Model Card for Model ID
This LoRA model is a product of the research paper "LLM QLoRA Fine-Tuning of Llama, DeepSeek, and Qwen: A Skyrim Case Study" (to appear in IEEE Access). The study investigates the interplay between model scale (~8B, ~13B, ~33B), architecture, and data formatting in adapting Large Language Models to knowledge-intensive domains.
By utilizing a multi-stage data cycling strategy and 4-bit Quantized Low-Rank Adaptation (QLoRA), this model was fine-tuned to master the lore of The Elder Scrolls V: Skyrim. The training process involved rigorous hyperparameter optimization, comparing unstructured, structured, and summarized datasets across varying LoRA ranks (16, 32, 64) to determine the optimal configuration for factual recall and narrative fluency. The resulting model demonstrates the capability to generate high-fidelity, hallucination-resistant character biographies and attribute lists, as validated by a comprehensive ensemble LLM-as-a-Judge evaluation framework.
Model Details
Model Description
- Skyrim Lore - Llama-3.1-8B
- Description: This model represents the "Pareto-optimal" configuration from the study. Fine-tuned on the Summarized dataset with a LoRA Rank of 64, it achieved the perfect balance of stylistic fluency (Perplexity ~4.8) and factual recall. Uniquely, this architecture demonstrated the highest resistance to catastrophic forgetting, retaining ~62% of its general knowledge while mastering the domain.
- Best Configuration: Summarized Dataset | Rank 64
- Base Model: unsloth/Meta-Llama-3.1-8B-Instruct
- Note: You can also use the pre-quantized version unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit for faster downloading and memory efficiency.
Framework versions
