decode(generated_ids)
print(decoded[0])

Training Datasets

Instruction Tuning Ver0.1

The following datasets were used for the instruction tuning.

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 Meta Research for releasing Llama 2 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

Llama 2 is licensed under the LLAMA 2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved.

Authors

Here are the team members: