| --- |
| tags: |
| - unsloth |
| - trl |
| - sft |
| --- |
| **Final Project** |
|
|
| **Project Overview** |
|
|
| This project involves fine-tuning or training a pretrained Large Language Model (LLM) to follow instructions in assigned languages. The objective is to enable the model to generate outputs in the same language as the input query. |
|
|
| **Project Objectives** |
| 1. Fine-tune an LLM to handle instructions in the assigned languages. |
| 2. Evaluate the LLM’s performance using multilingual input and output scenarios. |
|
|
| **Deliverables** |
| 1. Notebooks |
| Training Notebook: Includes the fine-tuning process, dataset preparation, and hyperparameter details. |
| Evaluation Notebook: Details the evaluation strategy, performance metrics, and multilingual outputs. |
| 2. Documentation |
| Model Selection: Describe the primary and secondary models chosen and the criteria for selection. |
| Hyperparameters: Document and explain hyperparameters used, including the ones that worked and failed. |
| Evaluation Strategy: Justify the evaluation methodology chosen. |
| 3. Results |
| Present outputs in assigned languages. |
| Include screenshots of Google Translate or equivalent tools to verify correctness. |
|
|
| **How to Use** |
| Clone the repository: |
| git clone <repository_url> |
| cd <repository_directory> |
| Install the required packages: Ensure you have the necessary libraries installed. Can be done by running: |
|
|
| pip install -r requirements.txt Run the Jupyter Notebook: Open the notebook in Jupyter: |
|
|
| jupyter notebook Training_Finnish.ipynb and Evaluation_Finnish.ipynb |
|
|
|
|
|
|
