AIKipling

Kipling-GPT-2: Fine-Tuned GPT-2 on Rudyard Kipling's Works

This repository contains a GPT-2 model fine-tuned on the works of Rudyard Kipling, designed to generate text in his unique literary style. The model can be used for generating Kipling-inspired prose, poetry, or creative writing.

Model Details

• Base Model: GPT-2
• Dataset: Works of Rudyard Kipling (custom dataset)
• Training Epochs: 3
• Fine-Tuning Task: Language modeling
• Output Size: ~500MB (GPT-2 small)

Repository File Structure

---

How to Use

You can load the fine-tuned model using the following code:

from transformers import GPT2LMHeadModel, GPT2Tokenizer

# Load the fine-tuned model and tokenizer
model = GPT2LMHeadModel.from_pretrained("path_to_model_directory")
tokenizer = GPT2Tokenizer.from_pretrained("path_to_model_directory")

# Example prompt
prompt = "If you can keep your head when all about you"
inputs = tokenizer(prompt, return_tensors="pt")

# Generate text
outputs = model.generate(
    inputs["input_ids"],
    max_length=100,
    num_return_sequences=1,
    do_sample=True,
    top_k=50,
    top_p=0.95,
    temperature=0.9
)

# Print generated text
print(tokenizer.decode(outputs[0], skip_special_tokens=True))