TheLastOfUs-QA: Specialized Qwen2.5-7B for The Last of Us Lore

adriangg04/TheLastOfUs-QA is a fine-tuned iteration of the Qwen2.5-7B-Instruct base model, meticulously adapted to serve as an expert on the rich universe of The Last of Us. This model is engineered to understand and generate content pertaining to the game's intricate story, diverse characters, pivotal events, and deep lore.

Key Capabilities

• Contextual Understanding: Provides coherent and detailed answers to queries about The Last of Us.
• Narrative Generation: Capable of generating explanatory or creative content based on the game's lore.
• Efficient Training: Utilizes QLoRA (4-bit) and PEFT methods, enabling training on a single NVIDIA T4 GPU with reduced VRAM usage while maintaining strong performance.
• Merged Adapters: The LoRA adapters are fully merged into the base model, simplifying usage, improving compatibility, and enhancing portability for direct deployment.

Good for

• Developing conversational assistants or chatbots focused on The Last of Us.
• Supporting creative projects that require accurate and detailed lore-based content.
• Generating narrative segments or explanations for fan wikis, guides, or interactive experiences related to the game.