FinSight AI - Financial Advisory Chatbot

A fine-tuned version of SmolLM2-1.7B optimized for financial advice and discussion

Model Details

Base Model: HuggingFaceTB/SmolLM2-1.7B-Instruct
Task: Financial Advisory and Discussion
Training Data: Curated dataset of 10,896 financial conversations (16.5M tokens)
Training Method: QLoRA (4-bit quantization with LoRA)
Language: English
License: MIT

Performance Metrics

BLEU Score Improvement: 135.36%
ROUGE-2 Score Improvement: 79.48%
ROUGE-L Score Improvement: 24.00%
ROUGE-1 Score Improvement: 12.57%
Financial Terminology Usage Increase: 90.3%

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("zahemen9900/finsight-ai")
tokenizer = AutoTokenizer.from_pretrained("zahemen9900/finsight-ai")

# Example usage
prompt = "What's a good strategy for long-term investing?"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(
    inputs.input_ids,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.95,
    do_sample=True
)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Example Interactions

Query: "What is dollar-cost averaging?"

Response: "Dollar-cost averaging (DCA) is an investment strategy in which an investor divides the total amount to be invested into equal parts and invests those parts at regular intervals, regardless of the asset's price. This approach aims to reduce the impact of market volatility by spreading out the investment over time. By purchasing more shares when prices are low and fewer shares when prices are high, DCA can potentially lower the average cost per share over the long term."

Limitations

Financial data and knowledge is current as of training data cutoff
Not connected to the internet for real-time information
Cannot provide personalized financial advice tailored to specific individual circumstances