--- language: - en license: apache-2.0 tags: - text-generation-inference - transformers - unsloth - gemma - trl - agriculture - farming - climate - biology - agritech base_model: unsloth/gemma-2b-it-bnb-4bit datasets: - CopyleftCultivars/Natural-Farming-Real-QandA-Conversations-Q1-2024-Update --- # Uploaded model - **Developed by:** Solshine - **License:** apache-2.0 - **Finetuned from model :** unsloth/gemma-2b-it-bnb-4bit Background: 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. Following this, the Gemma version was released. Updates for this model: We then revisited the data, adding four additional months of real-world in-field data from hundreds of users which was then editted by a domain expert in regenerative farming and natural farming (approximately 2,000 instruct examples.) This was combined with a small portion of synthetic datasets and semisynthetic datasets related to regenerative agriculture and natural farming, including some non-english language samples This gemma model was trained with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library. [](https://github.com/unslothai/unsloth)