Finance Lora Llama 3.2 1b

Finance Lora Llama 3.2 1b is a LoRA adapter designed for use with the mlx-community/Llama-3.2-1B-Instruct-4bit language model. The adapter has been fine-tuned on the gbharti/finance-alpaca dataset, which is focused on financial instruction data. The model card indicates that it is intended to be used for tasks involving financial language modeling, leveraging the MLX library for integration. Instructions are provided for downloading and using the adapter with MLX, including example code for loading the base model with the adapter applied. The evidence notes an improvement in test perplexity from 10.0 (base) to 7.359 (tuned) across three seeds, suggesting enhanced performance on the targeted dataset. The adapter is produced by slm-training. No information is provided regarding pricing, licensing, or intended user roles beyond its application to financial instruction data. The tool is distributed via the Hugging Face platform and is compatible with environments such as Google Colab, Kaggle, and local applications. Further details about its broader capabilities, audience, or licensing terms are not available in the evidence.