avinot/LoLlama-3.2-1B-lora-3ep-v3-instruct
avinot/LoLlama-3.2-1B-lora-3ep-v3-instruct is a 1 billion parameter instruction-tuned language model, fine-tuned by avinot, specifically specialized for the game League of Legends. Built upon a LoRA fine-tuned version of meta-llama/Llama-3.2-1B, this model excels at generating responses related to League of Legends lore, gameplay, and characters. It is designed for applications requiring deep, specialized knowledge within the League of Legends domain.
Loading preview...
Overview
avinot/LoLlama-3.2-1B-lora-3ep-v3-instruct is a 1 billion parameter instruction-tuned language model developed by avinot, specifically fine-tuned for the game League of Legends. This model is built on a LoRA fine-tuned version of meta-llama/Llama-3.2-1B and has undergone further instruction tuning using the TRL framework.
Key Capabilities
- Specialized Knowledge: Deep understanding and generation capabilities for League of Legends-specific queries.
- Instruction Following: Designed to respond to user instructions effectively within its domain.
- Efficient: A 1 billion parameter model, offering a balance between performance and computational efficiency for specialized tasks.
Training Details
The model was trained using Supervised Fine-Tuning (SFT) with the TRL library. Training involved monitoring the evolution of train and evaluation loss, as well as mean token accuracy on validation and train splits.
Good For
- League of Legends Chatbots: Creating conversational agents that can answer questions about champions, items, strategies, and lore.
- Game-Specific Content Generation: Generating descriptions, tips, or summaries related to League of Legends.
- Specialized AI Applications: Any application requiring an LLM with expert-level knowledge of the League of Legends universe.