Finance LLM Project
Overview
This project is a fine-tuned language model based on GPT-2 for answering finance-related questions in simple terms.
Model Details
- Base Model: GPT-2
- Fine-tuning Method: LoRA (Parameter Efficient Fine-Tuning)
- Framework: Hugging Face Transformers
Dataset
- Dataset used: finance-alpaca (subset)
- Size: 200 samples
Training
- Batch size: 1
- Epochs: 3
- GPU: Google Colab T4
Example
Input:
Explain stock market in simple terms
Output:
A stock market is a place where people buy and sell shares of companies...
Use Case
- Beginner-friendly finance explanations
- Educational chatbot
Limitations
- Small dataset → may give incomplete answers
- Not production-ready for financial advice
Author
Ayushi
