|
|
--- |
|
|
|
|
|
|
|
|
base_model: mistralai/Mixtral-8x7B-Instruct-v0.1 |
|
|
tags: |
|
|
- text-generation-inference |
|
|
- transformers |
|
|
- unsloth |
|
|
- mixtral |
|
|
- mixture-of-experts |
|
|
- qlora |
|
|
- code-generation |
|
|
- python-coder |
|
|
- code-alpaca |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
--- |
|
|
|
|
|
# Puxis97/Mixtral-8x7B-Python-Coder-CodeAlpaca 🐍 |
|
|
|
|
|
This model is a **Mixtral 8x7B Instruct** model fine-tuned using **QLoRA** on the **CodeAlpaca 20K** dataset to specialize in **Python code instruction following and generation**. |
|
|
|
|
|
- **Developed by:** Puxis97 |
|
|
- **License:** apache-2.0 |
|
|
- **Finetuned from model :** mistralai/Mixtral-8x7B-Instruct-v0.1 |
|
|
|
|
|
### Training Details |
|
|
|
|
|
This fine-tuned model was built for high-efficiency using **Unsloth's QLoRA optimizations** and the Hugging Face TRL library, resulting in a powerful, instruction-following code generation model that runs on consumer GPUs. |
|
|
|
|
|
| Setting | Value | |
|
|
| :--- | :--- | |
|
|
| **Base Model** | `mistralai/Mixtral-8x7B-Instruct-v0.1` | |
|
|
| **Dataset** | `HuggingFaceH4/CodeAlpaca_20K` | |
|
|
| **Method** | QLoRA (4-bit quantization) | |
|
|
| **Task** | Code Instruction Following / Python Coding | |
|
|
|
|
|
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth) |
