phrase-to-story-generator-v3

Trained Story Generation Model

This directory contains the fine-tuned GPT-2 model for phrase-to-story generation.

Model Files

• config.json
• model.safetensors
• tokenizer_config.json
• vocab.json
• merges.txt
• special_tokens_map.json
• generation_config.json
• training_args.bin
• model_info.json

Usage

from transformers import GPT2LMHeadModel, GPT2Tokenizer

# Load model
tokenizer = GPT2Tokenizer.from_pretrained("./model")
model = GPT2LMHeadModel.from_pretrained("./model")

# Generate story
phrase = "magical forest"
prompt = f"Story prompt: {phrase}\n\nStory:"
inputs = tokenizer.encode(prompt, return_tensors="pt")

outputs = model.generate(inputs, max_length=200, temperature=0.8, do_sample=True)
story = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(story.split("Story:")[-1].strip())

Model Details

• Base Model: GPT-2
• Training: Fine-tuned on creative writing prompts
• Use Case: Phrase to story generation
• Created: 2025-09-04 23:33:19

Integration

This model is used by the Streamlit application in this repository. The updated_cloud_pipeline.py loads this model for local inference.

File Sizes

• config.json: 0.00 MB
• model.safetensors: 474.71 MB
• tokenizer_config.json: 0.00 MB
• vocab.json: 1.00 MB
• merges.txt: 0.48 MB
• special_tokens_map.json: 0.00 MB
• generation_config.json: 0.00 MB
• training_args.bin: 0.01 MB
• model_info.json: 0.00 MB

Total size: 476.21 MB