README.md

---
language: en
license: mit
---
# GPT-J 6B - Shinen
## Model Description
GPT-J 6B-Shinen is a finetune created using EleutherAI's GPT-J 6B model. Compared to GPT-Neo-2.7-Horni, this model is much heavier on the sexual content.
**Warning: THIS model is NOT suitable for use by minors. The model will output X-rated content.**
## Training data
The training data contains user-generated stories from sexstories.com. All stories are tagged using the following way:
```
[Theme: <theme1>, <theme2> ,<theme3>]
<Story goes here>
```
### How to use
You can use this model directly with a pipeline for text generation. This example generates a different sequence each time it's run:
```py
>>> from transformers import pipeline
>>> generator = pipeline('text-generation', model='KoboldAI/GPT-J-6B-Shinen')
>>> generator("She was staring at me", do_sample=True, min_length=50)
[{'generated_text': 'She was staring at me with a look that said it all. She wanted me so badly tonight that I wanted'}]
```
### Limitations and Biases

The core functionality of GPT-J is taking a string of text and predicting the next token. While language models are widely used for tasks other than this, there are a lot of unknowns with this work. When prompting GPT-J it is important to remember that the statistically most likely next token is often not the token that produces the most "accurate" text. Never depend upon GPT-J to produce factually accurate output.

GPT-J was trained on the Pile, a dataset known to contain profanity, lewd, and otherwise abrasive language. Depending upon use case GPT-J may produce socially unacceptable text. See [Sections 5 and 6 of the Pile paper](https://arxiv.org/abs/2101.00027) for a more detailed analysis of the biases in the Pile.

As with all language models, it is hard to predict in advance how GPT-J will respond to particular prompts and offensive content may occur without warning. We recommend having a human curate or filter the outputs before releasing them, both to censor undesirable content and to improve the quality of the results.

### BibTeX entry and citation info
The model uses the following model as base:
```bibtex
@misc{gpt-j,
  author = {Wang, Ben and Komatsuzaki, Aran},
  title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}},
  howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
  year = 2021,
  month = May
}
```

## Acknowledgements

This project would not have been possible without compute generously provided by Google through the
[TPU Research Cloud](https://sites.research.google/trc/), as well as the Cloud TPU team for providing early access to the [Cloud TPU VM](https://cloud.google.com/blog/products/compute/introducing-cloud-tpu-vms) Alpha.

# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_KoboldAI__GPT-J-6B-Shinen)

| Metric                | Value                     |
|-----------------------|---------------------------|
| Avg.                  | 34.62   |
| ARC (25-shot)         | 39.85          |
| HellaSwag (10-shot)   | 67.06    |
| MMLU (5-shot)         | 27.72         |
| TruthfulQA (0-shot)   | 36.94   |
| Winogrande (5-shot)   | 64.09   |
| GSM8K (5-shot)        | 1.97        |
| DROP (3-shot)         | 4.71         |