This is the masterful Ai Task Instructions i started out with: here are two small paragraphs outlining instructions to fine-tune a GPT model for your desired conversation context: Contextual Conversation Training: To enable the GPT model to remember conversation context from the chat history, begin by preparing a training dataset that includes both user inputs and AI responses. Structure each conversation with the user's message followed by the AI's reply, building a chronological sequence. This setup helps the model understand the flow of conversation and context. For instance, if a user inquires about a product, the conversation history should provide the necessary context for the AI to give relevant responses. When fine-tuning the model, use a prompt that incorporates the chat history from the AI task input. An example prompt could be: "You are a customer support AI for tundrajunk.com. A user has provided the following chat history: [insert chat history]. The user is looking for assistance with [specific user query]. Provide a helpful response." Training the model on such prompts encourages it to maintain conversation context and generate more coherent and contextually relevant replies.