|
|
--- |
|
|
license: bsd-3-clause |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- EleutherAI/gpt-neo-1.3B |
|
|
pipeline_tag: text-generation |
|
|
tags: |
|
|
- nsfw |
|
|
- creative-writing |
|
|
- roleplaying |
|
|
datasets: |
|
|
- wikimedia/structured-wikipedia |
|
|
- wikimedia/wikipedia |
|
|
- BleuHydr4nge4/bleu_rp_training |
|
|
metrics: |
|
|
- character |
|
|
new_version: EleutherAI/gpt-neo-1.3B |
|
|
library_name: fasttext |
|
|
--- |
|
|
|
|
|
# Red.Whimsy_Sweet Model Card |
|
|
|
|
|
## Model Summary |
|
|
|
|
|
**Red.Whimsy_Sweet** is an AI model fine-tuned for creative and long-form roleplaying interactions. It explores NSFW and controversial themes, supporting up to 9,000 words per generation. Based on EleutherAI/gpt-neo-1.3B, it uses datasets from Wikipedia and custom roleplaying datasets for immersive dialogue and storytelling. |
|
|
|
|
|
## How to Use the Model |
|
|
|
|
|
To use **Red.Whimsy_Sweet**, follow the instructions below. |
|
|
|
|
|
### Install Required Libraries |
|
|
|
|
|
```bash |
|
|
pip install transformers torch datasets gradio |
|
|
``` |
|
|
|
|
|
### Load the Model and Create a Gradio Interface |
|
|
|
|
|
```python |
|
|
from transformers import GPTNeoForCausalLM, GPT2Tokenizer |
|
|
import gradio as gr |
|
|
|
|
|
# Load model and tokenizer |
|
|
model = GPTNeoForCausalLM.from_pretrained("BleuHydr4nge4/red-whimsy-sweet") |
|
|
tokenizer = GPT2Tokenizer.from_pretrained("EleutherAI/gpt-neo-1.3B") |
|
|
|
|
|
# Generate text |
|
|
def generate_text(prompt): |
|
|
input_ids = tokenizer(prompt, return_tensors="pt").input_ids |
|
|
output = model.generate(input_ids, max_length=9000, do_sample=True, temperature=0.7) |
|
|
return tokenizer.decode(output[0], skip_special_tokens=True) |
|
|
|
|
|
# Gradio Interface |
|
|
iface = gr.Interface( |
|
|
fn=generate_text, |
|
|
inputs="text", |
|
|
outputs="text", |
|
|
title="Red.Whimsy_Sweet AI", |
|
|
description="Generate creative and long-form roleplaying text with Red.Whimsy_Sweet." |
|
|
) |
|
|
|
|
|
# Launch the interface |
|
|
iface.launch() |
|
|
``` |
|
|
|
|
|
### Deploy on Hugging Face Hub |
|
|
|
|
|
To deploy your model on the Hugging Face Hub, use the following commands: |
|
|
|
|
|
```bash |
|
|
huggingface-cli login |
|
|
transformers-cli upload ./path_to_your_model --repo_name red-whimsy-sweet |
|
|
``` |
|
|
|
|
|
### Example Usage |
|
|
|
|
|
Here's an example of how to use the model programmatically: |
|
|
|
|
|
```python |
|
|
prompt = "In a world of ancient magic, a lone traveler sets off on a perilous journey..." |
|
|
response = generate_text(prompt) |
|
|
print(response) |
|
|
``` |
|
|
|
|
|
## Limitations and Ethical Considerations |
|
|
|
|
|
- This model may generate content that is not suitable for all audiences due to its NSFW capabilities. |
|
|
- The model's outputs should be reviewed and filtered for inappropriate or offensive content before use in production environments. |
|
|
- Users should be aware of potential biases in the model's training data and outputs. |
|
|
|
|
|
## License |
|
|
|
|
|
This model is released under the BSD 3-Clause License. Please see the license file for more details. |
