Sarashina1-7B

This repository provides Japanese language models trained by SB Intuitions.

How to use

Please set use_fast=False to use our tokenizer properly.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, set_seed
 
model = AutoModelForCausalLM.from_pretrained("sbintuitions/sarashina1-7b", torch_dtype=torch.float16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("sbintuitions/sarashina1-7b", use_fast=False)
generator = pipeline("text-generation", model=model, tokenizer=tokenizer)
set_seed(123)
 
text = generator(
    "おはようございます、今日の天気は",
    max_length=30,
    do_sample=True,
    pad_token_id=tokenizer.pad_token_id,
    num_return_sequences=3,
)
 
for t in text:
  print(t)
 
# These examples are generated by sarashina1-7b parameters model
# {'generated_text': 'おはようございます、今日の天気は晴れ!!最高気温は15度、最低気温は7度です。今日も1日頑張りましょー♪写真は、去年'}
# {'generated_text': 'おはようございます、今日の天気は曇り:cloud:です。 雨予報なので、洗濯物は家の中へ。 :city_sunrise:の見える時間。 今日は'}
# {'generated_text': 'おはようございます、今日の天気は、晴れ、気温も10度以上に上がるそうです、お日様が当たっていると15度くらいになると思います、朝の'}

Configuration

Parameters	Vocab size	Training tokens	Architecture	Position type	Layers	Hidden dim	Attention heads
7B	51200	1.0T	GPTNeoX	RoPE	32	4096	32
13B	51200	1.0T	GPTNeoX	RoPE	40	5120	40
65B	51200	800B	GPTNeoX	RoPE	80	8192	64

Training Corpus

We used a Japanese portion of the Common Crawl corpus, which is the largest Web corpus, as our training dataset. To clean the training corpus, we used CCNet and HojiChar. After cleaning, our corpus contains about 550B tokens.

Tokenization

We use a sentencepiece tokenizer with a unigram language model and byte-fallback. We do not apply pre-tokenization with Japanese tokenizer. Thus, a user may directly feed raw sentences into the tokenizer.

Ethical Considerations and Limitations

Sarashina1 has not been tuned to follow an instruction yet. Therefore, sarashina1 might generate some meaningless sequences, some inaccurate instances or biased/objectionable outputs. Before using sarashina1, we would like developers to tune models based on human preferences and safety considerations.

License

MIT License