qwen3_06B_lora_mathLORAQwen0.6B trained on the MetaMathQA dataset using Unsloth. Used to test ExecuTorch LoRA capabilities.