This is an experimental fine-tune of Google's Gemma-2B using the Dolphin-R1 dataset.
The goal is to enhance reasoning and chain-of-thought capabilities while maintaining efficiency with LoRA (r=32) and 4-bit quantization.
🚨 Disclaimer: This model is very much a work in progress and is still being tested for performance, reliability, and generalization. Expect quirks, inconsistencies, and potential overfitting in responses.
image/png
This is made possible thanks to @unsloth. I am still very new at finetuning Large Language Models so this is more of a showcase of my learning journey. Remember, it's very experimental, do not recommend downloading or testing.
