Quantization made by Richard Erkhov.

nekomata-14b - GGUF

Model creator: https://huggingface.co/rinna/
Original model: https://huggingface.co/rinna/nekomata-14b/

Name	Quant method	Size
nekomata-14b.Q2_K.gguf	Q2_K	5.41GB
nekomata-14b.IQ3_XS.gguf	IQ3_XS	6.12GB
nekomata-14b.IQ3_S.gguf	IQ3_S	6.31GB
nekomata-14b.Q3_K_S.gguf	Q3_K_S	6.31GB
nekomata-14b.IQ3_M.gguf	IQ3_M	6.87GB
nekomata-14b.Q3_K.gguf	Q3_K	7.16GB
nekomata-14b.Q3_K_M.gguf	Q3_K_M	7.16GB
nekomata-14b.Q3_K_L.gguf	Q3_K_L	7.44GB
nekomata-14b.IQ4_XS.gguf	IQ4_XS	7.37GB
nekomata-14b.Q4_0.gguf	Q4_0	7.62GB
nekomata-14b.IQ4_NL.gguf	IQ4_NL	7.68GB
nekomata-14b.Q4_K_S.gguf	Q4_K_S	7.96GB
nekomata-14b.Q4_K.gguf	Q4_K	8.8GB
nekomata-14b.Q4_K_M.gguf	Q4_K_M	8.8GB
nekomata-14b.Q4_1.gguf	Q4_1	8.4GB
nekomata-14b.Q5_0.gguf	Q5_0	9.18GB
nekomata-14b.Q5_K_S.gguf	Q5_K_S	9.34GB
nekomata-14b.Q5_K.gguf	Q5_K	10.14GB
nekomata-14b.Q5_K_M.gguf	Q5_K_M	10.14GB
nekomata-14b.Q5_1.gguf	Q5_1	9.96GB
nekomata-14b.Q6_K.gguf	Q6_K	11.46GB
nekomata-14b.Q8_0.gguf	Q8_0	14.03GB

Original model description:

thumbnail: https://github.com/rinnakk/japanese-pretrained-models/blob/master/rinna.png datasets: - mc4 - wikipedia - EleutherAI/pile - oscar-corpus/colossal-oscar-1.0 - cc100 language: - ja - en tags: - qwen inference: false license: other license_name: tongyi-qianwen-license-agreement license_link: https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT

`rinna/nekomata-14b`

Overview

We conduct continual pre-training of qwen-14b on 66B tokens from a mixture of Japanese and English datasets. The continual pre-training significantly improves the model's performance on Japanese tasks. It also enjoys the following great features provided by the original Qwen model.

The inclusive Qwen vocabulary (vocab size > 150k) enables the model to processs Japanese texts much more efficiently than the previously released youri series.
The model supports a maximum sequence length of 8192.

The name nekomata comes from the Japanese word 猫又/ねこまた/Nekomata, which is a kind of Japanese mythical creature (妖怪/ようかい/Youkai).

Library

The model was trained using code based on aws-neuron/neuronx-nemo-megatron.
Model architecture

A 40-layer, 5120-hidden-size transformer-based language model. Please refer to the Qwen paper for architecture details.
Continual pre-training

The model was initialized with the qwen-14b model and continually trained on around 66B tokens from a mixture of the following corpora
- Japanese CC-100
- Japanese C4
- Japanese OSCAR
- The Pile
- Wikipedia
- rinna curated Japanese dataset
Training Infrastructure

nekomata-14B was trained on 16 nodes of Amazon EC2 trn1.32xlarge instance powered by AWS Trainium purpose-built ML accelerator chip. The pre-training job was completed within a timeframe of approximately 7 days.
Contributors

Benchmarking

Please refer to rinna's LM benchmark page.

How to use the model

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rinna/nekomata-14b", trust_remote_code=True)

# Use GPU with bf16
# model = AutoModelForCausalLM.from_pretrained("rinna/nekomata-14b", device_map="auto", trust_remote_code=True, bf16=True)

# Use GPU with fp16
# model = AutoModelForCausalLM.from_pretrained("rinna/nekomata-14b", device_map="auto", trust_remote_code=True, fp16=True)

# Use CPU
# model = AutoModelForCausalLM.from_pretrained("rinna/nekomata-14b", device_map="cpu", trust_remote_code=True)

# Automatically select device and precision
model = AutoModelForCausalLM.from_pretrained("rinna/nekomata-14b", device_map="auto", trust_remote_code=True)

text = "西田幾多郎は、"
token_ids = tokenizer.encode(text, add_special_tokens=False, return_tensors="pt")

with torch.no_grad():
    output_ids = model.generate(
        token_ids.to(model.device),
        max_new_tokens=200,
        min_new_tokens=200,
        do_sample=True,
        temperature=1.0,
        top_p=0.95,
        pad_token_id=tokenizer.pad_token_id,
        bos_token_id=tokenizer.bos_token_id,
        eos_token_id=tokenizer.eos_token_id
    )

output = tokenizer.decode(output_ids.tolist()[0])
print(output)

Tokenization

The model uses the original Qwen tokenizer. It augments the cl100k tiktoken tokenizer and has a vocabulary size of 151,936. The inclusive vocabulary helps the model to reach a better tokenization efficiency, especially for Japanese texts.

We compared the Qwen tokenizer (as used in nekomata) and the llama-2 tokenizer (as used in youri) on different text collections and found that the Qwen tokenizer achieves a much better byte2token rate (i.e. the average number of tokens produced from 1 byte of text) as following. A lower byte2token rate indicates a better tokenization efficiency.

Tokenizer	Japanese	English	Multilingual
Qwen	0.24	0.27	0.27
llama-2	0.40	0.29	0.36

How to cite

@misc{rinna-nekomata-14b,
    title = {rinna/nekomata-14b},
    author = {Zhao, Tianyu and Kaga, Akio and Sawada, Kei},
    url = {https://huggingface.co/rinna/nekomata-14b}
}

@inproceedings{sawada2024release,
    title = {Release of Pre-Trained Models for the {J}apanese Language},
    author = {Sawada, Kei and Zhao, Tianyu and Shing, Makoto and Mitsui, Kentaro and Kaga, Akio and Hono, Yukiya and Wakatsuki, Toshiaki and Mitsuda, Koh},
    booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
    month = {5},
    year = {2024},
    pages = {13898--13905},
    url = {https://aclanthology.org/2024.lrec-main.1213},
    note = {\url{https://arxiv.org/abs/2404.01657}}
}

License

Tongyi Qianwen LICENSE AGREEMENT