README.md

---
language:
- en
license: mit
tags:
- text-generation-inference
- transformers
- unsloth
- gemma
- trl
- agriculture
- farming
- climate
- biology
- chemistry
base_model: unsloth/gemma-2b-it-bnb-4bit
---
# Gemma 2B Natural Farmer V1 by Copyleft Cultivars

![image/png](https://cdn-uploads.huggingface.co/production/uploads/654527ce2a13610acc25d921/MBzPSJ5cWgfCmqGo4pLy-.png)


- **Developed by:** Caleb DeLeeuw, Copyleft Cultivars (a nonprofit, protecting and preserving vulnerable plants)
- **License:** MIT
- **Finetuned from model :** unsloth/gemma-2b-it-bnb-4bit

Using real-world user data from a previous farmer assistant chatbot service and additional curated datasets (prioritizing sustainable regenerative organic farming practices,) Gemma 2B and Mistral 7B LLMs were iteratively fine-tuned and tested against eachother as well as basic benchmarking, whereby the Gemma 2B fine-tune emerged victorious. LORA adapters were saved for each model. The Natural Farming Gemma 2B was then loaded onto Hugging Face Hub in hopes it will help farmers everywhere and inspire future works.

Shout out to roger j (bhugxer) for help with the dataset and training framework.

Testing and further compiling to integrate into on-device app interfaces are ongoing. This project was created by Copyleft Cultivars, a nonprofit, in partnership with Open Nutrient Project and Evergreen State College. This project serves to democratize access to farming knowledge and support the protection of vulnerable plants.

This is V1 beta. It runs locally on Ollama with some expirimental configuring so you can use it off the grid and places where internet is not accessible (ie most farms I've been on.)

This Gemma model was trained with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)