---
title: "Nishanthan Karunakaran (@nishanthan) | Hashnode"
description: "An Software Engineer who loves to build end to end working products. Read the latest articles by Nishanthan Karunakaran on Hashnode."
url: "https://hashnode.com/@nishanthan"
---

@nishanthan

# Nishanthan Karunakaran

@nishanthan · Joined July 2024

An Software Engineer who loves to build end to end working products

### [Core Elements of Large Language Models (LLMs): Tokens, Context Windows, Temperature, and Sampling](https://nishanthan.hashnode.dev/core-elements-of-large-language-models-llms-tokens-context-windows-temperature-and-sampling?utm_source=hashnode&utm_medium=feed)

Learn the core elements of Large Language Models (LLMs), including tokens, context windows, temperature, Top-K, Top-P sampling, and repetition penalty

6 min read · [nishanthan.hashnode.dev](https://nishanthan.hashnode.dev/core-elements-of-large-language-models-llms-tokens-context-windows-temperature-and-sampling?utm_source=hashnode&utm_medium=feed)

[Oct 26](https://hashnode.com/posts/core-elements-of-large-language-models-llms-tokens-context-windows-temperature-and-sampling/6adf39c6f218f0c2c96fe74d)

### [How Query, Key, and Value (QKV) Work in Transformers](https://nishanthan.hashnode.dev/how-query-key-and-value-qkv-work-in-transformers?utm_source=hashnode&utm_medium=feed)

Every Transformer-based language model, from BERT to GPT, relies on a mechanism called attention to determine how different tokens in a sequence relate to one another. At the heart of this mechanism a

7 min read · [nishanthan.hashnode.dev](https://nishanthan.hashnode.dev/how-query-key-and-value-qkv-work-in-transformers?utm_source=hashnode&utm_medium=feed)

[Oct 22](https://hashnode.com/posts/how-query-key-and-value-qkv-work-in-transformers/6ad9c78da2619bfccea4aca7)

### [Understanding Transformer: The Architecture Behind Modern AI](https://nishanthan.hashnode.dev/understanding-transformer-the-architecture-behind-modern-ai?utm_source=hashnode&utm_medium=feed)

What Is a Transformer? A Transformer is a neural network architecture introduced by Google researchers in 2017 in the paper Attention Is All You Need. It uses attention mechanisms to help the model le

6 min read · [nishanthan.hashnode.dev](https://nishanthan.hashnode.dev/understanding-transformer-the-architecture-behind-modern-ai?utm_source=hashnode&utm_medium=feed)

[Oct 15](https://hashnode.com/posts/understanding-transformer-the-architecture-behind-modern-ai/6ad04aba652296781d0156e8)

### [Activation Functions in Neural Networks — A Practical Mental Model](https://nishanthan.hashnode.dev/activation-functions-in-neural-networks-a-practical-mental-model?utm_source=hashnode&utm_medium=feed)

Introduction Activation functions decide how a neuron responds to an input signal. They introduce non-linearity, enabling neural networks to learn complex patterns. This article focuses on the most commonly used activation functions: Sigmoid Sof...

4 min read · [nishanthan.hashnode.dev](https://nishanthan.hashnode.dev/activation-functions-in-neural-networks-a-practical-mental-model?utm_source=hashnode&utm_medium=feed)

[Dec 22, 2025](https://hashnode.com/posts/activation-functions-in-neural-networks-a-practical-mental-model/6948bb6728045e1aedad6132)

### [Backpropagation in Deep Learning: A Complete, Intuitive, and Practical Guide](https://nishanthan.hashnode.dev/backpropagation-in-deep-learning-a-complete-intuitive-and-practical-guide?utm_source=hashnode&utm_medium=feed)

Introduction: Why Backpropagation Matters Backpropagation is the learning engine behind modern deep learning. Every time a neural network improves its predictions — from recognizing faces to generating text — the improvement comes from one mechanism:...

6 min read · [nishanthan.hashnode.dev](https://nishanthan.hashnode.dev/backpropagation-in-deep-learning-a-complete-intuitive-and-practical-guide?utm_source=hashnode&utm_medium=feed)

[Dec 17, 2025](https://hashnode.com/posts/backpropagation-in-deep-learning-a-complete-intuitive-and-practical-guide/694223e88033609809fd65fe)
