Memory Books

Memory books serve as a supplemental memory system for AI to retain and reference extended context beyond the immediate interaction limits. Directly putting all information into the active context could overwhelm the AI’s immediate processing capacity. Memory books allow for a larger and more detailed backstory or context to be maintained without affecting the AI’s performance within the current session. They help maintain the richness of the interaction over time and across sessions.

Note: The AI model would not have intrinsic knowledge of the information stored in the memory book until it is referenced in the conversation. For instance, if you have a character with a pet named Jack, the model won’t actively use this information in the conversation unless it’s brought up or relevant to the context. The memory book acts as a reference point that the AI can draw from when needed, such as when you ask about the pet. This allows the AI to bring in details from the memory book to make the interaction more coherent and contextually rich.

Tip: Add multiple keywords to each memory book entry. This improves the AI’s information retrieval. For example, include dog, pet, Jack, animal instead of just dog, enabling the AI to quickly access relevant details when these concepts are mentioned.