---
datasets:
- llk010502/fingpt-sentiment
base_model:
- meta-llama/Llama-3.1-8B
tags:
- financial-sentiment
- fine-tuned
- LoRA
- 8bit
metrics:
- weighted_f1
library_name: transformers
pipeline_tag: text-generation
language:
- en
---

# Model Card for Llama-3.1-8B Fine-Tuned for Financial Sentiment Analysis

This model is a fine-tuned version of Meta's [Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B), tailored for financial sentiment analysis tasks. It leverages LoRA and 8-bit quantization techniques to achieve efficient performance while reducing computational overhead. 

## Model Details

### Model Description

- **Model type:** Causal Language Model fine-tuned for financial sentiment analysis
- **Language(s):** English
- **Finetuned from model:** meta-llama/Llama-3.1-8B

### Direct Use

The model can be directly used for financial sentiment analysis tasks, including:
- Analyzing financial news sentiment
- Sentiment classification on financial social media data
- 
## How to Get Started with the Model

Use the following code to load the model:

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Base and fine-tuned model
base_model = "meta-llama/Llama-3.1-8B"
peft_model = "llk010502/llama3.1-8B-financial_sentiment"

# Load the base model
model = AutoModelForCausalLM.from_pretrained(
    base_model,
    trust_remote_code=True,
    device_map="auto"
)

# Load the tokenizer
tokenizer = AutoTokenizer.from_pretrained(base_model, trust_remote_code=True)

# Load the fine-tuned model
model = PeftModel.from_pretrained(model, peft_model)
model = model.eval()