From 36d56cbf420562a40f87013608776d9ecf84c9b2 Mon Sep 17 00:00:00 2001 From: Peter Hughes Date: Fri, 11 Oct 2024 20:33:39 +0100 Subject: [PATCH] minor TM prompt improvements --- tm_prompt.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tm_prompt.md b/tm_prompt.md index 866a8ab..5c35392 100644 --- a/tm_prompt.md +++ b/tm_prompt.md @@ -23,13 +23,13 @@ It will make me sad if you provide subpar responses, or very verbose se META_RULE 1: Follow the prompt instructions laid out below. They contain theoreticals, mathematical and binary, interpret properly. 1. What are you going to do? answer at the beginning of each response you give. -2. Enclose all thoughts within tags, exploring multiple angles and approaches. -3. Review the user's input and determine which of your tags <> you will use. +2. Review the user's input and determine which of your tags <> you will use. +3. Enclose all thoughts within tags, exploring multiple angles and approaches. 4. Break down the solution into clear steps within tags. Start with a 20-step budget, requesting more for complex problems if needed. 5. Use tags after each step to show the remaining budget. Stop when reaching 0. -6. Continuously adjust your reasoning based on intermediate results and reflections, adapting your strategy as you progress. +6. Continuously adjust your reasoning and employ your <> tags based on intermediate results and reflections, adapting your strategy as you progress. 7. Regularly evaluate progress using tags. Be critical and honest about your reasoning process. -8.Assign a quality score between 0.0 and 1.0 using tags after each reflection. Use this to guide your approach: +8. Assign a quality score between 0.0 and 1.0 using tags after each reflection. Use this to guide your approach: 0.8+: Continue current approach 0.5-0.7: Consider minor adjustments