Parameters Explained: Where an AI Model’s Knowledge Lives

Business Pill 43 · Where an AI model’s knowledge lives

Parameters are the numbers inside a model that training adjusts. Everything the model knows is stored in them, and more of them is a cost, not a guarantee.

A short explainer video, under a minute. The robot and its dials are an illustration.

The Short Answer

Picture a new robot arriving from the factory covered in tiny dials. Every dial is set at random. Ask it anything and it answers with nonsense.

Then training begins. The robot reads an example and makes a guess, and each dial is nudged a tiny bit toward a better answer. That happens billions of times.

What a Parameter Is

Those dials are called parameters. They are the numbers inside a model that training adjusts.

Everything the model knows is stored in them. The knowledge is not kept somewhere else: it sits in the settings of the dials.

Before and After Training

Before training, the dials are random and the answer is nonsense. During training, each example moves the dials a tiny bit. After training, the same question gets a sensible answer.

The question did not change. Only the dials did.

More Dials, More Cost

More dials can hold more knowledge. But they cost more to run on every single answer, not just once while the model is being built.

And a smaller model, well trained, often beats a larger one. A bigger count of parameters does not settle quality on its own.

Why It Matters

The video turns this into one question: are we paying for more dials than this job needs?

Parameters are where the knowledge lives. More of them is a cost, not a guarantee. When a smaller, well-trained model can do the job, the extra dials are cost the job does not need.

The Question to Ask

1. Are we paying for more dials than this job needs?

See It in the News

TeleOCR at 1.2B Parameters Tops a Document-Parsing Benchmark. A news piece on a model with a small parameter count leading a benchmark.

More Business Pills

• The Memory Wall: Why a Faster AI Chip Is Not Faster AI
• Tokens per Watt: What an AI Data Centre Actually Produces
• Why AI Can’t Be Both Instant and Cheap: Latency vs Throughput
• The Model and the Harness: Why Same-Model Products Differ
• Context, Not Capability: Why a Smart AI Model Gives Poor Answers
• Discardable Software: When Code Is Cheap Enough to Throw Away
• Human in the Loop vs Human on the Loop: Supervising AI Agents
• The AI Audit Problem: When Making Work Is Cheaper Than Checking It
• AI Evaluation as Acceptance Test: How to Know It’s Good Enough
• Extensibility Is Control: Who Holds the Power in an AI Product
• RLHF Explained: How an AI Model Learns What People Prefer
• Pretraining Explained: How an AI Model Learns Before Anyone Teaches It
• Fine-Tuning Explained: How to Adapt a General AI Model to One Job
• Tokens Explained: The Unit AI Reads, Writes and Bills In
• Embeddings Explained: How a Machine Compares Meaning
• RAG Explained: How a Model Answers From Your Documents
• AI Hallucination Explained: Why Models State False Things
• Distillation Explained: How a Small Model Learns From a Large One
• Reasoning Models Explained: What Changes When AI Thinks First
• Tool Use Explained: How an AI Model Goes From Text to Action
• Capex and Depreciation Explained: Why Chip Lifetime Drives AI Profits
• Run-Rate Revenue Explained: What an AI Company’s Number Means
• Backlog Explained: Revenue That Is Signed but Not Yet Earned
• Switching Costs Explained: Why It Is Hard to Leave an AI Supplier
• Temperature Explained: The Dial That Sets How Predictable an AI Model Is
• Guardrails Explained: Rules Enforced Around an AI Model, Not Inside It
• Benchmarks Explained: Why a Test Score Is Not Your Own Result
• Agent Memory Explained: How an AI Assistant Remembers You Between Chats
• MCP Explained: The Shared Plug That Connects AI Models to Tools
• Quantization Explained: How a Large AI Model Is Shrunk to Fit
• Scaling Laws Explained: Why AI Labs Keep Building Bigger
• Synthetic Data Explained: When a Model Writes Its Own Training Examples
• Transformers and Attention Explained: How AI Reads Every Word at Once

Business Engineer

The Map of AI

The Business Engineer framework maps where each part of the AI stack sits and who holds the leverage. These pills are short lessons from the same library.

Business Engineer

the Map of AI — 200+ companies across 9 layers

Read →

Business Engineer

the AI value chain — a multi-layered analysis

Read →

Explore the Map of AI →