🔥 ChatGPT Tips Tuesday #5: Divide and Conquer!

💡 ChatGPT Tips for the day

Thanks to all the people who voted in the last issue. By popular demand, I am going to talk about tips when you have a very long prompt text to ChatGPT.

Is there a word limit for each prompt you can provide? According to ChatGPT, the answer is no.

So does it mean it would be okay to type as many words in the prompt as you would like? What about summarising a 4000-word essay? Well, ChatGPT also says no 👇️ 

So what is it then? At least with my test, I can see ChatGPT is okay with 2000 words, however, it can’t handle more than 3000. I am curious if you would see the same issue with your test, feel free to let me know if you see a different limit.

So here comes the secret weapon, which is breaking down the essay into parts and sending them through via multiple prompts. To do it properly, you would need to make sure this approach is communicated with ChatGPT in your first prompt.

As for the example, I have been using this essay to test ChatGPT.

With the prompt I have included in the beginning, I can get ChatGPT to accept the documents in parts:

So in conclusion, the approach I tried did reach the result I want which is providing a 4000+ word essay to ChatGPT to be summarised. Although the last part is a bit wonky, the final result is achieved so I am pretty happy. I would be playing with the wording of the prompt if any of those changes could change this unexpected outcome with the last part submission.

Try this in your next prompt

I am going to send a document in [no. of parts you are going to send] parts and will ask you a question after all the parts are provided.  

I will go through the process this way:
- I will send several prompts that contains the part, which start with [Part X starts] and end with [Part X ends]. Please reply "Got the part".
- After the last part of the document is provided, I will ask you a question. Please only provide your response then. 

Reply "Gotchaaa" if you understand the instructions. 

If you do find a better approach than the one I have suggested, do drop me a message!

🔤 Word of the day

Chain of thought prompting (CoT)

n. improves the reasoning ability of LLMs by prompting them to generate a series of intermediate steps that lead to the final answer of a multi-step problem.
(hint: this process is a bit like what I did in today’s tip)

To explore further about this, check this section on this Wikipedia

🖼️ Image of the day

I wonder what prompting is going to be like if ChatGPT continues to talk like this. Must be fun 😆 

💕 Share the love and get some cool avatars!

I am doing weekly draw to get people referring the newsletter to others with some cool avatars as a way to say thank you.

If you can refer at least 3 of your friends, you would be in this week’s draw of 3 to get 3 cool avatars using 3 of your chosen styles.

So if you enjoy this newsletter and want to get some cool avatars for yourself, here is your link to share your love of the Prompt Craftsman with your friends.