From 6c65865285720aafb14f2d5eb49ec63b326cbc24 Mon Sep 17 00:00:00 2001
From: Wania Kazmi <112770629+Wania-Kazmi@users.noreply.github.com>
Date: Sun, 7 Sep 2025 16:53:35 +0500
Subject: [PATCH] Clarify Top-K and Top-P definitions in readme

Enhanced explanations for Top-K and Top-P sampling methods.
---
 00_prompt_engineering/readme.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/00_prompt_engineering/readme.md b/00_prompt_engineering/readme.md
index bcb3464..91136f5 100644
--- a/00_prompt_engineering/readme.md
+++ b/00_prompt_engineering/readme.md
@@ -157,8 +157,8 @@ Before diving into prompt techniques, understand these key parameters that contr
 - Set appropriately for your task needs
 
 ### Top-K and Top-P (Nucleus Sampling)
-- **Top-K**: Limits choices to top K most likely tokens
-- **Top-P**: Limits choices based on cumulative probability
+- **Top-K**: Limits choices to top K most likely tokens (**Only consider the k most likely next words (e.g., top-5 words)**)
+- **Top-P**: Limits choices based on cumulative probability (**Keep adding words to the list until you reach p% total probability (like filling a bucket to 90%)**)
 - Work together with temperature to control randomness
 
 **Recommended starting points:**