Phase 2 · Math, when you need it · Probability & statistics
Distributions
Random variables and the shapes their outcomes take.
Open this lesson Create a free account
Why it matters
A language model's output at each step is a distribution: a probability for each of roughly 100,000 possible next tokens. Generating text is drawing samples from it. Conversion counts, orders per hour and order values all follow distributions too, and knowing the common shapes tells you what's normal noise and what's a real change. This lesson gives you the vocabulary; the next ones measure distributions' centers and spreads.
What you'll be able to do
- Describe a random variable and its distribution: which values, how likely
- Tell discrete distributions (counts) from continuous ones (measurements)
- Recognize Bernoulli, binomial, Poisson and categorical distributions in marketing data
- Sample from a categorical distribution, the way a language model picks its next token
Before this lesson
Next in this phase: Mean, variance and spread