Follow on chat

You will often want to continue chatting with LLM model, by asking follow up questions. Any previous questions and responses will be sent to the model as well as your new prompt, giving the LLM all the history for context. This means you don’t have to repeat your previous question.

Caution

Because the history of the chat is sent with every request when using following up questions, it will increase the number of tokens used, and therefore the cost of each chat.

Tip

To reduce cost, you can ask the AI to summarise the conversation so far, then copy the response and add it as part of your question in a new conversation.

summarise the conversation so far so I can use it in a new chat

Of course you will loose some of the detailed context, but it’s a compramise you sometimes have to make.

Enter the following question

How tall was the first species?

We are not mentioning humans at all here but the response will be related to human evolution.