Quantization made by Richard Erkhov.

Github

Discord

Request more models

japanese-stablelm-base-beta-7b - GGUF

• Model creator: https://huggingface.co/stabilityai/
• Original model: https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b/

Name	Quant method	Size
japanese-stablelm-base-beta-7b.Q2K.gguf	Q2K	2.36GB
japanese-stablelm-base-beta-7b.IQ3XS.gguf	IQ3XS	2.6GB
japanese-stablelm-base-beta-7b.IQ3S.gguf	IQ3S	2.75GB
japanese-stablelm-base-beta-7b.Q3KS.gguf	Q3K_S	2.75GB
japanese-stablelm-base-beta-7b.IQ3M.gguf	IQ3M	2.9GB
japanese-stablelm-base-beta-7b.Q3K.gguf	Q3K	3.07GB
japanese-stablelm-base-beta-7b.Q3KM.gguf	Q3K_M	3.07GB
japanese-stablelm-base-beta-7b.Q3KL.gguf	Q3K_L	3.35GB
japanese-stablelm-base-beta-7b.IQ4XS.gguf	IQ4XS	3.4GB
japanese-stablelm-base-beta-7b.Q40.gguf	Q40	3.56GB
japanese-stablelm-base-beta-7b.IQ4NL.gguf	IQ4NL	3.58GB
japanese-stablelm-base-beta-7b.Q4KS.gguf	Q4K_S	3.59GB
japanese-stablelm-base-beta-7b.Q4K.gguf	Q4K	3.8GB
japanese-stablelm-base-beta-7b.Q4KM.gguf	Q4K_M	3.8GB
japanese-stablelm-base-beta-7b.Q41.gguf	Q41	3.95GB
japanese-stablelm-base-beta-7b.Q50.gguf	Q50	4.33GB
japanese-stablelm-base-beta-7b.Q5KS.gguf	Q5K_S	4.33GB
japanese-stablelm-base-beta-7b.Q5K.gguf	Q5K	4.45GB
japanese-stablelm-base-beta-7b.Q5KM.gguf	Q5K_M	4.45GB
japanese-stablelm-base-beta-7b.Q51.gguf	Q51	4.72GB
japanese-stablelm-base-beta-7b.Q6K.gguf	Q6K	5.15GB
japanese-stablelm-base-beta-7b.Q80.gguf	Q80	6.67GB

Original model description:

---

language:

• ja

tags:

• japanese-stablelm
• causal-lm

pipeline_tag: text-generation
datasets:

• wikipedia
• mc4
• cc100
• oscar-corpus/OSCAR-2301
• oscar-corpus/OSCAR-2201
• cerebras/SlimPajama-627B

license:

• llama2

extragatedfields:
Name: text
Email: text
Country: text
Organization or Affiliation: text
I allow Stability AI to contact me about information related to its models and research: checkbox

---

Japanese-StableLM-Base-Beta-7B

!A cute robot wearing a kimono writes calligraphy with one single brush

A cute robot wearing a kimono writes calligraphy with one single brush — Stable Diffusion XL

Model Description

japanese-stablelm-base-beta-7b is a 7B-parameter decoder-only language model based on Llama-2-7b that has been fine-tuned on a diverse collection of Japanese data, with the intent of maximizing downstream performance on Japanese language tasks.

For an instruction-following model, check Japanese-StableLM-Instruct-Beta-7B. The base and instruct models are also available in larger 70b sizes. For a model that has faster inference times, see Japanese-StableLM-Base-JAVocab-Beta-7B, or the instruction-following version.

Usage

First install additional dependencies in requirements.txt:

pip install -r requirements.txt

Then start generating text with japanese-stablelm-base-beta-7b by using the following code snippet:

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_name = "stabilityai/japanese-stablelm-base-beta-7b"
tokenizer = AutoTokenizer.frompretrained(modelname)

The next line may need to be modified depending on the environment
model = AutoModelForCausalLM.frompretrained(modelname, torchdtype=torch.float16, lowcpumemusage=True, device_map="auto")

prompt = """
AI で科学研究を加速するには、
""".strip()

input_ids = tokenizer.encode(
    prompt,
    addspecialtokens=True,
    return_tensors="pt"
)

this is for reproducibility.
feel free to change to get different result
seed = 23  
torch.manual_seed(seed)

tokens = model.generate(
    input_ids.to(device=model.device),
    maxnewtokens=128,
    temperature=0.99,
    top_p=0.95,
    do_sample=True,
)

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

We suggest playing with different generation config (topp, repetitionpenalty etc) to find the best setup for your tasks. For example, use higher temperature for roleplay task, lower temperature for reasoning.

Model Details

• Model type: japanese-stablelm-base-beta-7b model is an auto-regressive language model based on the Llama2 transformer architecture.
• Language(s): Japanese
• License: Llama2 Community License.
• Contact: For questions and comments about the model, please join Stable Community Japan. For future announcements / information about Stability AI models, research, and events, please follow https://twitter.com/StabilityAI_JP.

Training Dataset

Roughly 100B tokens from a mixture of the following corpora were used for continued pre-training.

• Japanese/English Wikipedia
• Japanese mc4
• Japanese CC-100
• Japanese OSCAR
• SlimPajama (excluding the Books3 subset)

Use and Limitations

Intended Use

The model is intended to be used by all individuals as a foundation for application-specific fine-tuning without strict limitations on commercial use.

Limitations and bias

The pre-training dataset may have contained offensive or inappropriate content even after applying data cleansing filters which can be reflected in the model generated text. We recommend users exercise reasonable caution when using these models in production systems. Do not use the model for any applications that may cause harm or distress to individuals or groups.

Authors

This model was developed by the Research & Development team at Stability AI Japan, and the development was co-led by Takuya Akiba and Meng Lee. The members of the team are as follows:

• Meng Lee
• Fujiki Nakamura
• Makoto Shing
• Paul McCann
• Takuya Akiba
• Naoki Orii

Acknowledgements

We thank Meta Research for releasing Llama 2 under an open license for others to build on.

We are grateful for the contributions of the EleutherAI Polyglot-JA team in helping us to collect a large amount of pre-training data in Japanese. Polyglot-JA members includes Hyunwoong Ko (Project Lead), Fujiki Nakamura (originally started this project when he commited to the Polyglot team), Yunho Mo, Minji Jung, KeunSeok Im, and Su-Kyeong Jang.

We are also appreciative of AI Novelist/Sta (Bit192, Inc.) and the numerous contributors from Stable Community Japan for assisting us in gathering a large amount of high-quality Japanese textual data for model training.

How to cite

@misc{JapaneseStableLMBaseBeta7B, 
      url={https://huggingface.co/stabilityai/japanese-stablelm-base-beta-7b}, 
      title={Japanese StableLM Base Beta 7B}, 
      author={Lee, Meng and Nakamura, Fujiki and Shing, Makoto and McCann, Paul and Akiba, Takuya and Orii, Naoki}
}