---
title: "Inside a LLM Concept: Interactive Explainers on How LLMs Work"
description: "A series of interactive explainers on large language model concepts, from transformers to attention. Each runs real computations live in your browser."
keywords: "LLM concepts explained, how large language models work, interactive LLM explainer, transformer tutorial"
url: "https://withashish.com/inside-llm"
language: "en"
---

Series · learning in public

# Inside a LLM Concept

Interactive explainers on how large language models work, one concept at a time. Each one runs **real computations live in your browser** , so you can change the input and watch every step happen.

## Concepts

3 published

[ Concept 01  
Oct 2026 Inside a Transformer Follow one sentence from raw text to a next-token prediction. A tiny GPT-style transformer computes tokens, embeddings, attention, MLP and softmax live in your browser. ](/inside-llm/inside-a-transformer) [ Concept 02  
Oct 2026 Inside an LSTM Watch an LSTM read a sentence word by word: forget, input and output gates, the cell state and vanishing gradients, computed live with hand-set weights and PyTorch code. ](/inside-llm/inside-an-lstm) [ Concept 03  
Oct 2026 Inside CLIP Train a tiny CLIP in your browser: image and text encoders, cosine similarity, the symmetric contrastive loss, text-to-image search and zero-shot classification, with PyTorch code. ](/inside-llm/inside-clip)