gpt2-waiting-for-godot

GPT-2 Fine-tuned on Waiting for Godot

A GPT-2 Medium model fine-tuned on Samuel Beckett's "Waiting for Godot" for generating absurdist theatrical dialogue and prose.

Model Description

This model is part of the Deepstory project, which combines Natural Language Generation, Text-to-Speech, and animation technologies to create interactive storytelling experiences.

The model has been fine-tuned on the complete text of "Waiting for Godot" to capture the distinctive absurdist dialogue style, circular conversations, and existential themes of Beckett's masterpiece.

Model Architecture

This model is based on GPT-2 Medium architecture.

Usage

With Transformers Library

from transformers import GPT2Tokenizer, GPT2LMHeadModel
import torch

# Load model and tokenizer
tokenizer = GPT2Tokenizer.from_pretrained("thetobysiu/gpt2-waiting-for-godot")
model = GPT2LMHeadModel.from_pretrained("thetobysiu/gpt2-waiting-for-godot")

# Generate dialogue
prompt = "VLADIMIR: What do we do now?\nESTRAGON:"
input_ids = tokenizer.encode(prompt, return_tensors='pt')

output = model.generate(
    input_ids=input_ids,
    max_length=200,
    temperature=0.9,
    top_p=0.95,
    top_k=50,
    do_sample=True,
    num_return_sequences=1
)

generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
print(generated_text)

Generation Parameters

For best results when generating Beckett-style dialogue:

Character Format

The model works best when prompted with the character name format used in the play:

• VLADIMIR: or DIDI:
• ESTRAGON: or GOGO:
• POZZO:
• LUCKY:
• BOY:

Training Data

The model was fine-tuned on the complete text of "Waiting for Godot" (En attendant Godot) by Samuel Beckett, including both Act I and Act II.

Training Procedure

The model was fine-tuned using the Hugging Face Transformers library starting from the pre-trained GPT-2 Medium checkpoint.

Intended Use

This model is intended for:

• Experimental theater and creative writing
• Generating absurdist dialogue
• Exploring AI-generated dramatic text
• Research in language models and literary style
• Educational purposes in studying Beckett's style

Characteristics of Generated Text

The model tends to generate text with:

• Circular, repetitive dialogue patterns
• Existential themes and philosophical musings
• Short, punchy exchanges between characters
• References to waiting, time, and uncertainty
• The distinctive rhythm of Beckett's prose

Limitations

• The model is trained on a relatively small corpus (single play)
• Generated dialogue may become repetitive
• Context length is limited to 1024 tokens
• May not capture all nuances of Beckett's theatrical timing
• The training data is based on copyrighted material

Ethical Considerations

• This model is trained on copyrighted material and should be used for personal/research purposes
• Generated content should not be presented as original work by Samuel Beckett
• The model may generate text touching on existential and nihilistic themes

Citation

If you use this model, please cite the Deepstory project:

@misc{deepstory,
  author = {Siu King Wai},
  title = {Deepstory},
  year = {2020},
  publisher = {GitHub},
  url = {https://github.com/thetobysiu/deepstory}
}

License

This model is released under the MIT License. Please note that "Waiting for Godot" is the intellectual property of the Samuel Beckett estate.

Acknowledgments

• OpenAI for the original GPT-2 model
• Hugging Face for the Transformers library
• Samuel Beckett for the original play "Waiting for Godot"