# LLaMA2 Fine-tuning

T-LLaMA is a Tibetan large language model trained on the LLaMA2-7B model, constructed with a corpus containing 2.2 billion Tibetan characters, demonstrating excellent performance in text classification, generation, and summarization tasks.

A dialogue system developed based on the Llama-2-7B-chat model, supporting multilingual interaction and optimized for LangChain application scenarios

The first open-source large language model with instruction-following capability for explainable mental health analysis