đ Forensic Mistral-7B-Instruct-v0.3 Fine-tuned Model
This model is a fine-tuned version of Mistral-7B-Instruct-v0.3, optimized for answering questions in forensic investigations, aiming to support advanced reasoning, rapid knowledge retrieval, and high-precision assistance in the forensic domain.
đ Quick Start
This README provides detailed information about a fine - tuned model based on Mistral-7B-Instruct-v0.3 for forensic investigations.
⨠Features
- Optimized for answering forensic investigation questions.
- Trained on a specialized dataset for high - precision domain assistance.
- Achieved perfect scores on evaluation metrics within the evaluation dataset.
đĻ Installation
No installation steps are provided in the original document, so this section is skipped.
đģ Usage Examples
No code examples are provided in the original document, so this section is skipped.
đ Documentation
Summary
This model is a fine - tuned version of the Mistral-7B-Instruct-v0.3 optimised for answering questions in the domain of forensic investigations. The model has been trained using a specialised dataset titled Advanced_Forensic_Investigations_Knowledge_Library_v1, which consists of approximately 100 domain - specific question - answer pairs. The objective is to support advanced forensic investigative reasoning, rapid knowledge retrieval, and high - precision forensic domain assistance.
Model Details
Model Description
Property |
Details |
Model Type |
Instruction - following Language Model (LoRA - based fine - tuning) |
Language(s) |
English |
Fine - tuned from model |
Mistral-7B-Instruct-v0.3 |
Training Details
Training Data
Property |
Details |
Dataset |
Advanced_Forensic_Investigations_Knowledge_Library_v1 |
Data size |
~100 high - quality, domain - specific QA pairs |
Training Procedure
Preprocessing
- Template:
mistral
- Token truncation/cutoff: 2048
- No vocab resizing or prompt packing
Hyperparameters
- Finetuning type: LoRA
- Precision:
bf16
- LoRA rank: 16
- LoRA alpha: 32
- Batch size: 4
- Gradient accumulation: 8
- Learning rate:
3e - 4
- Epochs: 35
- LR scheduler: cosine
- Quantisation: 4 - bit (bitsandbytes)
- Cutoff length: 2048
Compute
- Training time: close to 30 minutes
- Framework: LLaMA - Factory
Evaluation
Testing Data, Factors & Metrics
- Metrics Used: BLEU - 4, ROUGE - 1, ROUGE - 2
- Results:
- BLEU - 4: 100%
- ROUGE - 1: 100%
- ROUGE - 2: 100%
These scores reflect perfect overlap with reference answers within the scope of the evaluation dataset.
Technical Specifications
Model Architecture and Objective
- Base: Transformer (Mistral - 7B architecture)
- Fine - tuning method: LoRA
- Objective: Instruction - following with forensic legal knowledge adaptation
Compute Infrastructure
- Hardware: 2xL40s
- Software: LLaMA - Factory, PyTorch, Transformers, bitsandbytes
đ§ Technical Details
The model is based on the Mistral - 7B architecture, using LoRA for fine - tuning. It is trained on a specialized forensic investigation dataset to adapt to the forensic legal knowledge domain. The preprocessing, hyperparameters, and training procedure are carefully designed to optimize the model's performance in answering forensic - related questions. The evaluation results show that the model has excellent performance within the scope of the evaluation dataset.
đ License
No license information is provided in the original document, so this section is skipped.