---
license: mit
language:
- ja
library_name: transformers
pipeline_tag: text-generation
tags:
- gpt_neox
- gpt-neox
- japanese
inference:
  parameters:
    max_new_tokens: 32
    do_sample: false
    repetition_penalty: 1.1
---

# stockmark/gpt-neox-japanese-1.4b

This repository provides a GPT-NeoX based model with 1.4B parameters pre-trained on Japanese corpus of about 20B tokens. This model is developed by [Stockmark Inc.](https://stockmark.co.jp/)

## How to use

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

# Use torch.bfloat16 for A100 GPU and torch.flaot16 for the older generation GPUs
torch_dtype = torch.bfloat16 if torch.cuda.is_available() and hasattr(torch.cuda, "is_bf16_supported") and torch.cuda.is_bf16_supported() else torch.float16

model = AutoModelForCausalLM.from_pretrained("stockmark/gpt-neox-japanese-1.4b", device_map="auto", torch_dtype=torch_dtype)
tokenizer = AutoTokenizer.from_pretrained("stockmark/gpt-neox-japanese-1.4b")

inputs = tokenizer("自然言語処理は", return_tensors="pt").to(model.device)
with torch.no_grad():
    tokens = model.generate(
        **inputs,
        max_new_tokens=128,
        repetition_penalty=1.1
    )
    
output = tokenizer.decode(tokens[0], skip_special_tokens=True)
print(output)
```

## Example:

- LoRA tuning: https://huggingface.co/stockmark/gpt-neox-japanese-1.4b/blob/main/notebooks/LoRA.ipynb

## Training dataset
- Japanese Web Corpus (ja): 8.6B tokens (This dataset will not be released.)
- Wikipedia (ja): 0.88B tokens
- CC100 (ja): 10.5B tokens

## Training setting
- Trained using HuggingFace Trainer and DeepSpeed (ZeRO-2)
- 8 A100 GPUs (40GB) at ABCI
- Mixed Precision (BF16)

## License
[The MIT license](https://opensource.org/licenses/MIT)

## Developed by
[Stockmark Inc.](https://stockmark.co.jp/)

## Author
[Takahiro Omi](https://huggingface.co/omitakahiro)