Model Card for Fine-Tuned Mistral 7B

Model Details

Model Name: Fine-Tuned Mistral 7B (Security Exploit Code Generation)
Developed by: Yamini
Funded by [optional]: Self-funded
Shared by [optional]: Yamini
Model Type: Causal Language Model (Decoder-only)
Language(s): English (Code-focused)
License: Apache 2.0
Finetuned from model: Mistral 7B

Uses

Direct Use

This model is fine-tuned for generating skin care recommendations.

How to Get Started with the Model

Use the code below to load and interact with the model:

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Yaminii/finetuned-mistral")
tokenizer = AutoTokenizer.from_pretrained("Yaminii/finetuned-mistral")

input_text = "<insert vulnerable code snippet>"
inputs = tokenizer(input_text, return_tensors="pt")
output = model.generate(**inputs, max_length=256)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Training Details

Training Data

The model was fine-tuned on Skincare dataset from Kaggle.

Training Procedure

Preprocessing:

Tokenization with Mistral’s tokenizer
Dataset cleaning and augmentation

Training Hyperparameters:

Base Model: Mistral 7B
Fine-tuning method: LoRA (Low-Rank Adaptation)
Optimizer: AdamW
Learning rate: 2e-5
Batch size: 8
Gradient accumulation steps: 4

Speeds, Sizes, Times [optional]

Training Time: ~12 hours on 4 A100 GPUs
Final Checkpoint: ./fine-tuned-mistral-merged

Environmental Impact

Using the Machine Learning Impact calculator (Lacoste et al., 2019), estimated carbon emissions are:

Hardware Type: 4x A100 GPUs
Hours Used: ~12 hours
Cloud Provider: AWS
Compute Region: US-East
Carbon Emitted: [More Information Needed]

Technical Specifications

Model Architecture and Objective

Transformer-based autoregressive model (decoder-only)
Supports token-level and sentence-level embeddings
Fine-tuned for security attack code generation

Compute Infrastructure

Hardware: 4x A100 GPUs
Software: PyTorch, Hugging Face Transformers, LoRA fine-tuning

Citation

BibTeX:

@article{yamini2025finetunedmistral,
  author    = {Yamini},
  title     = {Fine-Tuned Mistral 7B for Skincare Recommendations},
  year      = {2025},
  journal   = {GitHub Repository},
  url       = {https://huggingface.co/Yaminii/finetuned-mistral}
}

Model Card Authors

Yamini

Model Card Contact

For questions or concerns, contact Yamini via Hugging Face or GitHub.

Yaminii
/

finetuned-mistral