Instruction Tuning GPT-2: Fine-Tuning LLMs for Question Answering
This article explores the process of instruction tuning GPT-2, comparing vanilla performance with finetuned models, and detailing approaches to finetuning using supervised conversational data to improve question-answering capabilities.