Hacker News Reader

> Double check your response before sending it, and if it contains any of the above words, rewrite it.

Does this instruction have any effect in the output?

I'm curious if GPT has some kind of "output buffer" where it can see the generated answer and recompute if wrong before actually outputing it.

As an LLM dummy, I assumed it is just choosing next tokens based on the tokens in the context window.

View on HN

It has an effect on the output, but not because of any output buffer as far as I know. It's just my preferred way of encouraging the AI to be more precise.

Another prompt that would work the same way would be: "Ensure your response does not include any of the above words."

ChatGPT (at least 3.5, I cannot say anything about GPT-4) will "understand" better when you reiterate your most important constraints.

> ChatGPT (at least 3.5, I cannot say anything about GPT-4) will "understand" better when you reiterate your most important constraints.

That makes sense because the recent response is part of the context, which is not the case when you say "double-check" before the answer is generated.

Reply on news.ycombinator.com